Advances in Economics, Management and Political Sciences PDF Free Download
1 / 309/309
100%
AEMPS
Advances in Economics, Management and
Political Sciences
Proceedings of the 2nd International Conference
on Financial Technology and Business Analysis
Oxford, UK
November 8 - November 15, 2023
Volume 56
Editor
Javier Cifuentes-Faura
University of Murcia
ISSN: 2754-1169
ISSN: 2754-1177 (eBook)
ISBN: 978-1-83558-159-9
ISBN: 978-1-83558-160-5 (eBook)
Publication of record for individual papers is online:
https://aemps.ewapublishing.org/
Copyright © 2023 The Authors
This work is fully Open Access. Articles are freely available to both subscribers and the wider public with permitted reuse.
No special permission is required to reuse all or part of article, including figures and tables. For articles published under
an open access Creative Common CC BY license, any part of the article may be reused without permission, just provided
that the original article is clearly cited. Reuse of an article does not imply endorsement by the authors or publisher.
The publisher, the editors and the authors are safe to assume that the advice and information in this book are believed to
be true and accurate at the date of publication. Neither the publisher nor the editors or the authors give a warranty,
expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been
made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This imprint is published by EWA Publishing
Address: John Eccles House, Robert Robinson Avenue, Oxford, England, OX4 4GP
Email: info@ewapublishing.org
Committee Members
ICFTBA 2023
General Chair
Arman Eshraghi, Cardiff Business School
Organizing Chair
Canh Thien Dang, King's College London
Organizing Committee
Alina Cristina Nuta, Danubius University of Galaţi
Kevin McMeeking, Brunel University London
Daniel Balsalobre Lorente, University of Castilla-La Mancha
Kristina Lalova, University of Connecticut
Dhiroj P Koirala, University of Massachusetts Amherst
Javier Cifuentes-Faura, University of Murcia
Waseem Ahmad, University of Agriculture, Faisalabad
Muhammad Jamil, The University of Azad Jammu & Kashmir
Muhammad Hashim, National Textile University Faisalabad
Jeannievilyn Ola, De La Salle University
Zhongda He, Central University of Finance and Economics
Ziruo Li, Peking University
Yu Lu, University of Wisconsin-Madison
Ang Li, Lingnan University
Ivoslav Ganchev, Queen Mary University of London
Fiza Qureshi, University of Southampton Malaysia
Technical Program Chair
Naser Makarem, University of Aberdeen
Shima Amini, University of Leeds
Chinny Nzekwe-Excel, Birmingham City University
Runyu Chen, University of International Business and Economics
Mingfeng Dai, Chinese Academy of International Trade and Economic Cooperation
Xuezheng Qin, Peking University
Yuli Bian, Rutgers University
Weiwei Jiang, Tsinghua University
Mi Shen, Central University of Finance and Economics
Publicity Committee
Muhammad Hafeez, University of Agriculture, Faisalabad
Yu Liu, Beijing University of Posts and Telecommunications
Wei Meng,
Peking University
Preface
The 2nd International Conference on Financial Technology and Business Analysis (ICFTBA 2023)
is an annual conference focusing on research areas including finance, economics, business, and
management. It aims to establish a broad and interdisciplinary platform for experts, researchers, and
students worldwide to present, exchange, and discuss the latest advance and development in finance,
economics, business, and management.
This volume contains the papers of the 2nd International Conference on Financial Technology and
Business Analysis (ICFTBA 2023). Each of these papers has gained a comprehensive review by the
editorial team and professional reviewers. Each paper has been examined and evaluated for its theme,
structure, method, content, language, and format.
Cooperating with prestigious universities, ICFTBA 2023 organized four workshops in Cardiff,
London, Aberdeen and Galaţi. Prof. Arman Eshraghi chaired the workshop “Recent Bank Collapses
- A new Financial Crisis?”, which was held at Cardiff Business School. Professor Kevin McMeeking
chaired the workshop “Sustainable Finance Workshop: Accounting for Greenhouse Gases” at Brunel
University London. Dr. Naser Makarem chaired the workshop “Brexit and Earnings Management” at
University of Aberdeen. Professor Dr. Habil. Alina Cristina Nuţă chaired the workshop “Fintech
Tools and Cybersecurity Challenges - Finance Innovations That Shapes Our Lives” at Danubius
University of Galaţi.
Besides these workshops, ICFTBA 2023 also held an online session. Eminent professors from top
universities worldwide were invited to deliver keynote speeches in this online session, including Dr.
Shima Amini from The University of Leeds, Dr. Kevin McMeeking from the Brunel University
London, Dr. Naser Makarem from University of Aberdeen, etc. They have given keynote speeches
on related topics of finance, economics, business, and management.
On behalf of the committee, we would like to give sincere gratitude to all authors and speakers who
have made their contributions to ICFTBA 2023, editors and reviewers who have guaranteed the
quality of papers with their expertise, and the committee members who have devoted themselves to
the success of ICFTBA 2023.
Prof. Arman Eshraghi
General Chair of Conference Committee
Workshop
Workshop – Cardiff: Recent Bank Collapses - A New Financial Crisis?
September 1st, 2023 (GMT+1)
Finance and Chair of Finance and Investment, Cardiff Business School
Workshop Chair: Prof. Arman Eshraghi, Professor in Cardiff Business School
Workshop – London: Sustainable Finance Workshop: Accounting for Greenhouse Gases
13 October 2023 (GMT+1)
College of Business, Arts and Social Sciences, Brunel University London
Workshop Chair: Professor Kevin McMeeking, Professor at Brunel University London
Workshop – Aberdeen: Brexit and Earnings Management
October 18th, 2023 (GMT+1)
Department of Accountancy and Finance, University of Aberdeen
Workshop Chair: Dr. Naser Makarem, Assistant professor in University of Aberdeen
Workshop – Galaţi: Fintech Tools and Cybersecurity Challenges - Finance Innovations That
Shapes Our Lives
October 15th, 2023 (GMT+3)
Faculty of Economics and Business Administration, Danubius University of Galati
Workshop Chair: Professor Dr. Habil. Alina Cristina Nuţă, Danubius University of Galaţi
The 2nd International Conference on Financial
Technology and Business Analysis
ICFTBA 2023
Table of Contents
Committee Members······························································································································
Preface ·······················································································································································
Workshop ·················································································································································
The Development and Problems of the Live Broadcasting Industry under the Impact of
COVID-19 ··············································································································································· 1
Julie Ann Palumbo
Research on the Aging of Global Workforce and Solutions ···························································· 7
Xinyue Pu
The Influence That Jeonse Rental System Brought to the Housing Bubble in Korea ················· 15
Lancheng Fei
The Value and Promotion Role of Opinion Leaders in Media Communication ························ 21
Yixuan Pei
Research on the 2008 Financial Crisis Regarding Financial Network ·········································· 26
Shiyun Ding
The Impact of the Interest Rates Raised by Federal Reserve System on the Exchange Rate of
the US Dollar Against the Chinese Yuan ························································································· 34
Yuenan Chen
Analysis of the Real Estate Market in China’s Past and Present ··················································· 42
Hao Gong
Research on the Investment Portfolio Optimization Based on Efficient Frontier Model: A
Portfolio of AMD, NVIDIA, TXN, LRCX, AVGO, QCOM, INTC and MRVL ···························· 49
Ye Chen
The Impact of Systemic Financial Risks on the Shanghai Composite Index: Evidence from
ARIMA Model ····································································································································· 56
Yuqi Li
Research on the Relationship Between International Trade and Network ································· 64
Run Guo, Tongxi Wang, Chengyi Xu, Yichen Zhou
Stock Price Forecasts Based on KNN and LSTM ············································································ 70
Zihao Chen
The Impact of Corporate Social Responsibility on Corporate Financial Performance --An
Empirical Study of Chinese Energy Firms ······················································································· 78
Zhuoqun Xu
Three Applications of the Anchoring Effect ···················································································· 88
Yu Liu
A Case Study of Bright Food’s Acquisition of GNC in 2011 ························································· 92
GuangMei Zhou
An Analysis of the Applicability of Pricing Models in the Chinese Stock Market ····················· 98
Tianshu Wang
The Study of Risk Management of Financial Derivatives -Taking the Archegos Liquidation
Event as Example······························································································································· 104
Fengyi Sun
The Influence of Sino-US Trade War on Chinese Students Studying in the United States ····· 110
Xuemei Wang
Research on the Influencing Factors of Chinese Gasoline Price Fluctuation During the
COVID-19 Epidemic ·························································································································· 119
Chuhan Zhang
A Case Study of Bright Dairy’s Acquisition of Israel’s Tnuva Foods ········································ 124
Rui Zhong
Research on Zara’s Social Media Marketing Strategy in the Context of New Media ·············· 131
Jiayi Wang
An Analysis of Sexy Tea’s Marketing Strategy and Its Impact on Brand Communication of
New Chinese Milk Tea in China ······································································································ 139
Shihan Guo
Portfolio Optimazation Based on 10 US Stocks ············································································· 147
Tianzhou Yu
The Impact of Population Aging on Financial Services and Economic Development ············ 155
Hongyu Fang
Stock Price Prediction Based on ARIMA and Neural Network ·················································· 163
Xinyu Liao
Application of Mean-variance Model in Optimizing Stock Portfolio ········································ 172
Xingyu Qian
Validity Testing of Classical Asset Allocation Models: An Empirical Study ···························· 179
Chong Gao, Shengkai Xu
Research on the Brand Marketing Strategy of Netflix in the New Media Environment ········· 195
Huazhen Xiao
Imbalance Between Supply and Demand in China’s Labor Market Facing by the Graduates
······························································································································································ 203
Taoyun Lian
The Implementation of Modern Portfolio Theory on New Financial Assets: Evidence from
Cryptocurrencies ······························································································································· 209
Shuai Chen
Research on the Problems and Countermeasures Analysis of Old-age Security for the Elderly
in Rural China in the Context of Population Aging ····································································· 214
Le Guan
The Key Functions of Relevance and Relationship Effect in Marketing: Based on the Analysis
before and after Restructuring of Huiyuan Group ······································································· 222
Zongyi Li
Comparative Study on the Low Birth Rate Process in China and Japan and the Impact on
University Graduates’ Employment ······························································································· 228
Mingwei Wang
Research on Strategic and Financial Performance of Pop Mart ·················································· 238
Yaqi Guo
Strategies and Practices for Dealing with Inflation in New York ··············································· 247
Ying Wang
The Impact of Macroeconomic Factors on Non-Performing Loans of Commercial Banks: An
Empirical Study Based on 2022 Provincial-level Data in China ·················································· 256
Yixiang Li, Diyi Gao, Shu Fang, Xuesong Chen, Dan Wan
The Sino-US Trade Conflict’s Impact on Semiconductor Stock Index: China’s Case ·············· 263
Zhenxiang Liu
Application of Various Mechanisms on Students School Choice Problem ······························· 271
Zhenhao Cai
Research on Marketing Strategy of Lawson Convenience Store Brand from the Perspective of
New Media ········································································································································· 277
Jinsheng Tan
Stock Price Prediction for Technology Company ········································································· 284
Yuchen Wang
Applicatoin of TTCC Algorithm in House Reallocation Market in China ································ 291
Hanyu Xu
The Development and Problems of the Live Broadcasting
Industry under the Impact of COVID-19
Julie Ann Palumbo1, a, *
1Shanghai High School International Division, Shanghai, China
a. 1811000407@mail.sit.edu.cn
*corresponding author
Abstract: Faced with the convenience of the internet, people begin to rely more on the broad
internet and the leisure it brings, allows people to transcend the limitations of time and space,
experience and share exciting cultural, sports, artistic and other activities from around the
world together. Network broadcasting has blossomed in recent years, from the earlier game
broadcast, singing, and dancing to recent years where every event – no matter how big or
small – has the chance to be broadcasted. Following the sudden outbreak of COVID-19, live
streaming has been integrated into almost every person’s lives. With the increasing
popularization of an online lifestyle, live broadcasting has gradually integrated into
everyone’s life. Through research and analysis, this paper aims to probe into the impacts and
problems of the live-streaming industry, exploring its development since the pandemic and
analyzing its main competitors and its future.
Keywords: live-streaming, e-commerce, network, broadcasting
1. Introduction
1.1. Background
The outbreak of COVID-19 since 2020 inevitably had a negative impact on the economy. The
pandemic has claimed a large number of lives worldwide and presented unprecedented challenges to
public health and food systems. As factory closures and quarantines spread across the globe, limiting
the movement of people and commerce, companies are grappling with lost revenue and supply chain
disruptions. As global habits change to adapt to the new realities of the outbreak, consumers adapt to
the new lifestyle. This novel three-dimensional shopping method has attracted ordinary people under
the epidemic wave and brought great changes to the live streaming industry. In addition, live
broadcasting has become people’s main means of living, rather than an entertainment choice.
1.2. Related Research
Yu and Zhang conducted in-depth research and analysis on livestreaming from farmers containing
the quality of altruism in order to sell unmarketable products and alleviate farmers’ economic
difficulties caused by the pandemic. By collecting effective feedback from 475 Chinese consumers
who participated in farmers’ public welfare livestream, they explored the influencing factors on
consumers’ attitudes and purchase intentions towards agricultural products from three levels:
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231044
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
1
platform, product, and consumer [1]. Liu et al. investigated the mechanism of live broadcasting to
make up for defects and regain confidence [2]. Zhao and Bacao investigated the mental processes of
different populations during COVID-19 lockdowns. The structural equation model based on
covariance was used to analyze 374 verification data. The statistical results show a significant
difference between gender groups, which represents the level of engagement and immersion of users
shopping through LSSAs [3].
Mihelj et al. examined and quantified an enormous set of interviews and media diaries collected
in eastern European countries during the first wave of the pandemic [4]. Lee and Kwon conducted a
descriptive survey on changes in consumer demand in the cosmetics market, which has changed since
the COVID-19 outbreak [5]. Taking China as an example, Qiu et al. studied travel practices related
to live streaming. Their findings showed that live content mainly caused positive emotions, while
negative emotions were caused by illegal or boring content [6].
Luo analyzed the pre-covid and post-covid state of major streaming companies and their
competitions on prices and contents. Focusing on several different aspects, Luo examines the
psychology of customers regarding online streaming during the pandemic, providing data, case
studies, and recommendations for areas of improvement [7]. Ryu and Cho outlined and analyzed how
the COVID-19 pandemic is affecting different areas of the global entertainment industry, involving
the physicality and digitization of business. They analyzed four different scenarios to help readers put
the unprecedented impact on the entertainment industry into a more panoramic perspective, and come
up with two points of comparison: the direction of the COVID-19 impact and the continuity of change
[8]. Huang et al. used unique real-time datasets from livestreaming platforms to investigate how fans’
virtual gifts and livestreaming e-commerce decisions, as well as streamers’ sales, were affected by
COVID-19 [9]. Miah et al. investigated the impact of media on Bangladeshi consumers’ online
shopping behavior during the pandemic [10].
2. The Live Broadcasting Industry
2.1. The Rise of the Live Broadcasting Industry
In 2016, investors are optimistic about the development prospects of the live e-commerce industry,
and have invested several online live streaming platforms that has emerged in China. While the
number of viewers has surged, the initial live content is mostly in the entertainment field such as
games. However, in March 2016, Mogujie started taking the lead in expanding, introducing live
streaming of e-commerce content, increasing live shopping functions, and forming a platform of “live
streaming + content + e-commerce” to reduce costs and enhance consumer stickiness, so as to realize
traffic flow. Since then, Taobao, Jingdong and other platforms have launched live broadcast functions
and joined the e-commerce live broadcast army.
In 2019, the scale of live streaming e-commerce industry has witnessed explosive growth, many
stars and government TV stations have joined the live streaming army, anchors such as Li Jiaqi and
Wei Ya are well-known, and live streaming with goods is hotly debated by the whole people.
Under the impact of the epidemic in 2020, the “house economy” has developed rapidly, which has
contributed to the expansion of e-commerce live broadcasting. Mall cabinet sisters, business owners,
county mayors and other groups have joined the ranks to become anchors to carry out goods in order
to save the real economy on the verge of bankruptcy; The live broadcast environment became
diversified, from a single indoor live broadcast room to anchors seeking the suppliers. At the same
time, many relevant policies and regulations have been introduced and implemented.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231044
2
2.2. Factors That Influence Consumers’ Purchase Intentions
With the growth of live streaming users in the Internet environment and the birth and rise of network
celebrities, the live marketing scale has developed rapidly. This phenomenon allows consumers to
understand products in a new way and shorten the time to purchase. Internet celebrities gain
popularity through live streaming, while using their own popularity to promote products in live
streaming to enhance consumers’ willingness to purchase goods.
The most common promotion methods in live broadcasting are discounts, gifts, limited-time sales,
sweepstakes, cash rebates, coupons, etc. Researchers studied the perceived frequency of the use of
various promotional tools by college students and found out that the most commonly used promotion
method was discount, followed by direct gift. Various promotion methods, can promote consumers’
browsing of shopping websites, thus stimulating consumers’ impulsive purchase intentions [11].
2.3. Market Analysis
The scale of China’s live broadcast market from 2016 to 2026 have been analyzed and predicted, and
the results show that the scale of China’s live broadcast market will reach about 700 billion yuan in
2023 [12].
China’s game livestreaming market is worth 110 billion yuan, accounting for about 30 percent of
game revenue. Despite increasingly stringent gaming time restrictions for minors in the country, the
industry is still growing at a double-digit rate, largely driven by a surge in esports streaming. In
addition to viewer tips and advertising, game streaming platforms also generate revenue from
subscriptions, game operations and guessing games. In the coming years, their diversified
monetization models will reap the fruits of technological advances such as 5G, cloud computing, and
virtual reality.
With the advent of the 5G era, the live broadcast economy will be blessed by network speed
improvement, intelligent terminal upgrading, augmented reality (AR), AI and other technologies,
thereby greatly improving user experience, transaction efficiency and its own boundaries. The live
broadcast industry is developing from the incremental model to the stock era, but the live broadcast
economic concept has gradually penetrated the hearts of the people, and the industry tuyere has not
disappeared, but has come to the second half of the fine development. “+ live” is opening up a broader
development space for the industry.
2.4. Competitors in the Industry
China’s live streaming market is in a period of vigorous development. After the rapid development
in recent years, it is currently in a situation of fierce market competition, and it is forecasted that the
competition pattern will undergo profound changes. According to market research data analysis, the
main competitors in the Chinese market in 2020 include Tencent, Jingdong, iQiyi, etc. In 2023, the
market share is still dominated by Tencent, about 20%, followed by Jingdong, iQiyi (19% and 17%).
The rest of the market share is Pinduoduo (13%), Youku (7%), Mogujie (6%), Kuaishou (5%),
Taobao (4%) and so on. According to the scale analysis, on the whole, there are more than 1,000
enterprises in China’s live streaming e-commerce industry, and a large number of third-party service
providers actively participate in it, providing a full range of service support through technology and
investment; Entrepreneurial activities are also increasing, and they are taking advantage of their own
and partners’ advantages to seize the leading position in the industry in the face of the needs of users
in the new era. Therefore, new competitors such as NetEase Yanquan, Pinduoduo and Youku have
emerged in Guangdong’s live streaming industry. These emerging enterprises are gradually winning
the recognition of consumers, affecting and changing the competitive landscape of the market.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231044
3
In 2023, the competition pattern of China’s live streaming e-commerce market is still dominated
by Tencent, Jingdong and iQiyi, occupying an absolute market share advantage, while the emerging
live streaming e-commerce enterprises that focus on brand development, research and development
of innovative products and grasp the leading position of the industry are also gradually winning the
recognition of consumers, affecting and changing the market competition pattern.
3. Problems and Solutions
3.1. Problems
In recent years, the internet has facilitated people’s lives and changed their way of living, thinking,
and behaving, becoming an indispensable part of people’s lives. Nowadays, a mobile phone can be
said to solve various needs in our daily life. We can understand the news events around the world
without leaving the house, and we can buy high-quality goods around the world. At the same time,
because of it, many new industries have sprung up and developed vigorously, among which network
broadcast has become a new favorite of many people, especially young people. However, problems
can emerge with the development.
3.1.1. The Unhealthy Value Orientation for Teenagers
With the development of video broadcast platforms such as Tiktok and Kuaishou, network broadcast
has become the new favorite of people, especially youngsters. With the rapid development of network
broadcast, anchors have gradually become a professional representative with low threshold and high
income in people’s minds. Moreover, many young people’s future dream is to become a network
broadcast and become a network celebrity. Of course, recognition and encouragement should be given,
but attention should also be paid to the negative side of it. Researchers have also proved that some
online live broadcast platforms have very low personal requirements for anchors, only proposing the
basic conditions of being at least 18 years old and real-name authentication, without involving legal
concepts and moral awareness requirements, and without providing pre-job education and training.
With the rapid development of mobile Internet technology, mobile phones have become an
important window for children to understand the world [13]. In the past, when the internet highway
has not yet been built, the speed of information transmission and diffusion is relatively lagging, and
teachers, parents, newspapers, and television are the only sources of information for children. But
now, memes or rude and meaningless jokes seems to have become a route for many children to gain
attention among their peers.
Furthermore, some webcasts spread a unified but distorted aesthetic standard to the public. To meet
the demands of audiences, anchors go through plastic surgery, makeup, and other means of packaging
to portray the most popular awl face, showing an abandonment of self, chasing deformed aesthetic
standards while walking in the forefront of fashion. Living in this kind of abnormal aesthetic standard
can severely affect the aesthetic values of teenagers.
3.1.2. E-commerce Fraud and Selling Counterfeit Products
With the intensification of competition, illegal merchants sell fake and shoddy products and spread
false information through online live broadcast platforms, which has hit the market order and
consumer interests. This behavior has become a hot issue of social concern recently.
Fraud has existed since the birth of e-commerce. However, the surge in online sales and shopping
brought about by the COVID-19 pandemic has provided new opportunities for scammers. Online
shopping scams accounted for 38 percent of all scams reported globally in 2020, up from 24 percent
before the pandemic [14]. While that number has declined since the crisis subsided, security breaches
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231044
4
continue to hit the industry hard, with losses from online payment fraud totaling more than $40 billion
in 2022. In this context, the e-commerce fraud detection and prevention market is expected to more
than triple to over $100 billion between 2023 and 2027.
There are several reasons for this phenomenon. First, there is a lack of a certain gatekeeping system
for sellers. Due to the convenience of information dissemination and the lag of traditional
management, major network platforms have loopholes in reviewing the main account of live
broadcasting, and they cannot fully control the behavior of sellers, leading to some illegal businesses
to advertise and sell illegally. Secondly, sellers wish to pursue a rapid development, driven by sales
and profits. In recent years, the e-commerce industry has been booming, and a large number of online
live broadcast platforms and aggregation platforms have sprung up, such as Taobao, Tiktok and
Kuaishou, causing extremely fierce market competition. To quickly obtain sales and profits,
enterprises promote fake goods through the distribution platform and obtain a large number of profits
in a short period of time, regardless of employee welfare and consumer interests.
3.2. Solutions
At present, the relevant state departments have begun to strengthen the supervision of live streaming
e-commerce platforms. For example, industry and commerce, taxation, customs, and other
departments have standardized the registration, certification, taxation, import and export of live e-
commerce platforms. With the continuous improvement of regulatory measures, the development
prospects of live streaming e-commerce platforms will certainly be broader.
Live streaming e-commerce platforms should conduct qualification certification for anchors and
require anchors to provide truthful information. This can prevent the entry of bad merchants and
anchors from the source. In the process of qualification certification, live streaming e-commerce
platforms should strictly review the background, experience, reputation, and other aspects of the
anchor to ensure that the identity of the anchor is authentic and credible. In addition, the performance
of anchors can be evaluated by setting up a corresponding rating mechanism to encourage the
development of excellent anchors.
In order to ensure the healthy development of live e-commerce platforms, live content should be
reviewed by the platform before it can be broadcast. The focus of the review is to ensure the
authenticity, legality and morality of live content. Live content shall not appear false propaganda,
illegal information, etc., to ensure that the rights and interests of consumers are protected. At the same
time, anchors should also be encouraged to create useful and valuable content to improve the overall
quality of live streaming e-commerce platforms.
Moreover, live streaming platforms should supervise the quality of goods, ensure that the goods
sold meet relevant standards, and punish unqualified goods. For goods with quality problems, live
streaming e-commerce platforms should be removed in a timely manner, and punish merchants to
protect the legitimate rights and interests of consumers. In addition, it should also be possible to
conduct random inspections of goods through third-party quality inspection institutions to improve
the overall level of commodity quality.
4. Conclusion
With the rapid development of Internet technology, network broadcast has become a new industry in
today’s society, which has a profound impact on people’s lives. Joint efforts should be made from
government supervision, industry self-discipline, platform construction and other aspects to promote
the healthy and orderly development of China’s live streaming industry. Only regulations and
supervision are in place to ensure the healthy development of live streaming e-commerce platforms
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231044
5
and provide better services for consumers. With the continuous improvement of regulatory measures,
the development prospects of live streaming e-commerce platforms will be broader.
References
[1] Yu, Z., & Zhang, K. (2022). The Determinants of Purchase Intention on Agricultural Products via Public-Interest
Live Streaming for Farmers during COVID-19 Pandemic. Sustainability, 14(21), 13921.
https://doi.org/10.3390/su142113921.
[2] Liu, X., Yuan, Y., He, J., & Li, Z. (2022). Framing the travel livestreaming in China: a new star rising under the
COVID-19. Current Issues in Tourism, 1–20. https://doi.org/10.1080/13683500.2021.2023115
[3] Zhao, Y., & Bacao, F. (2021). How Does Gender Moderate Customer Intention of Shopping via Live-Streaming
Apps during the COVID-19 Pandemic Lockdown Period? International Journal of Environmental Research and
Public Health, 18(24), 13004. https://doi.org/10.3390/ijerph182413004
[4] Mihelj, S., Kondor, K., & Štětka, V. (2021). Audience Engagement with COVID-19 News: The Impact of Lockdown
and Live Coverage, and the Role of Polarization. Journalism Studies, 1–19.
https://doi.org/10.1080/1461670x.2021.1931410
[5] Lee, J., & Kwon, K. H. (2022). Mobile shopping beauty live commerce changes in COVID‐19 pandemic focused on
fun contents of MZ generation in Republic of Korea. Journal of Cosmetic Dermatology.
https://doi.org/10.1111/jocd.14442
[6] Qiu, Q., Zuo, Y., & Zhang, M. (2021). Can Live Streaming Save the Tourism Industry from a Pandemic? A Study of
Social Media. ISPRS International Journal of Geo-Information, 10(9), 595. https://doi.org/10.3390/ijgi10090595
[7] Luo, Y. (2020). The Streaming War During the Covid-19 Pandemic.
https://static1.squarespace.com/static/51d98be2e4b05a25fc200cbc/t/5fd79db79eb445300d4af437/160796613655
6/Streaming+Wars+During+the+Pandemic+Yujin+Luo+Final.pdf
[8] Ryu, S., & Cho, D. (2022). The show must go on? The entertainment industry during (and after) COVID-19. Media,
Culture & Society, 44(3), 016344372210795. https://doi.org/10.1177/01634437221079561
[9] Huang, Q., Xiao, S., & Zhuang, C. C. (2022). The heterogeneous impact of COVID-19 in a two-sided market:
evidence from a live-streaming platform. Applied Economics, 54(49), 1–20.
https://doi.org/10.1080/00036846.2022.2048787
[10] Miah, Md. R., Shikder, R., Hossain, A., Saha, T., & Neger, M. (2022). Evaluating the impact of social media on
online shopping behavior during COVID-19 pandemic: A Bangladeshi consumers’ perspectives. Heliyon, 8(9),
e10600. https://doi.org/10.1016/j.heliyon.2022.e10600
[11] Chen, Bing, et al. “Research on the Impact of Marketing Strategy on Consumers’ Impulsive Purchase Behavior in
Livestreaming E-Commerce.” Frontiersin.org, 16 June 2022,
www.frontiersin.org/articles/10.3389/fpsyg.2022.905531/full. Accessed 18 Aug. 2023.
[12] China: live streaming market size 2026. (n.d.). Statista. Retrieved September 24, 2023, from
https://www.statista.com/statistics/874591/china-online-live-streaming-market-
size/#:~:text=According%20to%20the%20forecast%2C%20the%20market%20size%20of
[13] Hillyer, Madeleine. (2020, November 18). Here’s how technology has changed the world since 2000. World
Economic Forum; World Economic Forum. https://www.weforum.org/agenda/2020/11/heres-how-technology-has-
changed-and-changed-us-over-the-past-20-years/
[14] cycles, T. text provides general information S. assumes no liability for the information given being complete or
correct D. to varying update, & Text, S. C. D. M. up-to-Date D. T. R. in the. (n.d.). Topic: E-commerce fraud.
Statista. https://www.statista.com/topics/9240/e-commerce-fraud/
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231044
6
Research on the Aging of Global Workforce and Solutions
Xinyue Pu1,a,*
1Clarkson Secondary School, 2524 Bromsgrove Rd, Mississauga, Ontario
a. 1911411112@mail.sit.edu.cn
*corresponding author
Abstract: The twenty-first century has seen significant shifts in fertility attitudes and
demographics, leading to an aging global workforce. This paper explores how different
regions - the United States, Western Europe, Japan, and China - address the challenges and
opportunities posed by their aging populations in terms of healthcare, social security, and
employment. The main findings reveal diverse strategies. The United States combines public
and private healthcare components through Medicare and private insurance. Western Europe
emphasizes publicly funded healthcare, exemplified by the UK’s NHS and Germany’s multi-
layered pension system. Japan’s universal coverage and pioneering Long-Term Care
Insurance stand out, while China adapts healthcare for both urban and rural elderly residents.
Social security systems range from Germany’s multi-faceted pensions to Japan’s
comprehensive structure, and China is working to bridge urban-rural disparities. Employment
trends vary, with the United States witnessing older adults extending work lives, the
Netherlands introducing innovative retirement schemes, Japan promoting “active aging,” and
China focusing on skill enhancement. The study suggests cross-country learning and
adaptation. Each nation can benefit from the best practices of others to enhance its own
strategies. Collaborative approaches can drive innovative solutions for aging populations,
fostering resilient healthcare systems, inclusive social security, and age-friendly employment
environments. This comparative analysis offers insights for creating a harmonious and
sustainable future for aging populations globally.
Keywords: aging workforce, healthcare, social security, employment
1. Introduction
1.1. Background
Over the course of the twenty-first century, fertility attitudes have begun to change in many regions
as the overall quality of life of people around the globe has improved. This has led to a year-on-year
decline in the number of newborns, the effects of which have gradually become apparent over the
decades, leading to the emergence of an aging average age of the workforce. The global workforce is
experiencing significant demographic shifts, with an increasing aging population becoming a
prominent feature. Besides, advancements in healthcare, improved living conditions, and declining
fertility rates have led to higher life expectations worldwide. This demographic transformation has
economic, social, and business implications, affecting labor markets. As older workers delay
retirement and younger generations enter the workforce, employers face challenges in managing an
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231045
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
7
age-diverse workforce and addressing age-related biases. Creating inclusive work environments that
value age diversity becomes crucial in harnessing the experience and expertise of older workers.
1.2. Related research
Diving into the multifaceted landscape of aging dynamics, an array of research perspectives coalesce
to provide a comprehensive understanding of its implications across various domains. Pit et al.’s
illuminating study examines the reverberations of the COVID-19 pandemic on older workers and
ageism. Their exploration encompasses 15 case studies from diverse countries, showcasing how the
pandemic has exacerbated inequalities among senior employees. Their proposed solutions, including
funding support, intergenerational contact interventions, and educational activities, underscore the
urgent need for inclusive workplaces that combat ageism and foster the well-being of older workers.
This resonates with the World Health Organization’s call for global initiatives to support older
individuals within the workforce [1].
Venturing further into the labyrinth of aging, Vodopivec and Arunatilake’s research unravels the
labor market consequences of Sri Lanka’s aging population. Their insights reveal a stark reality where
a substantial proportion of older workers find themselves entrenched in the informal sector, grappling
with extended working hours and meager remuneration [2]. The intricacies of retirement motivations
highlight disparities between formal and informal sector workers, shedding light on the multifaceted
nature of labor market duality in later life. Serban’s examination of demographic conditions within
European Union countries underscores the paramount role of education in navigating the challenges
precipitated by an aging populace. Education emerges as a beacon that can bolster labor market
adaptability and provide resources to sustain future generations of older individuals [3].
Widening the lens, Börsch-Supan’s exploration into the impact of population aging on the labor
market and public policy in Germany provides valuable insights. The need for enhanced human
capital formation through education and training is emphasized to address the decline in the
economically active population. Changes in the age structure are shown to affect the demand for
goods, necessitating labor mobility to adapt to shifting employment patterns across different sectors
of the economy. While population aging and decline are related phenomena, examples from the USA
illustrate that they need not always occur simultaneously [4].
Ince Yenilmez’s study delves into the ripple effects of population aging, with a focal point on the
labor market and care sector. Their research dissects the consequences of declining fertility rates and
escalating life expectancies, resulting in a burgeoning elderly population and an ensuing labor
shortage. The policy measures they advocate, including extending retirement age and promoting
women’s labor force participation, emphasize a proactive approach to managing demographic
transformations. Alongside this, the importance of comprehensive social protection programs
emerges as a cornerstone to address the multifaceted challenges posed by population aging [5].
Meanwhile, Foot’s exploration in the Canadian context paints a unique picture, unveiling how
demographic and economic dependency levels are at historical lows due to augmented labor force
participation rates. This suggests the potential of increased engagement of older workers in mitigating
the effects of population aging, providing an alternative perspective on sustaining economic stability
[6].
Turning to Lisenkova et al.’s dynamic model of Scotland, the intricate effects of labor force decline
and aging on the labor market and macroeconomic variables become evident. Their evaluation
highlights the connection between aging and government expenditure, with population aging leading
to lower output. This provides a nuanced perspective on the multifaceted ramifications of aging
dynamics [7].
From the paper by Mehri et al., it becomes evident that Iran is grappling with rapid population
aging due to declining fertility rates and increased life expectancy. The projected statistics paint a
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231045
8
significant transformation where around 31% of the population is expected to be aged 60 and older
by 2050. In the face of these shifts, crucial concerns around gender-related issues, socio-economic
security for the elderly, and holistic societal adaptations come to the forefront. Mehri et al. aptly
emphasize that while population aging poses challenges, it should not be seen as a catastrophe; rather,
it demands a comprehensive response that addresses its impact across various societal sectors,
including the economy, pension system, policies, labor force, and healthcare [8].
Turning the spotlight to strategies that embrace an aging workforce, Denton and Spencer’s
research emphasizes the policy-driven potential of tapping into the skills and expertise of older
workers. Their analysis underscores the pivotal role that policy initiatives play in encouraging labor
force participation among seniors, thereby mitigating the adverse consequences of population aging
on living standards [9]. In parallel, Cristea et al.’s investigation traverses the complex intersections
of aging, public health expenditure, and labor market performance within the European Union. Their
study reveals substantial divergences across developed and developing countries, emphasizing the
importance of balanced policies to address the intricate web of challenges and opportunities that arise
from the confluence of population aging and health-related dynamics [10]. Collectively, this body of
research provides a rich tapestry of insights, weaving together the multifaceted interplays of aging
dynamics across labor markets, societal implications, and strategies for economic resilience and
inclusion.
1.3. Objective
This paper aims to study the complexities and opportunities associated with the aging global
workforce, offering insights to inform effective strategies and policies that promote sustainable,
inclusive, and dynamic workforce practices for the future.
2. Status of National Responses to Aging
2.1. Medical Care
Medical care for the aging exhibits a striking diversity across different countries, highlighting the
influence of distinct healthcare systems and strategic approaches. In the United States, a multifaceted
blend of public and private components collectively addresses the healthcare needs of the elderly
populace. Medicare, a prominent federally funded initiative, plays a pivotal role by extending health
insurance coverage to individuals aged 65 and above. This comprehensive program encompasses a
wide spectrum of medical services, including hospital care, medical treatments, and essential
prescription drugs. Complementing this, Medicaid steps in to offer coverage for seniors with low-
income backgrounds, while the presence of private health insurance options further enriches the
landscape, affording seniors a range of choices based on their preferences and needs.
Turning the spotlight to Western Europe, a predominant reliance on publicly funded healthcare
systems funded through taxation is evident. This foundational financial structure facilitates
comprehensive healthcare coverage for aging citizens. This encompasses essential services like
doctor consultations, hospital treatments, and long-term care provisions. Notably, the United
Kingdom’s National Health Service (NHS) stands out as a sterling example. Through the NHS,
healthcare services, including specialized care for the elderly, are extended without the imposition of
direct charges, ensuring equitable access for all. Similarly, the healthcare paradigm in Germany is
underpinned by a social insurance-based framework that underscores expansive medical coverage for
its aging demographic. The harmonization of these elements forms a cohesive structure that addresses
the unique healthcare needs of older adults.
Meanwhile, Japan’s progressive healthcare system entails a universal health insurance mechanism
that offers comprehensive coverage for medical expenditures incurred by senior citizens. This system
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231045
9
is fortified by the groundbreaking Long-Term Care Insurance (LTCI) system, a hallmark initiative
that provides essential support to the elderly requiring extended care, irrespective of the care setting.
In the rapidly evolving healthcare landscape of China, where a burgeoning aging population
necessitates nuanced strategies, distinctive healthcare programs have emerged. Notable examples
encompass the Basic Medical Insurance System catering to urban elderly residents, and the New
Rural Cooperative Medical Scheme, specifically designed for seniors residing in rural areas.
2.2. Social Security
Social security programs for the aging population in the United States, Western Europe, Japan, and
China display both similarities and distinct characteristics. In the United States, social security serves
as a vital safety net, providing financial assistance to retirees and disabled individuals. Contributions
made during a person’s working years culminate in benefits upon reaching retirement age. These
benefits can take the form of Social Security Retirement Benefits or Social Security Disability
Insurance (SSDI), addressing retirees’ and disabled individuals’ needs respectively.
Across Western Europe, comprehensive social security systems form a cornerstone of elderly
support. The United Kingdom’s State Pension guarantees a consistent income to individuals attaining
the state pension age, the sum determined by their National Insurance contributions. Germany has
fostered a multi-faceted pension system encompassing public, occupational, and private pensions,
ensuring a diversified retirement income strategy.
Japan’s social security structure centers on pensions and welfare initiatives. Pensions from
programs such as Employees’ Pension Insurance and the National Pension System provide financial
stability in retirement, dependent on contribution history and age. In addition to pensions, Japan offers
welfare services such as nursing care insurance and long-term care insurance to cater to the elderly’s
specialized needs. China, confronting demographic shifts, has been reshaping its social security
apparatus to cater to its aging population. The Basic Old-age Insurance for Urban Workers and
Residents anchors urban retirees’ financial well-being, contingent on their employment history and
contributions. Meanwhile, rural areas implement their pension and social security programs, tailored
to support elderly individuals in less urbanized settings.
2.3. Employment Landscape
The employment landscape for the aging population is undergoing transformations in the United
States, Western Europe, Japan, and China, driven by diverse factors and responses. In the United
States, a notable trend emerges as older adults increasingly opt to work beyond the traditional
retirement age, motivated by financial stability and personal fulfillment. Nevertheless, challenges like
age discrimination and skill relevance pose obstacles for older workers. The Age Discrimination in
Employment Act (ADEA) in the US addresses age-based employment discrimination, safeguarding
individuals aged 40 and above. Meanwhile, forward-looking companies adopt strategies such as
flexible work arrangements, targeted training, and phased retirement options to retain and attract
experienced employees.
Western Europe presents a mosaic of labor markets, where nations implement varying policies to
bolster older workers’ employment prospects. The Netherlands introduced the “Life Course Savings
Scheme,” allowing employees to accumulate funds for early retirement or reduced work periods.
Germany pioneers “flexible retirement,” enabling older workers to gradually scale down their hours
while receiving partial pension payments.
Japan’s rapidly aging demographic has spurred initiatives for prolonged workforce participation.
Embracing the concept of “active aging,” the Japanese government enforces the “Employment of
Elderly Persons Act,” promoting extended work lives by incentivizing companies to raise the
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231045
10
retirement age. Moreover, innovative workplaces prioritize age-friendly environments with
ergonomic designs and adaptable work setups. China grapples with demographic shifts, advocating
an employment-friendly ecosystem for its aging population. The incremental increase in the
retirement age aligns with the aspiration for extended work lives. Governmental encouragement
propels employers to invest in skill enhancement and training programs for older employees,
enhancing their employment ability.
However, employment scenarios are complex and multifaceted, intertwined with cultural norms,
economic contexts, labor regulations, and governmental policies. Opportunities for older workers
diverge across sectors and regions. For real-time insights into the nuanced employment environment
for the aging populace in these nations, recent reports, official sources, and labor market studies serve
as essential resources. Understanding and adapting to this evolving landscape empowers societies to
harness the experience and expertise of their senior citizens, ensuring a resilient and inclusive
workforce for the future.
In addressing the challenges of aging populations, countries’ approaches to healthcare, social
security, and employment reflect their distinct priorities. The United States balances public and
private components through Medicare, Medicaid, and private insurance, catering to seniors’ diverse
healthcare needs. Western Europe champions publicly funded healthcare systems, exemplified by the
UK’s NHS and Germany’s multi-layered pension strategy.
Japan’s healthcare model emphasizes universal coverage and its pioneering LTCI system,
promoting “active aging.” China adapts social security systems to accommodate extended work lives
and urban and rural disparities. Employment environments are equally diverse. The US enforces anti-
age discrimination laws and encourages flexible work arrangements. Western Europe innovates
phased retirement options and savings schemes. Japan advocates “active aging” and ergonomic
workplaces, while China promotes skill enhancement for seniors.
3. Discussion
3.1. Difference
In the discourse surrounding the intricate issue of aging populations, a comparative analysis of
healthcare, social security, and employment strategies across the United States, Western Europe,
Japan, and China illuminates distinct approaches that unveil both strengths and areas for growth
within each nation’s context.
Commencing with healthcare, the United States stands as a formidable contender with its
multifaceted healthcare system catering to the needs of the elderly. Notably, the federal initiative
Medicare extends comprehensive health insurance coverage to individuals aged 65 and above,
underscoring a commendable commitment to addressing geriatric healthcare. However, the
intricacies of the American healthcare system can at times manifest as a labyrinth of complexity,
resulting in coverage gaps and challenges in equitable accessibility, thus marking an area for
improvement. Conversely, Western Europe, with the Netherlands as a prime exemplar, has embraced
publicly funded healthcare systems funded through taxation. This model ensures that essential
medical services are accessible to aging citizens without substantial financial burdens, reflecting a
best practice. Yet, the United Kingdom’s National Health Service (NHS), though noble in its aim,
grapples with resource constraints, compromising its effectiveness in delivering timely care.
Meanwhile, Japan’s progressive healthcare model, rooted in the principles of universal coverage and
supported by the innovative Long-Term Care Insurance (LTCI) system, sets an impressive benchmark.
China, amidst its rapid healthcare evolution necessitated by a burgeoning aging population, is in the
process of tailoring healthcare programs like the Basic Medical Insurance System to urban and rural
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231045
11
elderly residents. Nonetheless, urban-rural disparities persist as an area of concern, warranting further
attention.
Transitioning to the realm of social security, Germany’s multi-layered pension system
distinguishes itself as a well-rounded approach within the Western European context. The integration
of public, occupational, and private pensions creates a diversified foundation for retirees’ financial
stability, exemplifying a best-case scenario. However, sustainability concerns loom over this
approach, casting a shadow on its long-term effectiveness. Parallelly, Japan has formulated a
comprehensive social security structure centered on pensions and welfare initiatives, with programs
like Employees’ Pension Insurance and the National Pension System offering a multi-faceted safety
net. In contrast, China’s dynamic landscape is in the midst of reshaping its social security apparatus
to accommodate an aging population. The Basic Old-age Insurance for Urban Workers and Residents
stands as a substantial step, yet disparities persist between urban and rural settings, warranting a
comprehensive resolution. Meanwhile, the United States’ social security system acts as a vital safety
net for retirees and disabled individuals, providing financial assistance. However, questions of
sustainability and adequacy often cloud its efficacy.
The United States presents a compelling narrative as older adults increasingly opt to extend their
working lives beyond traditional retirement ages, driven by financial stability and personal fulfillment.
This growing trend is supported by anti-age discrimination laws and initiatives promoting flexible
work arrangements, positioning the United States favorably. However, the challenge of age
discrimination remains an ever-present concern. Within Western Europe, the Netherlands introduces
the innovative “Life Course Savings Scheme,” allowing employees to accumulate funds for early
retirement or reduced work periods, offering a flexible approach. Nevertheless, the widespread
effectiveness of such schemes varies across the region. Japan’s proactive stance towards “active aging”
is evident through the “Employment of Elderly Persons Act,” fostering longer working lives and age-
friendly workplaces. China, grappling with demographic shifts, is advocating for an employment
landscape conducive to extended work lives, with incremental retirement age increases and skill
enhancement programs. The integration of such programs into the broader employment landscape
remains a dynamic process.
Each country navigates the complexities of aging populations with unique strategies, showcasing
strengths and areas for enhancement. By synthesizing best practices and addressing challenges, these
nations can collectively contribute to a global discourse on effectively managing the multifaceted
challenges presented by aging populations. As the world’s demographic landscape evolves, these
comparisons offer valuable insights into tailoring policies that prioritize the well-being and
contributions of older citizens while forging paths toward a more inclusive and sustainable future.
3.2. Directions for Future Improvement in China
China, with its rapidly aging population, can draw valuable insights from the approaches of the United
States, Western Europe, and Japan. The experiences of these countries underscore the importance of
balancing public and private healthcare components to provide comprehensive coverage. China’s
distinctive healthcare programs, such as the Basic Medical Insurance System, should continue to
adapt to cater to the diverse needs of urban and rural elderly residents.
In terms of social security, China’s evolving social security programs should address urban and rural
disparities while ensuring financial stability for retirees. Learning from Western Europe’s multi-
faceted pension strategies and Japan’s specialized care initiatives, China can enhance its social safety
net to support its aging population effectively.
Regarding employment, China’s emphasis on skill enhancement aligns with the global trend of
encouraging extended work lives. By incorporating age-friendly workplaces and phased retirement
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231045
12
options, China can harness the expertise of its senior citizens and contribute to a resilient and inclusive
workforce.
The diverse strategies adopted by the United States, Western Europe, Japan, and China to address
aging populations’ challenges reflect their unique priorities and contexts. Drawing from the strengths
of each approach, China can shape its future direction by further refining its healthcare, social security,
and employment policies to foster a thriving environment for its aging citizens.
4. Conclusion
In this paper, a comprehensive comparative analysis of healthcare, social security, and employment
strategies for aging populations in the United States, Western Europe, Japan, and China has been
conducted. The primary objective was to discern the unique approaches each country employs to
address the challenges posed by their aging demographics and to identify both strengths and areas for
improvement within their respective systems.
The main findings of this study underscore the remarkable diversity in strategies and policies
adopted by the four regions. In healthcare, the United States combines public and private components,
with Medicare playing a pivotal role, while Western Europe relies on publicly funded systems like
the UK’s NHS and Germany’s multi-layered approach. Japan’s universal coverage and pioneering
LTCI system set it apart, and China is evolving its healthcare programs to cater to urban and rural
elderly residents. Social security systems vary from Germany’s multi-faceted pensions to Japan’s
comprehensive structure, while China seeks to bridge urban-rural disparities. The United States’
social security system serves as a safety net, but sustainability concerns persist. Employment trends
show older adults working beyond retirement age in the United States, while the Netherlands
introduces innovative schemes, Japan promotes “active aging,” and China focuses on skill
enhancement.
The suggestions and implications derived from these findings highlight the potential for cross-
country learning and adaptation. Each country can draw inspiration from the best practices of others
to enhance its own strategies. Collaboration and the exchange of experiences can drive innovative
solutions to the challenges of aging populations, fostering resilient healthcare systems, inclusive
social security, and age-friendly employment environments. The paper’s insights emphasize the
importance of continually refining policies to address the evolving needs of the elderly while
capitalizing on their contributions to society. In a world undergoing demographic shifts, this
comparative study provides valuable insights for nations striving to create a harmonious and
sustainable future for their aging populations.
References
[1] Pit, S., Fisk, M., Freihaut, W. et al. COVID-19 and the ageing workforce: global perspectives on needs and solutions
across 15 countries. Int J Equity Health 20, 221 (2021). https://doi.org/10.1186/s12939-021-01552-w
[2] Vodopivec, M., Arunatilake, N. Population Aging and Labour Market Participation of Old Workers in Sri Lanka.
Population Ageing 4, 141–163 (2011). https://doi.org/10.1007/s12062-011-9045-5
[3] Serban, A. C. (2012). Aging Population and Effects on Labour Market. Procedia Economics and Finance, 1, 356-
364. https://doi.org/10.1016/S2212-5671(12)00041-X
[4] Börsch-Supan, A. (2003). Labor Market Effects of Population Aging. LABOUR, 17, 5-44.
https://doi.org/10.1111/1467-9914.17.specialissue.2
[5] Ince Yenilmez, M. Economic and Social Consequences of Population Aging the Dilemmas and Opportunities in the
Twenty-First Century. Applied Research Quality Life 10, 735–752 (2015). https://doi.org/10.1007/s11482-014-
9334-2
[6] Foot, D.K. Public expenditures, population aging and economic dependency in Canada, 1921–2021. Popul Res
Policy Rev 8, 97–117 (1989). https://doi.org/10.1007/BF00124282
[7] Lisenkova, K., Mérette, M., & Wright, R. (2013). Population ageing and the labour market: Modelling size and age-
specific effects. Economic Modelling, 35, 981-989. https://doi.org/10.1016/j.econmod.2013.09.007
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231045
13
[8] Mehri, N., Messkoub, M. & Kunkel, S. Trends, Determinants and the Implications of Population Aging in Iran.
Ageing Int 45, 327–343 (2020). https://doi.org/10.1007/s12126-020-09364-z
[9] Denton, F. T., & Spencer, B. G. (2009). Population aging, older workers, and Canada’s Labour Force. Canadian
Public Policy, 35(4), 481–492. https://doi.org/10.3138/cpp.35.4.481
[10] Cristea, M., Noja, G. G., Stefea, P., & Sala, A. L. (2020). The Impact of Population Aging and Public Health Support
on EU Labor Markets. International Journal of Environmental Research and Public Health, 17(4), 1439.
https://doi.org/10.3390/ijerph17041439
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231045
14
The Influence That Jeonse Rental System Brought to the
Housing Bubble in Korea
Lancheng Fei1, a, *
1Panyaden International School, Chiangmai, 50230, Thailand
a. 2012004@panyaden.ac.th
*corresponding author
Abstract: The bubble burst, and the South Korean property market plummeted. Thousands of
Koreans should not be able to afford their debts and choose to commit suicide. Prior to this,
South Korea’s real estate market soared rapidly, and no one noticed the abnormality and the
future direction of the real estate market. That was a miracle, but it only lasted for a while,
and one of the most important factors that accelerated its destruction was a rental system
called Jeonse. The purpose of this paper is to explore the impact of Jeonse, a unique rental
system, on the Korean real estate market. Housing prices in South Korea have soared amidst
the global pandemic. By studying macroeconomics, the global capital chain, and policy-
related factors that affect the market, it seems necessary for the government to understand the
importance of controlling leverage and properly using quantitative easing. Those methods
can theoretically prevent the economic crisis, slow down the creation of bubbles, or decrease
the influence made by the contagions it creates.
Keywords: South Korea, housing market, Jeonse, leverage, capital chain
1. Introduction
From the economic crisis in 2008 to the ups and downs of Japanese real estate, to the current collapse
of South Korea’s real estate market. For decades, the real estate market has seemed to be a place of
opportunity and pitfalls. Many ordinary people made money either because of luck or because they
really seized the opportunity, but at the same time, many ordinary people also lost everything in the
chaos of real estate. Therefore, understanding the real estate market and real estate bubbles is
beneficial for both individuals and institutions. Understanding market trends and bubbles can help
investors seize opportunities and obtain good returns; buyers can buy at low prices and avoid
excessive consumption; governments can optimize policies and maintain market stability. However,
ignoring the housing bubble could trigger a serious crisis. The bursting of the bubble could lead to
the collapse of the financial system and affect the economic stability of the entire region. Purchasing
overvalued properties when a bubble is about to burst is also likely to leave individuals in financial
distress. So now that the real estate market has become an irreplaceable concept, an unignorable
opportunity, it seems all the more important that people care about it.
The discovery of housing in Korea started in 1968. Burns and Tjioe see the concept of housing as
a tool but not a final target and introduce the International Housing Productivity Study to assess the
social and economic returns of housing investment in monetary terms. Finally, they stated that
investment in housing should be viewed as a productive tool for development, producing outputs
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231046
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
15
similar to investments in health and education. Their preliminary results at that time suggest that
housing improves human resources and can compete as a productive investment. This perspective
should guide future resource allocation and policy decisions [1].
Now, there are already a lot of researchers who have decided to put their eyes on the housing
market in Korea. One of the most popular areas that people chose to research on is the macroeconomic
and microeconomic related to the housing market. Back in 2003, Yoon examined issues such as rising
deposit costs worldwide, economic changes, low-interest rates, and redevelopment projects, as well
as addressing challenges such as volatile rent-to-let prices, security of tenure, and disputes between
landlords and tenants. Investigate the factors leading to the rise of the full-rent system and shift to
monthly rent in the Korean housing market. It was determined that the increase in the price of full
rent and the conversion of monthly rent affected the affordability of tenants. Instability in the pass-
on rental market, evictions of tenants during rising house prices, and disputes between landlords and
tenants are prominent issues. Differences in payment method preferences further complicate the rental
housing market [2]. Also in 2004, Kim analyzed data on residential investment, housing wealth,
housing finance evolution, price trends, and their macroeconomic linkages. Thereby investigating the
relationship between housing and the Korean economy, focusing on investment, housing financing,
price trends, and policy implications. Finally, it was stated that housing is crucial to the Korean
economy, accounting for a large part of GDP and investment, and also affecting housing price trends.
Government intervention has also been suggested to affect housing supply and price volatility. Real
estate market dynamics affect consumption, inflation, and macroeconomic conditions[3]. Lately, in
2023, The author Suh. employed a general equilibrium model that incorporates two region- and
housing-market-specific shocks. Simulations analyze consumption, housing services, debt, and
welfare impacts. Study the macroeconomic consequences of different housing cycles in different
regions and assess the impact of stabilization policies. Finally, it finds that housing cycles in specific
regions create economic inequality, asymmetrically affecting consumption, housing services, debt,
and welfare. Housing supply policies are beneficial to affected areas, countercyclical loan-to-value
ratio policies are unfavorable to borrowers, and headwind monetary policies have a limited effect on
stabilizing housing prices in areas [4].
Another relatively popular area is to focus on “speculation”. Back in 2004, researcher Chung. And
Kim worked on using policy to combat speculation using regression analysis and three estimation
techniques. Emphasizes the influence of speculation in Gangnam District, compares it with the
Japanese real estate bubble, and proposes anti-speculative measures. The presence of foam was
verified using different methods. Finally, they find that speculative demand significantly drives up
housing prices; bubbles exist to varying degrees. Speculative demand outpaced normal demand,
driven by falling interest rates, rising money supply, and loose housing policies [5]. Same in 2009,
the author Xiao. and Park. analyzed the relationship between the apartment house price index and
GDP, wages, population, and rational bubble proxies from 1998 to 2006. Aiming to investigate the
role of rational speculative bubbles in the sharp rise in apartment house prices in Seoul, South Korea.
Assess the impact of fundamental and speculative demand factors on house prices. In the end, it was
found that rational speculative bubbles drove the surge in apartment prices in Seoul. Condominium
prices are more responsive to fundamental and speculative demand than townhouses or single-family
homes [6].
Noticeably, politics-related areas are also a field that people chose to dig into. In 2020, the author
Jang. And Song. wishes to evaluate the South Korean government’s policies to stabilize the housing
market and prices by analyzing apartment prices from 2012 to 2019, concerning policy and
institutional changes. Finally, three results are emphasized: the policy restrains the price rise,
improves the market efficiency, and makes the price tend to be balanced, which proves the
effectiveness of the policy in stabilizing prices and improving the environment [7]. And lately, in
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231046
16
2023, the author, Kim examines the impact of South Korea’s housing policy on the housing market
in the Gangnam region. By using the AITS-DID approach, it is found that policies do not stabilize
prices, leading to unintended consequences. Finally, self-sufficient communities require long-term
urban planning to address housing market imbalances [8].
Other than these differing research orientations, there is also research on land use and city designs.
In 2017, Kang evaluated the effect of spatial accessibility and centrality on housing prices in Seoul,
South Korea by using a multilevel hedonic price model combined with street configuration and
building density indicators to assess the impact of land use on housing prices (2010). Finally, the
mixed-use of land is emphasized in the conclusion, highlighting the need for an efficient transport
network, buffer zones are crucial to mitigate the negative impact of industrial land use, etc [9].
Another study by Yun also presents a youth universal rental housing program in South Korea that
utilizes a unique universal tenure system. It demonstrates how the program can have a positive impact
on beneficiaries without negatively impacting the community. Adopting the hedonic price model and
the hierarchical linear model, it highlights that the plan combines the policy characteristics of the
demand side and the supply side. This innovative attempt provides new insights into housing
challenges [10].
There were not many people focused on the term Jeonse and the influence it brings to the housing
market. Therefore, this paper will focus on the concept “Jeonse”, and analyze from different
perspectives linking to the bubble it creates and what the government can do to deal with it.
2. Jeonse Description
Jeonse is a method of leasing real estate that has become particularly popular in Korea in recent years.
On the surface, this form of leasing seems to be no different from free. Under this structure, when the
deal is closed, tenants will pay the landlord a deposit of 50% to 80% of the purchase price. It is worth
noting that when the landlord receives the deposit, they have the right to do anything with the money.
After this process, tenants can move in at any time. The important thing is that when the rental period
expires, the landlord will be required to return the deposit received from the tenant when renting the
house to the tenant intact. At this point, it looks like the tenant is living in someone else’s house for
free for a while.
3. The Evolution of Jeonse
Since the 1900s, the structure of Jeonse has already existed. However, after the establishment of the
Republic of Korea in 1958, governments acknowledged its legality and clarified some rules within
the structure of Jeonse in their civil code. Therefore, seeing 1958 as a threshold, this new and special
method of house renting started gradually getting more popular. With the rapid economic growth of
South Korea in the 1970s, a large number of people flooded into the city. The model of Jeonse quickly
became popular. Due to the serious shortage of urban housing, and the situation a lack of housing
forms like public rental housing that can replace Jeonse while the housing price was significantly
increasing. Like this, until the 21st century, a majority of the population in Korea already decided to
put their money on Jeonse to stably gain more benefits from it. Plus the increase in the demand
specifically for Jeonse was caused by the lower interest rates every year. Until 2020 when the covid
- 19 reached a climax globally, the interest rate in Korea already was lowered to 0.5% by the
government, and this is also a significant reason for the creation of the housing bubble.
Simultaneously, this movement caused more people to take their assets out of banks and put them
into the market of Jeonse.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231046
17
4. A Utopian Benefit Triangle? Three Winners at the Same Time?
For a normal resident in Korea, it is worth it if they only need to pay some rental deposit that is going
to be given back when renting time runs out, then basically you can live in the house for free. In this
situation, tenants are usually happy about the deal. On the other hand, as a landlord or the owner of
the house. In a situation where the interest rate is extremely low and using the deposit from tenants
to start more cycles, Jeonse can gain comparatively more benefits than leaving those deposits in
banks. Therefore, with extremely high leverage, a lot of people decided to focus on Jeonse and bought
a big number of houses. As a result, the majority of them made a lot of money with the cycle of
buying a house and using Jeonse to attract tenants. In this situation, most landlords are satisfied with
their results. At the same time, banks also gained a lot of benefits according to the extremely broad
range of their housing-related business scale. Most banks get a huge amount of money based on the
spread of income. That can mostly replace the loss of money deposited by people. And in this
situation, these banks are staying very positive about their financial results. Certainly, if the housing
prices are not falling, the three sides: tenants, landlords, and banks would always be in a situation of
triple-win, where everyone is gaining profit. However, the reaction of the Korean government to
prevent losses when the FED was raising their interest rates was to increase their interest rate. When
the interest rate got higher, it might be more valuable to put money into the bank than doing any
Jeonse that has 0 income from interest.
5. Fall
The triple-win situation of the housing market in Korea created by the system of Jeonse dramatically
raised the housing price. The housing index in Korea rose from around 80 in 2020 to more than 100
after 2022. But these rises are only because of the low interest rate in Korea. Therefore, when the
interest rate was raised by the government in 2022, the triple-win would surely be destroyed. The
Jeonse creates a capital chain that seems ideal and sustainable, but behind the appearance, when the
housing market starts falling, the landlord would likely be in danger. With the increases in the interest
rate, a lot of tenants chose to take their money back and put them back into the bank. But at the same
time, landlords who use high leverage to buy more houses are now under a higher debt. This caused
a lot of them to lose the ability to give tenants their deposits back as they should. With that, a lot of
landlords flee with tenants’ money, whereabouts unknown. This again leads to a disaster on the side
of tenants. Those who get loans to pay for the Jeonse housing then get no payback from the landlord,
so they are in trouble paying the loans to banks. Therefore, the capital chain ruptures, bubbles burst,
and the entire housing market collapses. Those disasters became the haze over Korea, a huge number
of people deficit and lost everything, and many of them chose to end their lives themselves. And
looking back to the bubble at the beginning of everything, the Jeonse system is truly a factor that
accelerated the rise of the housing market, and the creation of the bubble that destroyed everything in
the end.
During the creation of the bubble, the Korean government did not notice that there would be a
dangerous bubble soon caused by the Jeonse rental system. The Covid-19 pandemic forced Korea to
lower its interest rate. However, even though they knew that lower interest rates meant higher market
liquidity, they were ruling the entire housing market with a relatively loose policy. In this kind of
economic environment, a lot of people chose to do something about leverage and believed that they
were soon going to be rich. Without many kinds of protective behavior in the housing market, the
housing price increases dramatically with the fake “ideal benefits triangle”. Not long after that, the
bubble was finally ready to be popped.
To effectively avoid the creation of a bubble when the “ideal benefits triangle” is increasing the
housing price rapidly, the supervision of the risk of capital chains is relatively important. The collapse
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231046
18
of the capital chain is one of the major issues that pop the bubble. When landlords lose the ability to
pay back the deposits back, tenants would accordingly have no money to pay for their loans from the
bank as they should get that money from their landlords. At the same time, plenty of tenants fall into
the same trap, so this somehow pushes the problem into a regional issue. And to avoid this, a harsh
law is surely necessary to solve the problem with power. The government plays an irreplaceable role
in the economic system, it is usually the powerful player that has authority to change the foundation
of the system. Therefore, with some rules set by the government, it might be possible to actually avoid
the disaster.
To find some ways that effectively avoid the collapse of a capital chain, looking at the risk of
trades and business could be a solution. Therefore, “leverage”, the degree that measures risks could
be a spot to focus on. When the leverage is high, it means that the loan on this person is a lot more
than the money he had originally. Simultaneously, when the leverage is lower, the risk of the business
is comparatively lower. Therefore, the aim of the regulation is simply to supervise and control the
degree of leverage people take to try to gain profits. With a limited leverage degree, landlords will
not be able to expand their range of business by getting loans from banks. On the other hand, it will
also prevent a large number of tenants from getting into the Jeonse market and renting houses with
the money of the bank. With the regulation, the creation of a housing bubble will be a lot slower
according to the low leverage. Therefore, the risk of an economic collapse would be proportionally
lower when limiting the degree of leverage.
On the other hand, imagine that the bubble has already popped. There should be other ways for
the government to control the disaster and minimize the loss. Before digging into the field of networks
to find something useful, it is necessary to understand that since the time when the bubble was first
created, it will surely explode someday in the future. There is no way to make a housing bubble
completely disappear, what can be done is just to minimize the influence it causes by setting
something up in a limited amount of time. Between the influence of the pooped bubble spread along
the entire country, there is a short period for the ruler to react and make serious decisions. Mentioned
that no one is able to completely stop the influence after it is already there, so the target here is just
to somehow reduce the impact of the contagion, and find several methods to achieve a soft landing
in the crisis. Eventually, the economic disaster won’t damage the economy entirely.
If governments want to deal with the aftermath of the contagion, they have to lose some other
things to fix the issue. The main problem of this contagion is the lack of market liquidity caused by
the collapse of the capital chain within the “ideal benefits triangle”. Aiming to increase the market
liquidity, an easy way is to simply decrease the interest rate again. However, if the gap between the
Korean interest rate and the rate of the FED is too big, a lot of funds in Korea will flow to the US.
This is surely going to somewhat make the situation better, but compared to preventing the
consequences of a bubble in the housing market, it is more important to follow the US raising their
own interest rate to prevent capital outflows, maintain exchange rate stability, and curb inflation.
Other than decreasing the interest rate directly, there is another emergency method to use to increase
the market liquidity. It could work if the government directly injected money into the market, which
is “Quantitative easing”. Noticeably, this method is also going to decrease the interest rate, but the
degree of interest rate it lowers is a lot less than directly decreasing the interest rate. At the same time,
during quantitative easing, the central bank also buys government bonds or financial assets to increase
the money supply in the market. This stimulates the market that is in danger just like doing CPR to a
shocked person. With the risk of inflation and the consequences that money is flowing to the US,
Quantitative easing would theoretically decrease the impact of the contagion made by the bubble and
achieve a soft landing. Overall, quantitative easing is a possible solution facing the contagions made
by the popped bubble, but at the same time, it would also damage the economic system to various
degrees. Governments need to weigh the pros and cons of doing that and make their choice.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231046
19
6. Conclusion
Until now, the paper discussed what happened during the bubble in the housing market in Korea, how
Jeonse accelerated and strengthened the power of the contagions after the bubble, what the possible
preventive measures for the creation of the bubble, and the solutions that can save the market after
the bubble is popped. During the creation of the bubble, the Joence system caused the “ideal benefits
triangle” where the tenants, landlords, and banks were all happy about their gains. However, as soon
as Korea decreases its interest rate, the high degree of leverage directly breaks the capital chain. The
dangerous contagions created by the high degree of connection within the benefits triangle quickly
spread throughout the entire Korea. On the other hand, to effectively prevent it, controlling the
leverage would effectively lower the risk of collapsing capital chains. If the bubble has already
popped, quantitative easing is the last solution that would simulate the market and carry out a
successful soft landing. In the end, economic crises seem like something that’s unavoidable or it is
simply associated with the development of the financial system. However, I believe that since more
researchers are starting to dig into this field, there will be ways found to eventually predict a crisis or
a bubble.
References
[1] Burns, L. S., & Tjioe, B. K. (1968). Housing and human resource development. Journal of the American Institute of
Planners, 34(6), 396–401. https://doi.org/10.1080/01944366808977568
[2] Yoon, J. (2003). Structural changes in south korea’s rental housing market: The rise and fall of the jeonse system.
The Journal of Comparative Asian Development, 2(1), 151–168. https://doi.org/10.1080/15339114.2003.9678376
[3] Kim, K.-H. (2004). Housing and the Korean economy. Journal of Housing Economics, 13(4), 321–341.
https://doi.org/10.1016/j.jhe.2004.09.001
[4] Suh, H. (2023). Regionally heterogeneous housing cycles and housing market stabilization policies: Evidence from
Korea. Economic Modelling, 120, 106192. https://doi.org/10.1016/j.econmod.2023.106192
[5] Chung, H. S., & Kim, J. H. (2004). Housing speculation and housing price bubble in korea. SSRN Electronic Journal.
https://doi.org/10.2139/ssrn.535882
[6] Xiao, Q., & Park, D. (2009). Seoul housing prices and the role of speculation. Empirical Economics, 38(3), 619–
644. https://doi.org/10.1007/s00181-009-0282-x
[7] Jang, H., Song, Y., & Ahn, K. (2020). Can government stabilize the housing market? The evidence from South Korea.
Physica A: Statistical Mechanics and Its Applications, 550, 124114. https://doi.org/10.1016/j.physa.2019.124114
[8] Kim, C., & Ko, J. (2023). Unintended consequences of housing policies: Evidence from south korea. Sustainability,
15(4), 3407. https://doi.org/10.3390/su15043407
[9] Kang, C.-D. (2017). Effects of spatial access to neighborhood land-use density on housing prices: Evidence from a
multilevel hedonic analysis in Seoul, South Korea. Environment and Planning B: Urban Analytics and City Science,
46(4), 603–625. https://doi.org/10.1177/2399808317721184
[10] Yun, S. (2020). Neighborhood effects of housing program using Jeonse in Korea. International Journal of Housing
Markets and Analysis, ahead-of-print(ahead-of-print).
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231046
20
The Value and Promotion Role of Opinion Leaders in Media
Communication
Yixuan Pei1, a, *
1BDA School of The High School Affiliated to Renmin University of China, Beijing, China
a. unberlee.pei@stu.nwupl.edu.cn
*corresponding author
Abstract: The research background of this paper is the continuous development of modern
science and technology, the continuous popularization of the Internet links people in various
regions, and people who have more knowledge in each field gradually gain a lot of fans on the
Internet, and their comments are also recognized by more people. These people are called
opinion leaders. However, the value and influence that opinion leaders can bring are
gradually expanding and affecting our lives. The goal of this paper is to find out the influence
and value that opinion leaders can bring to different objects such as society, individuals and
companies. The main research content is what kind of value can opinion leaders create and
promote the development of public opinion by using their high appeal and credibility on the
Internet. And that affects the economy and politics and different aspects of the whole country.
The influence of opinion leaders is huge and can bring a lot of value, but whether ordinary
people can distinguish the truth and falsehoods of opinion leaders’ speeches is very important.
The independent thinking of ordinary people should not be obliterated by the big trends on
the Internet.
Keywords: social media, opinion leader, economic development, the effect on the mind
1. Introduction
1.1. Background
The Internet is developing more and more rapidly in today’s society. People exchange feelings and
express opinions on the Internet, learn about different people’s lives, learn more knowledge and
learn more news. And some people on the Internet get a lot more attention, and that’s the opinion
leaders on the Internet. Opinion leaders usually refer to individuals and organizations who are
familiar with the media and can better understand the issue because they have more product
information and professional knowledge in a specific field. They usually carry out secondary
dissemination. They are trusted because they have a certain level of professional information.
Through the information they publish on various channels such as social media, many people will
support their proposals or buy the products they promote because of their trust. This is the value and
promotion role of an opinion leader in the dissemination of social media. Not only that, on some
social platforms, if you have enough appeal and public trust in you, your words and posts can even
affect the economy and politics of the entire country to a certain extent.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231048
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
21
1.2. Related Research
Thomas describes a social media phenomenon that shows just how influential opinion leaders are
on the Internet. The article cited several examples, reflecting that having a high number of fans on
the Internet only needs a post to promote their own products or opinions to get a high heat, greatly
increasing the attention and sales of products. This is a very frightening appeal [1]. Alexandre
Barbeira taking Musk as an example, this paper introduces how the CEO of a company reflects the
professionalism and value of his company through the content published by his account and other
channels on social media. How to control the thoughts and emotions of shareholders on social
media to achieve the ultimate goal of profit for your company [2]. Juntiwasarakij introduces some
preferential treatment that celebrities get on the Internet. First, young people nowadays pay much
attention to successful people who can reflect their independent personality, such as self-made
people, and celebrities on the Internet will be allocated more attention and traffic when they express
their own views. Celebrity culture is becoming increasingly important and ingrained in the new
generation. The benefits and influence of celebrity titles are expanding [3].
Schafer and Taddicken show us that the influence of public opinion on social media is still very
great. People will still be influenced by the opinions of opinion leaders and most people, and tend to
be more accepting of others [4]. Valence and Davis present a new theory of innovation and describe
a new method of innovation dissemination by using the convincing power of opinion leaders. A
model is presented, and the advantages and limitations of this model are discussed and tested [5].
Meng et al. introduced that in online shopping, sometimes we pay more attention to the
recommendation and promotion content of some bloggers, rather than the feedback of buyers on the
network platform. This is a kind of consumption behavior influenced by opinion leaders [6].
Lin et al. show us how to make reasonable use of the appeal and conviction of an opinion leader
on an online network platform to accumulate popularity, increase exposure and create value for
promoting a product or service [7]. Bodendorf and Kaiser introduced us to the more and more
extensive aspects of data that are circulating on the Internet now, and you can learn about all aspects
of a company on the Internet. Through the numbers on the Internet, we can also understand how the
ideas and opinions of opinion leaders are generated, and at the same time, we can identify the data
authenticity of the words spoken by opinion leaders [8].
Dong and Zhang introduced to us how to find Opinion leaders in a virtual community. The
definition of Opinion leader was first introduced. Secondly, we can learn that Opinion leaders are
actually “messengers” of information. Product information is generally interpreted by Opinion
leaders before being transmitted to everyone. Therefore, Opinion leaders have a great influence on a
social media or network platform [9]. Winter and Neubaum introduce to us the emotions and
psychology of a person who plays himself as an opinion leader on social media, and how they are
eager to influence and change others through the dynamics posted on their account [10].
1.3. Objective
This paper aims to study what kind of value can the influence of opinion leaders on the Internet
bring. The second chapter studies the influence of opinion leaders on the value of assets. It includes
the impact of online opinion leaders’ promotion and evaluation on the price of commodities and the
impact of some company executives and partners’ promotion and communication on the stock
market value of a company. From these influences, we can find that the opinions of opinion leaders
on the Internet have caused changes in our lives. The third chapter studies some problems and
solutions about opinion leaders cheating consumers that often happen on the Internet. Many
bloggers publish their ideas freely on the Internet because they have a lot of fans, and their opinions
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231048
22
will be paid more attention by more people. Many people take bad product advertisements to
promote products whose quality and safety cannot be guaranteed.
2. The Influence and Reasons of Opinion Leaders’ Propaganda on Asset Prices
2.1. The Way, The Impact, The Path
The influence of the publicity of opinion leaders on asset prices is reflected in many aspects. With
the continuous development of science and technology and the progress of social media, people pay
more attention to the lives of celebrities and learn from them. A large part of the groups using the
Internet are young people and teenagers, they are more willing to accept new things and advocate a
free and independent way of life and work, so some independent celebrities and opinion leaders on
the Internet will get more attention from them.
Opinion leaders can post some videos and pictures on the Internet, or live broadcasts online to
show that they are using a certain product or that they trust a certain company. At this time, opinion
leaders have high visibility on the Internet and millions of fans, and because of their trust, many
people will buy the products promoted by opinion leaders or trust the stocks of enterprises. Few
people understood and chose to buy many goods before becoming the so-called network red goods,
but when it suddenly became popular on the Internet, such as “dirty bag”, “thumb Sheng Sheng
Bao” and other ordinary food, the prices under the packaging of network red and countless shop
bloggers can be bought from the original ten dollars, soaring to dozens of dollars at that time are
difficult to find a share. At the same time, being an opinion leader on the Internet also helps to
accumulate your credibility and reputation. When you become popular on the Internet and get more
attention and fans, you will also “appreciate” yourself.
2.2. Reason
The reason is that people are more likely to be influenced when facing a field that they do not know
very much about. In our concept, opinion leaders are “insiders” who have professional knowledge,
are familiar with products, and understand the market, and the choices they make are usually
reasonable and trustworthy. Therefore, we will be influenced by their opinions and suggestions to
our judgment, their accounts usually have a lot of fans, and their visibility is also high, it looks at
their perspective in the eyes of most people who are looking up, so it is difficult for us to doubt
what problems they make suggestions and say.
2.3. Case Analysis
During Kardashian’s pregnancy, she posted a Twitter feed touting the benefits of a morning
sickness medication she took during pregnancy. Because of her high popularity and visibility on
social media, her every post is very hot. Subsequent media reported that the production company of
this morning’s sickness medicine spent a huge amount of money on this dynamic, but at the same
time, it did receive the corresponding return. Since Kardashian posted this news, more people have
chosen the company’s drugs when buying morning sickness medicines, and sales have increased by
about 20 percent.
3. Problems and Solution
3.1. Problem
Now the information on the Internet is mixed, and many opinion leaders who play a large role in the
Internet will also publish some false information for their own interests on the Internet. It’s hard to
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231048
23
judge. Some commodities themselves have problems to a certain extent, and through the
propaganda of opinion leaders, many people will choose to ignore these problems. Many opinion
leaders post some politically biased words on social media, and many young fans or people who
don’t know much about this matter will blindly choose to follow the trend.
3.2. The Phenomenon on the Internet
3.2.1. Casual Evaluation
The fact that anyone can say anything online has led many people to believe that they are hiding
behind a virtual account and therefore can’t take responsibility for their words and actions. There is
a “dazui” person who has made many improper remarks on the Internet. These improper statements
may not only be false claims about a product, disinformation about a person, or even slander against
a country. The power of language is very powerful, when a person’s speech continues to ferment on
the Internet, there will be tens of thousands of people who see this news, and some of them may
choose to believe this news. This is a very dangerous effect.
3.2.2. Defraud Consumers
With the increasing number of influencers and celebrities on the Internet. There are more and more
successful people who are simply packaged for profit, and they promote the product to their fans by
taking advertisements on the Internet, hiding the problems and side effects of the product, and only
presenting the good side of the product to everyone. The safety of products on the Internet is not
guaranteed, and pirated or counterfeit products are often bought in the hands of some broadcast
rooms or celebrities. This has caused serious impact and harm to our life safety and health.
3.3. Prospects
It is mentioned in the mob that the way people think, the standard of judgment and even a person’s
personality can be swallowed up by the blindness of the group. When a person is a lonely individual,
he has the characteristics of a very distinct individual. But when he is integrated into a group of
people, that is, a group, the group’s thoughts and behaviors will affect him, so that his thoughts will
be replaced by the group’s thoughts.
4. Conclusion
This paper has studied the influence of opinion leaders on the public, including the sales volume of
goods, the public opinion of the society, and even the whole country. With the development of the
media, new opinion leaders are constantly emerging, and their appeal on the Internet is also
constantly expanding and improving. The main finding is that in this era of Internet development,
the words of opinion leaders can sometimes have a great impact. The promoted products will be
recognized by more people, and the same people who often voice and publicize the company on the
Internet will also drive up the company’s stock price. Now the value of opinion leaders has become
more and more great. People can lose their own ideas because they are carried away by what the
masses think. In the face of the collective consensus one’s own identity will be erased. When facing
opinion leaders on the Internet, we should also keep our own ideas and not be easily controlled and
influenced by the opinions of opinion leaders. Some opinion leaders have a certain degree of
commercial characteristics, so the words of opinion leaders may also be based on the goal of profit.
The Internet is not a place outside the law, and opinion leaders should also be cautious when
expressing their own opinions and promoting them, and not publish inappropriate views at will.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231048
24
References
[1] Thomas, K. (2019). Key opinion leaders supercharged by the internet. BMJ: British Medical Journal, 365.
[2] Corte, M. A. B. (2020). How social media usage by managers affects corporate value: the case of Elon Musk
(Doctoral dissertation).
[3] Juntiwasarakij, S. (2018). Framing emerging behaviors influenced by internet celebrity. Kasetsart Journal of Social
Sciences, 39(3), 550-555.
[4] Schäfer, M. S., & Taddicken, M. (2015). Opinion leadership| mediatized opinion leaders: new patterns of opinion
leadership in new media environments?. International Journal of Communication, 9, 22.
[5] Valente, T. W., & Davis, R. L. (1999). Accelerating the diffusion of innovations using opinion leaders. The Annals of
the American Academy of Political and Social Science, 566(1), 55-67.
[6] Meng, F., Wei, J., & Zhu, Q. (2011, May). Study on the impacts of opinion leader in online consuming decision. In
2011 International Joint Conference on Service Sciences (pp. 140-144). IEEE.
[7] Lin, H. C., Bruning, P. F., & Swarna, H. (2018). Using online opinion leaders to promote the hedonic and
utilitarian value of products and services. Business horizons, 61(3), 431-442.
[8] Bodendorf, F., & Kaiser, C. (2009, November). Detecting opinion leaders and trends in online social networks. In
Proceedings of the 2nd ACM workshop on Social web search and mining (pp. 65-68).
[9] Zhang, X., & Dong, D. (2008). Ways of identifying the opinion leaders in virtual communities. International Journal
of Business and Management, 3(7), 21-27.
[10] Winter, S., & Neubaum, G. (2016). Examining characteristics of opinion leaders in social media: A motivational
approach. Social Media+ Society, 2(3), 2056305116665858.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231048
25
Research on the 2008 Financial Crisis Regarding Financial
Network
Shiyun Ding1,a,*
1Shenzhen College of International Education, Shenzhen, 518043, China
a. s20654.ding@stu.scie.com.cn
*corresponding author
Abstract: The rapid withdrawal in the financial market due to the collapse of Silicon Valley
Banks in 2023 brought reflections on the 2008 financial crisis. Past research has focused on
modeling different types of financial networks that help explain bankruptcy’s contagion
process. This paper discusses how these networks help to explain the 2008 financial crisis,
followed by shedding light on the global impacts (European, Latin American, and Asian
markets) of the crisis in terms of networks. The last section further discusses potential policies
that could be used to prevent or predict severe contagion depending on different network
structures, including adjusting the level of portfolio diversification for international
investment, for example. This paper sheds light on the importance of financial networks
regarding their contagion channel and wide application in the real economy.
Keywords: network, financial crisis, interbank, contagion
1. Introduction
The recent failure of the Silicon Valley Bank and Signature Bank, followed by the First Republic
Bank, ranked within the top 30 by total assets among the biggest U.S. banks, signaled an alarm to the
global economic participants. Russell and Zhang [1] found that the total assets of these banks
surprisingly surpassed the total assets of defaulted federally insured U.S. banks in the 2008 financial
crisis after adjusting for inflation. The rapid withdrawal of banks for both cases indicates the
underlying network among financial institutions may help to explain the spread of global contagion.
To briefly introduce the 2008 financial crisis, two hedge funds owned by The Bear Stearns
Company (BSC) collapsed, leading to the Lehman Brothers (LB) stock price plummeting by 93% in
July 2007 [2]. As a heavily mortgage-backed securities US bank, LB faced bankruptcy with the
Reserve Primary Fund holding about $785 million in debt from it [3], making the prominent safe
short-term investment funds in the money market risky to invest. At that point, consumer confidence
in the money market had crashed, followed by the interbank lending market drying up because the
banks were highly unsure of the solvency of each other (incomplete information). At the same time,
the housing bubble burst due to the significant mortgage defaults and fall in mortgage-backed
securities that drove the housing price down. Moreover, by 2009, this recession had spread worldwide,
affecting economies such as Australia, Canada, the UK, Hong Kong, India, and China [4-11].
To define, a basic network consists of nodes and the links between them [12]. Albeit there is a
broad ackrussenowledgment of rich existing network models and theories in the field of this network
[13-16], its complexity is worth further investigation to assist in the explanation of underlying
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231049
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
26
mechanisms among institutions. This paper will review the literature on how financial network
structures could help explain the 2008 financial crisis and then discuss its impact on the global
economy, focusing on European and Asian markets.
2. Network Structures
2.1. Multilayer Interbank Credit Contract Network
Between bank contract structure explains the interbank connection in terms of internal assets, external
assets, and liability on balance sheets [17]. For example, Figure 1 illustrates the contract network
between three banks {1, 2, 3}. The red arrow represents loans from one bank to another, and the blue
one represents investments from bank to external. In this case, if Bank 3 has a loan from Bank 1, then
part of Bank 1’s asset {a1B} became Bank 3’s Liabilities {l3B}. The importance of this multilayer
model is the intertwined relationship between each institution, which will be further explained in the
next part.
Figure 1: Example of a contract network among three banks [17].
2.2. 2-D Lattice Model
Agata Aleksiejuk and Holyst introduced a 2-D lattice simple representation of the contagion process
and large-scale bankruptcies among interbank credit networks [13]. The network consists of vertical
and horizontal links between each node; each link represents a direct connection between two banks,
and each node represents a bank (see Figure 2.) Assuming that one bank is experiencing bankruptcy
caused by withdrawal or bad credit, the other two randomly chosen neighboring banks will face
insolvency if they have given a loan to that bank. For example, if bank {7} is bankrupted, then bank
{2, 6} would face insolvency, followed by bank {1, 3, 11}, and so on until all vertexes (banks) are
infected (see Figure 2.)
Figure 2: Bankruptcy spreading in banking network based on square lattice [13].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231049
27
Moreover, it noticed that vertex {7} and vertex {25}, for example, are indirectly connected, but
{25} was still affected by the insolvency of {7}. Recalling the interbank contract structure, the
indirect interaction between the financial institution, regarding the complexity of the network and
irresistible feedback mechanism, could lead to a greater effect in the financial market compared to
the initial shock for one bank [18]. Moreover, empirical research found the contract network to be
sparse, heavily tailed degree distributed, high clustering, and short average path length, suggesting a
surge in the speed of contagion in reality [19-21]. Thus, the banks are less likely to mitigate the
systematic risk they face, especially when facing a financial crisis, which could lead to a high social
cost [17].
2.3. Three Layers Contagion Channel
Bookstaber and Kenett provided a multiplex model to demonstrate how the collapse of the two
subprime mortgage-backed hedge funds of Bear Stern in 2007 impacted the financial market [22].
The three dimensions of the multilayer network consist of funding, collateral, and asset layers, as
shown in Figure 3. Bear Stern represents the Trading Desk Dealer labeled {T} in lilac, which it plays
in more than one financial activity, and its Hedge funds are presented as {HF} in yellow. On the
funding level, Bear Stern invested a significant amount in commercial paper and repurchase
agreement (repo) provided by Cash Provider {F}, which were both short-term fundings; plus, the two
hedge funds held as high as $18 billion assets from prime brokers {PB}. On the collateral level, the
two hedge funds received a $16 billion repo loan after the collateral to the prime brokers. On the asset
level, a remarkable amount of essential assets were traded through the desk of Bear Sterns (lilac
shaded in Figure 4.) Therefore, the collapse caused a wide disruption in the financial market within
all three layers affecting other institutions.
Figure 3: Three-dimensional Multilayer Network [22].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231049
28
Figure 4: Financial Network Map Showing Relationships among Market Participants [22].
2.4. Topology: the Core-Periphery Model
Characterizing the underlying topology of interbank networks is one of the mainstreams in the
financial network investigation. The majority of the research suggested a core-periphery structure,
which consists of the core, containing highly connected banks, and the periphery, including the other
banks that are connected to the core but not internally [23-26]. The core banks were further identified
as money-centered banks [12] and intermediary banks [24]. Moreover, Kojaku et al. suggested a
transition from a pure core-periphery model to a bipartite structure, and they discovered that the eMID
financial market had multiple pairs of core-periphery structures [26].
One key feature of this structure is a relatively higher clustering coefficient in the core compared
to the periphery layer (see Figure 5.) Bear Stern was classified as one of the core banks during the
2008 financial crisis, so its contagion could be widespread. Adding to it, an active decrease of links
from core banks to the other banks reduced the solvency of banks to face bankruptcy, aggregating to
the severe situation [24]. Moreover, the systematic risk level in this type of structure was found to be
high [25].
Figure 5: Core of Fedwire Interbank Payment Network [12].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231049
29
3. Global Impact
3.1. European Market
Empirical research on the core-periphery structure in the European market found a clear downward
trend of active banks in 2008 [24, 27]. This is because foreign bank interactions with the Italian banks,
for instance, had significantly dropped during the crisis.
Furthermore, McGuire and Von Peter suggested that solely evaluating the balance sheet data for
the European financial market during the crisis could be misleading [28]. Figure 6 shows that the
European banks funded subprime portfolios using a “bidirectional” method. As the cash inflow and
outflow from Europe to the US and vice versa were in a similar amount, the net cash flow would be
almost negligible. In this case, the emerging market investors were not significantly impacted by the
subprime mortgage securities [28]. However, European countries were highly dependent on US
dollars in trading, so they were directly impacted during the crisis. Other European countries, such as
Spanish and Greece, had experienced severe recessions from the contagion.
Figure 6: US dollar-denominated cross-border claims (billions of US dollars) [28].
3.2. Latin America
Empirical evidence suggested a noticeable change within the Latin American financial networks
before and after the crisis [29]. The pair-wise conditional correlations between the US and the Latin
American stock markets (Argentina, Brazil, Chile, and Mexico) helped explain this alternation and
its impact [29]. In addition, studies have shown, in contrast with the other countries, that the financial
crisis led to several significant positive effects on the Latin American stock market. This includes
reducing the volatility of the stock market and consumers gaining higher expected utility during their
investments [30].
3.3. Asian Market
The tree network diagram in Figure 7 helps explain the contagion spread from the US stock market
to the Asian stock market [31]. It demonstrated a chain reaction. Hong Kong was first affected by the
collapse in the US because of its highly active and liberal stock market. Followed by a rapid fall in
the Shanghai Stock Market index, then to Korea and other smaller markets in countries such as the
Philippines, Japan, and India.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231049
30
Figure 7: Spread of crisis from the US stock market to the Network of ASMs [31].
Moreover, Aswani suggested the density of the Asian financial market connection increased from
the pre-crisis period to the post-crisis (see Figure 8, Figure 9, and Figure 10) [31]. This proves,
contrary to the European market, the connectivity between Asian financial institutions had increased
pre- to post-crisis.
Figure 8: Network of Asian Stock Market Pre-crisis Period [31].
Figure 9: Network of Asian Stock Market in Crisis Period [31].
Figure 10: Network of Asian Stock Market post-crisis [31].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231049
31
4. Discussion
For further discussions, the financial network could be useful for financial regulators and the
government to evaluate, predict, and prevent future crises. Moreover, they could adjust their
international portfolio diversification based on their network structures [29]. For example, for a low
level of portfolio diversification, banks could be more resilient to systematic risk if the diversification
is low; while for a high level of portfolio diversification, banks could be more resilient to systematic
risk if the diversification is high [32]. The subtle level to determine whether it was a higher or lower
level may be difficult.
However, this suggests that it is essential for the government of related policymakers to implement
policies according to different financial networks in that area or otherwise lead to a greater risk of
failure. This research is similar to the idea of adjusting the clustering coefficient for the interbank
credit contract network; for example, it would be suitable to decrease the clustering coefficient so that
the contagion would spread slower, and financial institutions may have time to act.
Moreover, research has suggested network fragility had decreased post-crisis, which means it
became more resilient [33]. However, this signals to the financial regulators that the network is
dynamic over time, so paying attention to the network evolution is also vital.
Alternatively, the government could utilize other methods to mitigate financial crisis damage to
the economy. For example, the US government applied significant quantitative easing (QE) and
provided short-term loans to banks through the Federal Reserve Discount Window Program (FRDW).
Furthermore, Turnbull & Pirson showed the possibility of avoiding the 2008 financial crisis if
network governance is presented hypothesis, which emphasizes the high potential application of
financial network structures [34].
5. Conclusion
In conclusion, this paper discussed how the network between financial institutions helped explain the
2008 financial crisis by reviewing the literature on the 2-D lattice model, three layers contagion
channel, core-periphery model; and In the global impact section, this paper discussed the 2008
financial impacts on mainly European and Asian financial markets regarding networks. Overall, this
essay provides a general analysis of the financial networks, but the research gap in investigating the
underlying mechanism of financial decisions is still worth further discussion.
References
[1] Russell, K., & Zhang, C. (2023, March 11). 3 Failed Banks This Year Were Bigger Than 25 That Crumbled in 2008.
The New York Times. https://www.nytimes.com/interactive/2023/business/bank-failures-svb-first-republic-
signature.html
[2] CFI Team. (2023, March 16). Lehman Brothers. Corporate Finance Institute.
https://corporatefinanceinstitute.com/resources/capital-markets/lehman-brothers/
[3] Mcwhinney, J. (2019). Money Market Mayhem: The Reserve Fund Meltdown. Investopedia.
https://www.investopedia.com/articles/economics/09/money-market-reserve-fund-meltdown.asp
[4] Richardson, G., Taylor, G., & Lanis, R. (2015). The impact of financial distress on corporate tax avoidance
spanning the global financial crisis: Evidence from Australia. Economic Modelling, 44(ISSN 0264-999), 44–53.
https://doi.org/10.1016/j.econmod.2014.09.015
[5] Luchtenberg, K. F., & Vu, Q. V. (2015). The 2008 financial crisis: Stock market contagion and its determinants.
Research in International Business and Finance, 33, 178–203. https://doi.org/10.1016/j.ribaf.2014.09.007
[6] Ritchie, J. R. B., Amaya Molinar, C. M., & Frechtling, D. C. (2009). Impacts of the World Recession and Economic
Crisis on Tourism: North America. Journal of Travel Research, 49(1), 5–15.
https://doi.org/10.1177/0047287509353193
[7] Iqbal, A., & Kume, O. (2014). Impact of Financial Crisis on Firms’ Capital Structure in UK, France, and Germany.
Multinational Finance Journal, 18(3/4), 249–280. https://doi.org/10.17578/18-3/4-3
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231049
32
[8] Lee, S., Guo, W., Tsang, A., Mak, A. D. P., Wu, J., Ng, K. L., & Kwok, K. (2010). Evidence for the 2008 economic
crisis exacerbating depression in Hong Kong. Journal of Affective Disorders, 126(1-2), 125–133.
https://doi.org/10.1016/j.jad.2010.03.007
[9] Joseph, D. A. (2013). Global Financial Crisis: Reflections on Its Impact on India. SSRN Electronic Journal.
https://doi.org/10.2139/ssrn.2260772
[10] Wim Naudé. (2009). The financial crisis of 2008 and the developing countries. WIDER Discussion Paper, No.
2009/01(ISBN 978-92-9230-171-2). The United Nations University World Institute for Development Economics
Research (UNU-WIDER), Helsinki.
[11] Siddiqui, K. (2009). Financial Crisis and Its Impact on the Economies of China and India. Research in Applied
Economics, 1(1). https://doi.org/10.5296/rae.v1i1.183
[12] Soramäki, K., Bech, M. L., Arnold, J., Glass, R. J., & Beyeler, W. E. (2007). The topology of interbank payment
flows. Physica A: Statistical Mechanics and Its Applications, 379(1), 317–333.
https://doi.org/10.1016/j.physa.2006.11.093
[13] Agata Aleksiejuk, & Hołyst, J. A. (2001). A simple model of bank bankruptcies. Physica D: Nonlinear Phenomena,
299(1-2), 198–204. https://doi.org/10.1016/s0378-4371(01)00296-5
[14] Allen, F., & Gale, D. (2000). Financial Contagion. Journal of Political Economy, 108(1), 1–33.
https://doi.org/10.1086/262109
[15] Iori, G., & Jafarey, S. (2001). Criticality in a model of banking crises. Physica A: Statistical Mechanics and Its
Applications, 299(1-2), 205–212. https://doi.org/10.1016/s0378-4371(01)00297-7
[16] Li, S., He, J., & Zhuang, Y. (2010). A network model of the interbank market. Physica A: Statistical Mechanics and
Its Applications, 389(24), 5587–5593. https://doi.org/10.1016/j.physa.2010.08.057
[17] Battiston, S., Caldarelli, G., May, R., Roukny, T., & Stiglitz, J. E. (2015). The Price of Complexity in Financial
Networks. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2594028
[18] Stiglitz, J. E. (2010). Risk and Global Economic Architecture: Why Full Financial Integration May Be Undesirable.
American Economic Review, 100(2), 388–392. https://doi.org/10.1257/aer.100.2.388
[19] Demiralp, S., Preslopsky, B., & Whitesell, W. (2006). Overnight interbank loan markets. Journal of Economics and
Business, 58(1), 67–83. https://doi.org/10.1016/j.jeconbus.2005.04.003
[20] Bech, M. L., & Atalay, E. (2010). The topology of the federal funds market. Physica A: Statistical Mechanics and
Its Applications, 389(22), 5223–5246. https://doi.org/10.1016/j.physa.2010.05.058
[21] Boss, M., Elsinger, H., Summer, M., & Thurner, S. (2004). Network topology of the interbank market. Quantitative
Finance, 4(6), 677–684. https://doi.org/10.1080/14697680400020325
[22] Bookstaber, R., & Kenett, D. Y. (2016). Looking Deeper, Seeing More: A Multilayer Map of the Financial System.
Office of Financial Research.
[23] In ‘t Veld, D., & van Lelyveld, I. (2014). Finding the core: Network structure in interbank markets. Journal of
Banking & Finance, 49, 27–40. https://doi.org/10.1016/j.jbankfin.2014.08.006
[24] Fricke, D., & Lux, T. (2014). Core–Periphery Structure in the Overnight Money Market: Evidence from the e-MID
Trading Platform. Computational Economics, 45(3), 359–395. https://doi.org/10.1007/s10614-014-9427-x
[25] Silva, T. C., Guerra, S. M., Tabak, B. M., & de Castro Miranda, R. C. (2016). Financial networks, bank efficiency
and risk-taking. Journal of Financial Stability, 25, 247–257. https://doi.org/10.1016/j.jfs.2016.04.004
[26] Kojaku, S., Cimini, G., Caldarelli, G., & Masuda, N. (2018). Structural changes in the interbank market across the
financial crisis from multiple core–periphery analysis. The Journal of Network Theory in Finance, 4(3), 33–51.
https://doi.org/10.21314/jntf.2018.044
[27] Botta, A., & Tippet, B. (2021). Secular stagnation and core-periphery uneven development in post-crisis eurozone.
Competition & Change, 102452942110038. https://doi.org/10.1177/10245294211003864
[28] McCauley, R., Shin, H. S., & Avdjiev, S. (2016, September 28). Conceptual challenges in international finance.
CEPR. https://cepr.org/voxeu/columns/conceptual-challenges-international-finance
[29] Hwang, J.-K. (2014). Spillover Effects of the 2008 Financial Crisis in Latin America Stock Markets. International
Advances in Economic Research, 20(3), 311–324. https://doi.org/10.1007/s11294-014-9472-1
[30] Zhu, Z., Bai, Z., Vieito, J. P., & Wong, W.-K. (2018). The Impact of the Global Financial Crisis on the Efficiency
and Performance of Latin American Stock Markets. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3208090
[31] Aswani, J. (2017). Impact of global financial crisis on network of Asian stock markets. Algorithmic Finance, 6(3-
4), 79–91. https://doi.org/10.3233/af-170192
[32] Li, S., & Wang, C. (2021). Network structure, portfolio diversification and systemic risk. Journal of Management
Science and Engineering, 6(2), 235–245. https://doi.org/10.1016/j.jmse.2021.06.006
[33] Silva, T. C., de Souza, S. R. S., & Tabak, B. M. (2016). Structure and dynamics of the global financial network.
Chaos, Solitons & Fractals, 88, 218–234. https://doi.org/10.1016/j.chaos.2016.01.023
[34] Turnbull, S., & Pirson, M. (2011). Could the 2008 US financial crisis have been avoided with network governance?
International Journal of Disclosure and Governance, 9(3), 238–263. https://doi.org/10.1057/jdg.2011.26
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231049
33
The Impact of the Interest Rates Raised by Federal Reserve
System on the Exchange Rate of the US Dollar Against the
Chinese Yuan
Yuenan Chen1, a, *
1Management school, University of Liverpool, Liverpool, L69 3BX, The United Kingdom
a. hsych132@liverpool.ac.uk
*corresponding author
Abstract: On March 16, 2022, the U.S. raised interest rates for the first time in response to
inflationary pressures, which announced a 25-basis point increase. After the latter four rate
hikes happened in May, July, September and November, the cumulative increased total is 425
basis points. This change had a huge impact, especially on the exchange rate between the US
dollar and the Chinese currency. This paper uses data on the exchange rate of the US dollar
to the Chinese yuan after 2010, and ARIMA model is built to model and analyze the data in
order to study the impact of the Federal Reserve’s interest rate hike on the exchange rate of
dollar and yuan. Therefore, the analysis discovered that the Fed’s interest rate increase caused
the appreciation of the USD and the depreciation of the CNY. Furthermore, the model helps
to predict the trend of the future exchange rate, which in turn will provide suggestions for
China’s future development.
Keywords: interest rate, exchange rate, US, China, ARIMA model
1. Introduction
The Federal Reserve is the central bank of the United States of America, and its primary responsibility
is to maintain maximum employment and price stability by regulating monetary policy. By regulating
the federal funds rate, it influences market interest rates, which in turn affects the money supply and
economic activity. With the outbreak and continuation of the COVID-19, its impact on individual
countries has been multifaceted and severe. The pandemic has led to supply chain disruptions,
demand recovery and fiscal stimulus of the United States, which contribute to the increase of the
consumer index of America, reaching its highest level in four decades. In addition, the inflation rate
far exceeding expectations and resulted in severe damage of the economy. Therefore, the Federal
Reserve should take some action. For example, raising interest rate to tighten monetary policy to
dampen inflationary expectations and actual inflation.
In response to economic pressures in the United States, the Federal Reserve has raised interest
rates, which has also pushed up the exchange rate of the dollar. For the yuan, the US dollar appreciated
while the yuan depreciated. If the yuan continues to depreciate, it will also bring some impacts and
challenges to China. For example, the burden of servicing foreign debts will increase; exports will
fall; imports will increase, and financial markets will become unstable. What should be the future
trend of the exchange of the United States dollar for the RMB? What will be the impact on China’s
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231054
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
34
economic development if the RMB continues to depreciate? What strategies and recommendations
should the Authority of China people make to cope with it? It is very meaningful to discuss these
issues.
The thesis is based on information on the USD/CNY exchange rate from 2010 to just before the
first Fed rate hike. These data were broken down into three types of time series data that were
collected on a daily, weekly, and monthly basis. Each category of data was then utilized to create an
artificial control group that was not impacted by the Fed’s interest rate hike. As a result, the impact
and outcomes of the Fed rate hike can be examined after comparison with the real exchange rate data
set. This study is based on the exchange rate data of the US dollar to the Chinese yuan after 2010,
which is divided into daily, weekly and monthly data, and builds an ARIMA model to forecast the
future trend of the exchange rate after the Fed’s interest rate hike in the form of a time series.
2. Literature Review
Since the COVID-19 happened in 2020, it has had a profound effect on the world. For the United
States, many people lost their jobs because of the outbreak, and some were unable to find work
immediately after it ended, causing unemployment rate to soar. Mahmoudi claims the unemployment
rate increased from 3.5% in February 2020 to the highest level seen since the 1930s in April 2020,
and it peaked at 4.9% in October 2021. For comparison, the employment rate in early 2022 was about
3 million lower than it had been before to the COVID-19 epidemic [1]. The CPI index of the United
State continued to rise, reaching a high of 9.1% in June 2022, a figure that was a 40-year high. This
is well above the 2% inflation target set by the Federal Reserve and too much inflation can cause
serious damage to the economy [2]. Additionally, the Fed initially raised the interest rate on 16 March
by 25 basis points and expected the inflation to stabilize by several months later. Then, In Fed
Chairman Jerome Powell press remarks, who stated that the rate-setting committee’s main attention
in the upcoming months would be on inflation data [3]. Besides, Siegel points out the central bank
continued the faster pace of tightening that was started in June on July 27th by increasing its policy
rate by an additional 75 basis points. However, The Fed will continue to tighten monetary policy at a
faster rate, though, if inflation is higher than presently anticipate [4]. Moreover, Ming suggests the
CNY exchange rate movement against the USD has exhibited the characteristics of “consolidation-
rapid depreciation-consolidation-rapid depreciation” from the beginning of 2022 until the present.
The median price of CNY against the USD decreased from 6.38 to 7.07 between the end of 2021 and
September 27, 2022, a decline of 10.9% and the exchange rate for the remaining months of 2022 is
assumed to swing between 6.9 and 7.3 [5]. However, Tintin proposes the steep fall of the CNY need
not cause the market too much concern. The market is not overly concerned about a sudden drop in
the value of the CNY [6]. The Chinese central bank currently has enough policy options to address
the CNY exchange rate’s fast decline. On the one hand, the central bank has been stabilizing the
currency market by utilizing methods for managing the foreign exchange market. On the other hand,
China has undertaken counter-cyclical factors, capital flow control, and macro-prudential
management at the level of foreign exchange [7]. Nevertheless, the Fed is having trouble controlling
inflation as it becomes clear that changing monetary policy won’t solve all of the economy’s problems.
It is not advisable to utilize the cure-all for economic issues repeatedly or excessively [8]. Furthermore,
the depreciation of the RMB is not all bad. According to the research of Zhangpeng, it promotes the
optimization of the structure of export products [9]. Fluctuations in the exchange rate between the
renminbi and the US dollar have led to trade frictions between the US and China, and challenges in
the US-China relationship [10].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231054
35
3. Research Design
3.1. Data Source
This paper collects the data of the exchange rate between USD and CNY from the Investing.com,
which is the world’s fourth largest financial information website. A sample of 13 years of data from
2010 was gathered, and the data acquisition was divided into three forms, including daily, weekly,
and monthly data. In addition, the exchange rate after announcing the initial increase of the interest
rate by Fed was highlighted. These data provide the basis and source for practical analyses of the
impact of the Fed’s rate hike on the exchange rate between the RMB and the USD.
3.2. Weak Stationarity Test
The ADF test (unit root test), which determines if the series has a unit root and is smooth or not, is
the first stage in implementing the model. After respectively importing the daily, weekly and monthly
data into Stata and conducting the ADF test, Table 1 shows that the p-value of three types for the log-
prices are zero after diverse order differencing, contradicting the original hypothesis that the model
is not smooth. This shows that the model is stable and viable.
As is shown in the Table 1, the difference order (d) of daily test is 1. Besides, weekly and monthly
test both have the same difference order (d), which d equals 2.
Table 1: Weak stationarity test.
t
p
Daily
Ln price
-2.064
0.5664
1st order difference
-38.878
0.0000
Weekly
Ln price
-4.880
0.0003
1st order difference
-20.561
0.0000
2nd order difference
-33.023
0.0000
Monthly
Ln price
-2.249
0.4625
1st order difference
-7.108
0.0000
2nd order difference
-13.182
0.0000
3.3. ARIMA Model
The AR(p) model ∑∅!𝑅"#!
$
!%1, which forecasts the future by using the history exchange rate between
the U.S. dollar and the Chinese yuan, is represented by equation (1) above, while equation (1) also
shows that 𝛼!−∑∅!𝛼"#!
&
!%1, which employs previous volatility to forecast the exchange rate’s future
trend.
𝑅"= ∅0+∑∅!𝑅"#! + 𝑎"
$
!%1 −∑𝜃!𝑎"#!
&
!%1 (1)
In this study, the AR model uses historical data on the exchange rate prior to the Fed’s first interest
rate increase on March 16 2022 whereas the MA model forecasts the future using an error term.
The ARIMA model is used for forecasting. It is impossible to build an ARMA (p, q) model when
the original sequence looks to be unstable, hence differencing should be used instead. The sequence
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231054
36
becomes smooth when the difference order is d, at which point the ARMA model can be represented,
and at this moment the model is the ARIMA (p, d, q) model.
For daily data, the original series is not smooth. After first order differencing, the unit root test
rejects the original hypothesis and then ARMA model can be performed (i.e., ARIMA model). In
addition, both weekly and monthly data were differenced twice to ensure that the series were stable
so that the ARIMA model could be applied.
This paper sets 16 March, the date of the Fed’s first-rate hike in 2022, as moment zero (T0). This
allows us to predict the movement of the exchange rate after a rate hike based on past data.
4. Empirical Findings and Analysis
4.1. Determine the Order
The first step in this section is to use the PACF and ACF pairings to order the first ln_price series.
The results are displayed below.
When it comes to the daily data, The part beyond the x-axis is 9 according to the fixed order result
of the two images in the first row in Figure 1, which means AR(p) is of order 9 and MA(q) is also of
order 9. As a result, p and q have values of 9.
Regarding weekly information, Figure 1 shows AR(p) is of order 4 and MA(q) is of order 1, which
means p equals 4 and q equals 1.
As for monthly measured data, the order of AR(p) and MA(p) are 9 and 1 respectively. Therefore,
the value of p and q are 9 and 1.
PACF
ACF
Daily
Weekly
Monthly
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231054
37
Figure 1: ARMA (p, q) identification.
Photo credit: Original
4.2. Forecasting Results and Interpretation
After establishing the order, a residual test was conducted to determine whether the data were white
noise. As shown in Table 2, both the daily and weekly data are white noise data, demonstrating the
viability and sufficiency of the data.
Table 2: Residual test.
Model
Portmanteau (Q) statistic
Prob > chi2
Daily-ARIMA (9,1,9)
22.8397
0.9866
Weekly-ARIMA (4,2,1)
70.4057
0.0021
Monthly-ARIMA (9,2,1)
27.3978
0.9351
For the forecast of the daily exchange rate, the model evolves the likely trend of the exchange rate
after the rate hike from 16 March to 27 March 2022 based on historical data. As can be seen from the
Figure 2, the forecast of the exchange rate remains relatively flat at 6.37 after the Fed rate hike.
However, after 16 March the actual exchange rate data experienced a fluctuation from about 6.34 to
6.37. The actual value experienced a peak at about 6.37 on the 23 March and then declined. However,
the overall actual exchange rate has been consistently lower than the forecast rate. In addition, the
fact that there are fewer days in the forecast and a time lag before the dollar appreciates as a result of
the interest rate hike may be the cause of the actual exchange rate being lower than the predicted
exchange rate.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231054
38
Figure 2: Actual value and fitted value, daily.
Photo credit: Original
Figure 3 prognosticates the USD-CNY exchange rate for the subsequent four weeks following an
increase in interest rates by the Fed. The forecasts show a continued appreciation of the dollar, rising
from around 6.35 and surpassing the actual exchange rate figure around 25 March, reaching around
6.38 on 10 April. The real exchange rate, however, remained relatively flat around 6.37, which was
very close to the predicted value.
Figure 3: Actual value and fitted value, weekly.
Photo credit: Original
As can be seen in Figure 4, based on historical exchange rates, the exchange rate should stay
essentially flat at approximately 6.4 for the following nine months after March if the Fed does not
raise rates. In actuality, after the Fed’s rate hike on 16 March, the US dollar continued to appreciate
and the Chinese yuan continued to depreciate. Additionally, the dollar strengthened rapidly from
around 6.4 to around 7.3 in October. Then it experienced a decline to roughly 6.9.
6.31
6.32
6.33
6.34
6.35
6.36
6.37
6.38
2022-03-03
2022-03-05
2022-03-07
2022-03-09
2022-03-11
2022-03-13
2022-03-15
2022-03-17
2022-03-19
2022-03-21
2022-03-23
2022-03-25
2022-03-27
Actual value Fitted value
6.31
6.32
6.33
6.34
6.35
6.36
6.37
6.38
6.39
2022-02-20
2022-02-27
2022-03-06
2022-03-13
2022-03-20
2022-03-27
2022-04-03
2022-04-10
Actual value Fitted value
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231054
39
Figure 4: Actual value and fitted value, monthly.
Photo credit: Original
5. Conclusion
The purpose of this thesis is to investigate how a Fed interest rate hike would affect the exchange rate
of the US dollar to the Chinese yuan. After the rate hike in March, a forecasting model based on
monthly data illustrates that the US dollar will continue to strengthen and appreciate in the remaining
months of 2022, implying a similar depreciation of the Chinese yuan, which is in line with previous
research.
The forecasting involves several economic concepts and implications. On the one hand, due to the
potential for larger returns on assets denominated in that nation’s currency, when a country’s central
bank, like the Federal Reserve in the United States, boosts its interest rates, it frequently attracts
higher levels of international investment. As a result, there is a rise in the demand for that currency,
which raises its value in relation to other currencies. In this situation, the US currency gains strength.
By contrast, the Chinese Yuan has depreciated. As for trade implications, a stronger US dollar may
increase the cost of US exports for international consumers, thereby dampening demand for American
goods and services. Conversely, a weaker Chinese Yuan may make Chinese exports more affordable
on the international market, potentially increasing Chinese exports.
To avoid excessive currency fluctuations that could cause instability, nations may need to
coordinate their policies. Firstly, policymakers should keep an eye on capital flows and any possibility
for speculation in reaction to changes in currencies. In addition, to ensure overall economic stability,
policymakers might need to make changes to their monetary and fiscal strategies. In the case of China,
the authorities may need to manage the foreign exchange reserves and align their short-term responses
with long-term economic goals.
Investors should also carefully assess the situation. Diversification helps mitigate the impact of
currency fluctuations on the overall portfolio, so Investors ought to maintain a well-diversified
portfolio comprises a variety of asset classes (stocks, bonds, real estate, etc.) and geographies.
Moreover, Investors that have exposure to either the US dollar or the CNY on a considerable scale
should evaluate the potential effects on their holdings. A portfolio may need to be rebalanced to take
into account shifting market dynamics if it is highly influenced by certain currencies. Besides,
markets for currencies can be erratic. Investors should continuously monitor developments and be
ready to modify their strategies if the situation alter.
6.2
6.4
6.6
6.8
7
7.2
7.4
2021-07-01
2021-08-01
2021-09-01
2021-10-01
2021-11-01
2021-12-01
2022-01-01
2022-02-01
2022-03-01
2022-04-01
2022-05-01
2022-06-01
2022-07-01
2022-08-01
2022-09-01
2022-10-01
2022-11-01
2022-12-01
Actual value Fitted value
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231054
40
It is confirmed by the ARIMA model that the Fed’s interest rate hike led to the appreciation of the
US dollar and the depreciation of the RMB. Besides, the results of this study are in line with previous
researches. For this situation, this thesis proposes relevant policy recommendations including
maintaining the stability of capital inflows and the trade, keeping an eye on macroeconomic stability
and foreign reserves management. Similarly, it also suggests that investors can make opportunistic
investments when the exchange rate changes, but they need to pay attention to the market fluctuations.
Furthermore, investors are recommended to diversify their risks and improve their overall
consciousness in order to analyze the global economy.
References
[1] Mahmoudi, M. (2023). COVID lessons: was there any way to reduce the negative effect of COVID-19 on the United
States economy? Journal of Economic Studies (Bradford), 50(5), 896–920. https://doi.org/10.1108/JES-01-2022-
0052
[2] ‘Understanding US inflation during the COVID era’ (2022) Premium Official News, 09 Sep, NA, available:
https://link-gale-com.liverpool.idm.oclc.org/apps/doc/A717044760/GBIB?u=livuni&sid=bookmark-
GBIB&xid=c4857d8d [accessed 23 Aug 2023].
[3] ‘Fed raises interest rates by another 75 basis points’ (2022) Country Report: United States, 29 Jul, NA, available:
https://link-gale-com.liverpool.idm.oclc.org/apps/doc/A712154405/GBIB?u=livuni&sid=bookmark-
GBIB&xid=6df9ab4a [accessed 23 Aug 2023].
[4] Siegel, R. (2021). Fed projects new interest rate hike in 2022, signals upcoming easing of supports if economy
continues to heal. Washington Post, NA. https://link-gale-
com.liverpool.idm.oclc.org/apps/doc/A676431460/AONE?u=livuni&sid=bookmark-AONE&xid=756927c3.
[5] M. Zhang, (2022). The depreciation of the RMB against the US Dollar: Manifestations, Causes, Responses and
prospects. Tsinghua Financisal Review.10:69-70.
[6] T. T. Zhu, (2022). Volatility Analysis and Trend Forecast of RMB Exchange Rate against US Dollar. Zhongnan
University of Economics and Law.
[7] M. Zhang, Y. M. Chen, (2022). No need to worry about a sharp depreciation of the RMB against the US dollar.
China’s currency market, 11:22-26.
[8] H. Z. Qi, Y. Y. Chen and Y. Y. Liu, (2023). Spillover effect and Countermeasures of the Federal Reserve’s interest
rate hike Cycle in 2022. Journal of Changchun Finance College, 01:5-14.
[9] Z. P. Zhang, (2022). Study on the Impact of Exchange Rate Fluctuation on China’s Import and Export Trade. North
China University of Technology.
[10] N. S. Cooray, P. Thangavel, (2022). The impact and implication of the COVID-19 on the trade relationship between
China and the United States: the political economy perspectives. Transnational Corporations Review 14.1.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231054
41
Analysis of the Real Estate Market in China’s Past and
Present
Hao Gong1,a,*
1School of Business, University of Nottingham Ningbo China, Ningbo City, Zhejiang Province,
315199, China
a. biyhg4@nottingham.edu.cn
*corresponding author
Abstract: The real estate sector, which plays a significant role in land financing and is a
foundational sector of the Chinese economy, is very important. By primarily examining data
from the China Statistics Bureau, the China Statistical Yearbook, and real estate-related
statistics, this paper examines the present growth pattern of China’s real estate business and
highlights possible concerns in the current real estate industry. The real estate market is
clearly in a downturn right now, but there are also hidden risks of overinvestment that might
undermine the stability of the financial system. This is what this paper’s data investigation
reveals. As a result, the economic growth model that relies heavily on land financing cannot
continue, and appropriate regulatory changes are needed to lower the risks associated with
potential hidden hazards in real estate. Finally, it is recommended that the government
carefully regulate the real estate market and, in the future, promote high-tech companies and
financial industries to lessen the country’s reliance on land financing.
Keywords: real estate, land finance, Chinese economy
1. Introduction
China’s local economy is highly dependent on land finance, and most of the region’s economic
development is driven by the real estate sector. After the global financial crisis broke out in 2008,
China adopted expansionary policies to stimulate the economy. While this has contributed to
sustained economic growth, it has also resulted in a rapid increase in the leverage ratio of various
sectors and the macro leverage ratio. The high debt of real estate enterprises has also become more
and more prominent, which hides a large risk potential.
To prevent the debt risk of real estate enterprises from spreading to the entire financial system,
China introduced the “three red lines” and the “centralized management system for real estate
enterprise loans” in 2020, to strictly prevent funds from flowing into the real estate market in violation
of the law [1]. China’s supervision of real estate enterprises to become more stringent effectively
curbed the real estate enterprise leverage rate continued to climb, the real estate enterprise gearing
ratio in 2021 fell by 0.38 percentage points [1], but also caused Evergrande Group as the
representative of the part of the real estate enterprise cash flow difficulties, real estate enterprises have
appeared in the event of default on debt, the real estate market and the relevant financial institutions
have brought about a greater impact. The real estate market as well as the related financial institutions
have been greatly impacted. In the past two years, China’s real estate prices, in general, appeared
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231056
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
42
downward in the background, real estate enterprises’ high leverage to China’s economy is the hidden
danger that caused great concern.
This paper will examine the development history, current situation, and possible pitfalls of China’s
real estate in the past few years, and give some insights and suggestions.
The purpose of this paper is to study and analyze the sustainability of the government’s
implementation of an economic development approach that is highly dependent on land finance, and
to provide corresponding insights and recommendations.
This paper will first introduce the development history of China’s real estate industry, explaining
how the real estate industry has been an important part of China’s economic development and the
relationship between the real estate industry and China’s economy. Secondly, it will introduce the
current situation of real estate and its hidden dangers. Finally, some suggestions will be given and the
article will be summarized.
2. History of China’s Real Estate Industry
2.1. Economics of Local Governments’ Development Model Centered on Land Finance
After the reform of the tax-sharing system in 1994, part of the financial power of local governments
was transferred to the central government, while the power of affairs remained almost unchanged,
which led to the excessive responsibility of local governments and a huge gap between fiscal revenues
and expenditures. In order to maintain the normal operation of local governments and the continuous
development of the local economy, it is not enough to rely only on the local government’s own income
and transfer payments from the central government, so local governments at all levels actively look
for financial resources. At this time, the land as a resource in the hands of the local government could
be freely disposed of, and its huge economic value gradually appeared, local governments at all levels
through “land sales” to obtain a huge amount of income, this part of the huge income can solve the
situation of the local financial resources shortage. Therefore, local governments will naturally
produce dependence on “land finance” [2].
Figure 1: Revenues from land finance and their share of fiscal revenues.
Data source: National Bureau of Statistics
Photo credit: Original
1 1 1 1 2 5 5 6 8 12 10 17 27 32 28
41 40 31 37
52
65
78 84 87
67
5% 5% 4%
8%
13%
25%
21%
19%
21%
24%
17%
25%
33%
31%
24%
32%
29%
20%
23%
30%
35%
41%
46%
43%
33%
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
0
10
20
30
40
50
60
70
80
90
100
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
Thousand
Total Land Premiums Land Finance Dependence
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231056
43
As can be seen in Figure 1, China’s dependence on land finance has been increasing since the
housing reform in 1998, and especially in 2015-2020, China’s dependence on land finance continues
to increase dramatically, reaching nearly half of the public budget revenue in 2020.
The reason why the government relies on land finance is that land finance can quickly pry local
development: firstly, the government through the allocation and agreement to transfer land and other
ways to attract investment to promote the manufacturing industry, real estate industry and the
construction industry extraordinary development, access to high sales tax, corporate income tax, and
other local taxes, and with the land concessions, the government will continue to increase the
periphery of the other project inputs, which will lead to the whole region urbanization and Economic
growth; Secondly, the government through the recruitment, auction, hanging and other ways to collect
land premiums, and the right to use the land and the right to obtain the land financing, so as to drive
the local economic development; Finally, the government can use the land for bank mortgages to
obtain the urban infrastructure funds, to promote the construction of urban infrastructure [3]. In this
way, land finance has become an indispensable part of local governments to build cities.
2.2. Relationship Between the Chinese Real Estate Market and Economy
Land finance and the real estate market are tightly linked through land acquisition fees, and higher
house prices can bring higher land finance revenues to the government, thus further boosting the local
urbanization process. Therefore, one of the consequences of relying heavily on land finance is to push
up local housing prices and to further deepen the link between the real estate sector and the Chinese
economy.
The direct contribution of the real estate industry to China’s GDP in 2020 is more than 12% (value
added share), of which 7.3% comes from real estate services and 4.8% from real estate construction
[4]. In addition, according to the 2017 China Input-Output Table, it is expected that the indirect
contribution of the real estate industry to GDP through inter-industry supply chain transmission may
also be above 12%. Overall, the real estate sector’s pull on China’s GDP growth could be around 25%,
i.e. if real estate activity falls by 10% overall, the combined direct and indirect drag on GDP growth
(including multiple rounds of transmission) could be around 2.5 percentage points. At the same time,
the economic situation will also affect the real estate industry in the direction of economic growth on
real estate investment is greater than the impact of real estate investment on economic growth, which
means that once the economic downturn fluctuations will lead to dramatic fluctuations in real estate
investment [5].
Overall, the real estate industry is inextricably linked to the Chinese economy.
3. Current Situation of Real Estate in China
3.1. Concerns About the Viability of the Real Estate-Centric Strategy of Economic
Development
Whether China’s real estate industry is still in a golden period should mainly depend on the main
indicators such as the scale of investment, the scale of sales, the number of employees, and the rate
of increase.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231056
44
Figure 2: Employment in China’s real estate industry.
Data source: China population and employment statistical yearbook
Photo credit: Original
Figure 2 shows that employment in the real estate sector has decreased over the past ten years,
with all growth rates being negative.
Figure 3: Real estate investment as a share of GDP.
Data source: National Bureau of Statistics
Photo credit: Original
As can be seen in Figure 3, the share of real estate investment in GDP peaked in 2014, and then
gradually became a downward trend.
476 467 371 365 331 321 263 221 215 196 157
-1.94%
-20.50%
-1.65%
-9.30%
-3.03%
-17.95%
-16.06%
-2.84%
-8.84%
-19.87%
-25%
-20%
-15%
-10%
-5%
0%
0
100
200
300
400
500
600
2011 2012 2013 2014 2015 2016 2017 2020 2021 2018 2019
Thousand
Working Population Growth Rate
4.99%
5.68%
6.41%
7.42%
8.16%
8.49%
8.84%
9.37%
9.77%
10.39%
11.72%
12.67%
13.33%
14.50%
14.76%
13.94%
13.75%
13.20%
13.08%
13.40%
13.95%
12.84%
10.98%
4.00%
6.00%
8.00%
10.00%
12.00%
14.00%
16.00%
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231056
45
Secondly, according to the National Bureau of Statistics, in 2022, China’s real estate industry, the
area of new housing construction decreased by 39.4% compared to 2021, the area of housing sales
decreased by 24.3% compared to 2021, and the market supply of new housing decreased, and the
listing of second-hand housing continued to improve [6].
Taken together, the number of real estate-related employees in China is also decreasing, while the
supply of real estate has shrunk, suggesting that real estate is shrinking in size.
In addition to this, changes in population size and age structure are long-term factors driving the
real estate market. Changes in the demographic structure make it difficult to support rigid demand in
the real estate market. This contains two factors, the first factor is the population growth rate and its
structural changes, a long-term factor driving the development of the real estate market [7]. The
second is the number of marriages, marriage, and home ownership has been one of the main
components of consumer demand for commercial real estate in China.
Figure 4: Population Structure.
Data source: National Bureau of Statistics
Photo credit: Original
According to Figure 4, China’s population is undergoing structural changes, the 65+ population is
increasing, while the population under 14 years old is decreasing. The population growth rate has
declined to less than 1.5% after 1990, and the current population growth rate is about 0.5%, it is
expected that the population growth rate will still decline in the future. Secondly, the number of
marriages has been growing negatively since 2015 [7]. Marriage and home ownership are one of the
main components of consumer demand for commercial properties. However, from the Civil Affairs
Bureau published data on new marriages in previous years, it is expected that the number of newly
married pairs in the country will decline from 13.5 million pairs in 2013 to 8 million pairs in 2020
year by year, with an average annual growth rate of -7%, which can be seen that the demand for
marriage and home ownership is declining year by year in the next five years [7]. The impact of
demographic factors on the real estate industry is complex and may lead to changes in market supply
and demand, fluctuations in price trends, and strategic adjustments by developers. The current
population decline means that the demand for property will plummet and house prices will fall,
leading to a further contraction in the size of real estate.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
Population aged 0 to 14 Population aged 15 to 64 Population aged 65 and +
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231056
46
3.2. Problems That may Arise From the Crash of Real Estate Companies Represented by
Evergrande
In 2021, Evergrande Group, the second-largest Chinese real estate developer in terms of
comprehensive strength, had debt repayment problems. The endgame of the Evergrande case is not
yet completely clear, but the spillover effects have begun to emerge. The real estate research team at
UBS Securities believes that Evergrande’s debt restructuring may be difficult to avoid in the coming
weeks or months [4]. Possible ways for the spillover effects of the Evergrande incident include: 1)
tightening financing conditions for other developers and suppliers; 2) impacting the project delivery
capacity of real estate-related suppliers; 3) dragging down confidence in the real estate market as well
as real estate sales and construction activities; and 4) dragging down the production and investment
in real estate-related industries, especially building materials [4]. The UBS Bank research team did
stress tests on the possible impact on bank loan quality under different scenarios, such as a default by
Evergrande, defaults by more high-risk developers, and a significant downturn in real estate activity,
and the main conclusion was that the overall risk exposure of bank loans is manageable. Of course,
the impact could be even greater if banks’ off-balance sheet exposures and indirect exposures from
real estate-related sectors and local government debt are taken into account. In the most pessimistic
scenario, the drag on economic growth will continue to be highlighted.
3.3. Pitfalls in the Real Estate Sector
The Chinese Academy of Social Sciences released the “2007 China’s Social Situation Analysis and
Forecast” Blue Book pointed out that the current “land finance” model of investment overheating risk
and financial risk [8].
The Shell Research Institute released the “2022 Survey Report on Housing Vacancy Rate in Major
Chinese Cities”, which shows that the average housing vacancy rate in 28 large and medium-sized
cities is 12% [9], which is relatively in a high range. This means that many houses are not fully utilized
and there is a waste of land resources, which is also a reflection of overheated investment. The
oversupply of housing may mean that there is a bubble in real estate assets, and the bursting of asset
bubbles may trigger financial risks and economic instability, which in turn affects the long-term
sustainable development of the economy.
Data released [10] by the China Index Research Institute shows that in 2022, a total of 606,000
foreclosed homes were listed for auction in China, an increase of 35.7% year-on-year, and the size of
the nation’s foreclosure market has increased to 1.4 trillion yuan. The surge in foreclosures means a
rise in the risk of loan defaults, meaning that some borrowers are unable to repay their loans on time.
This could lead to an increase in non-performing loans by banks and financial institutions, thus posing
a threat to the stability of the financial system.
Pitfalls such as over-investment and impacts on the stability of the financial system may affect the
long-term health of the economy, so the government should transition from a model that relies heavily
on land finance to develop the economy.
4. Conclusion
As the cornerstone sector of China’s economic development, real estate is under intense development
pressure as a result of the real estate industry’s abrupt move away from high-speed expansion in recent
years.
This essay first describes China’s approach to economic development, which is based on land
finance, before going on to discuss the role that real estate plays in China’s economic growth. It then
examines the current state of real estate development and the potential risks. The market is in a slump,
as seen by the drop in real estate employment and the percentage of real estate investment; however,
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231056
47
by examining the vacancy rate of Chinese properties and the number of foreclosed properties, it is
determined that there are potential risks associated with overinvesting in real estate. Finally, it offers
pertinent advice from the government’s perspective on how to keep China’s economy growing
sustainably and healthily.
The haphazard growth of the real estate sector in China should be tightly controlled, as should the
leverage and debt-to-equity ratio of real estate companies. Second, the government should actively
promote the reform of industrial institutions to further achieve tax growth, diversify fiscal revenues,
and energetically create high-tech companies focused on research and technology. To draw more
investment into the actual economy, it should also boost the financial market’s innovation and
openness. At the same time, the growth of the bond market and the capital market should lessen
reliance on bank loans and the real estate market.
References
[1] China Index Academy Limited. http://www.beijingleather.com.cn/house/105035.html
[2] China population and employment statistical yearbook.
https://kns.cnki.net/kns/Navi?DBCode=CYFD&BaseID=YZGRL
[3] Fong, W. (2011). Real estate investment, real estate credit and China’s economic growth. Economic Theory and
Economic Management (01), 59-68.
[4] Ge, Yang & Qian, Chen. (2014). The role of “land finance” in promoting economic growth and its transformation.
Social Science Research (01), 28-34.
[5] Lin, Tsai-Yi. (2016). China’s real estate market has entered a downward cycle. China Real Estate Finance (04),
94-95.
[6] National Bureau of Statistics homepage. http://www.stats.gov.cn/
[7] Ruxin, Lu Xueyi, Li Peilin, Xu Xinxin, Chen Guangjin, Li Wei. (2007): Analysis and Forecast of China’s Social
Situation. Beijing. Social Science Literature Publishing House,2006-12
[8] Shell Research Institute. (2022). Survey Report on Housing Vacancy Rate in Major Cities of China,
2022.https://www.djyanbao.com/report/detail?id=3273840&from=search_list
[9] Tian, Yuan Yuan. (2023). Research on the Impact of Leverage Ratio of Real Estate Enterprises on Systemic
Financial Risk (Master’s thesis, Lanzhou University of Finance and Economics).
[10] Wang, T., Zhang, N. (2021).: How important is China’s real estate sector to China’s economy from a macro
perspective. Real Estate Journal (11), 16-19.
[11] Yu, N. (2012). Correlation analysis of urbanization development and land finance in China (Master’s thesis, Tianjin
University of Finance and Economics).
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231056
48
Research on the Investment Portfolio Optimization Based on
Efficient Frontier Model: A Portfolio of AMD, NVIDIA,
TXN, LRCX, AVGO, QCOM, INTC and MRVL
Ye Chen1, a, *
1Department of Economic, University College London, London, United Kingdom
a. yjmsyc5@ucl.ac.uk,
*corresponding author
Abstract: Optimal investment portfolio is beneficial for investors to make decisions and
construct investment strategy. This paper uses effective frontier methods to build an
investment portfolio. This study selects 8 different companies from the USA, which are AMD,
MRVL, LRCX, QCOM, INTC, AVGO, and TXN, and collects the data to optimize
investment portfolio. The study finds that the optimal portfolio point is (0.1407, 0.0506), the
risk is 0.1407, and the return is 0.0506. In this case, for AMD, MRVL, NVIDA, LRCX, INTC,
QCOM, AVGO, and TXN, the weights for each company are 0.0502, -0.3867, 0.5289, 0.4367,
-0.9987, -0.2805, 1, and 0.6501, respectively. The study allows for short selling and buy
mechanisms, since a negative number is actually equivalent to a short sale. As a result, the
study has an optimal portfolio that enables investors to make optimal investment decisions
for the company the study choose. Through this research, it can help people to increase their
life quality, because investment can help with more finance in the daily life, which can let
people consume a better goods.
Keywords: portfolio investment, efficient frontier, NVIDIA
1. Introduction
Investment has many benefits for our lives, enabling us to achieve our life goals, such as accumulating
enough retirement pension to enjoy the old age, establishing an education fund to consider our
children’s future, accumulating a certain amount of funds to buy a car or house, or accumulating a
sum of funds to travel around the world, and some people even plan to establish their own businesses,
etc. The achievement of these goals and investment are conducive to providing a better economic
foundation. Moreover, people often encounter unexpected situations in their daily lives, such as
illness, injury, death of loved ones, natural disasters, theft, unemployment, etc., which can lead to a
decrease in personal wealth. In order to offset these unexpected and disasters, it is necessary to carry
out scientific investment planning, reasonably arrange income and expenditure, in order to have
sufficient financial support and successfully overcome difficulties in the event of unexpected and
disasters; In the absence of unforeseen circumstances or disasters, it is possible to establish a “risk
fund” and increase its value, thereby improving the quality of life. The importance of the investment
including, firstly, realizing the preservation and appreciation of wealth (resistance to inflation).
Secondly, improve family risk resistance. Thirdly, improve the life quality (improve financial
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231057
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
49
conditions and meet different levels of consumer demand). Finally, investment is closely related to
national economic growth (the basic driving force of economic growth. An increase in investment
will cause a K-fold increase in income).
In addition, there are some other researchers doing the similar research, De Bondt and Thaler used
empirical evidence from CRSP monthly earnings data and the overreaction hypothesis, the majority
of people tend to analyze whether the behavior of “overreacting” to unexpected and dramatic news
events will affect stock prices in violation of Bayesian rules, and summarize the substantial forms of
weakness under low market efficiency [1]. McGrattann and Prescott studied whether the market was
overvalued at a level where the stock value of American companies was close to 1.8 times the US
GDP in the first half of 2000, a standard growth model based on the theory of aggregate economy
was constructed, and the conclusion was drawn that the market was not overvalued [2]. Guiso and
Sapienza et al. studied the impact of a general lack of trust on stock market participation, using micro
data from the Netherlands and Italy as well as cross-border data, and found that a lack of consistent
evidence of trust is an important factor in explaining the problem of limited participation [3]. Mazur,
Dang and Vega used the knowledge of currency circulation and the worst performing 8K and
DEF14A file data to analyze the performance of the U.S. stock market during the stock market crash
in March 2020 caused by COVID-19, and concluded that natural gas, food, health care and software
stocks have achieved high positive returns, while the stock values of oil, real estate, entertainment
and hotel industries have declined significantly. Besides, loss-making stocks exhibit extreme
asymmetric volatility, which is negatively correlated with stock returns [4].
Discussion on market effectiveness, Samuelson proposed that the stock market is micro efficient,
but macro ineffective. It indicates that the efficient market hypothesis has a better effect on individual
stocks than on the overall stock market. Therefore, researchers proposed a simple test based on
regression and simple scatter plots to prove the authenticity of Samuelson’s aphorism about US stock
market data since 1926 [5]. In order to study the main indicators of the development of the Pakistani
stock market and their possible correlation with actual economic activity during the post liberalization
period, Iqbal investigated market liberalization, integration of the market with global markets, and
issues on corporate governance, then compared them with a sample of developing and developed
markets. The results showed that the size of the Pakistani stock market was relatively small, the
sources of capital mobilization are relatively insignificant, and these factors decrease the value of the
stock market in promoting economic activity. In addition, due to the influence of noise traders and
speculators, the market seems too volatile. From a positive perspective, the market seems to have
brought huge returns to investors, compensating for the intensification of market volatility [6]. When
managers make decisions, should they follow the signals sent by the market and stock market, even
if this may differ from their assessment of fundamentals. In order to find the answer, Blanchard, Rhee
and Summers reviewed theoretical arguments and tested empirical evidence. They observed
investment behavior during the economic crashes of 1929 and 1987 and concluded that the role of
market valuation was limited under fundamental conditions [7].
Scholars used the latest developments in cointegration theory to examine the connections and
dynamic interactions between stock market trends. And it was found that before October 1987, there
was a strong linkage among stock markets in different countries. After October 1987, the international
linkage between stock price indices significantly increased, with the Nikkei Index being the only
exception. Moreover, the US stock market had a great impact on the French, German and English
Market markets afterwards. The results suggest that the reaction of French, German and English
Market markets to the innovation of American stock markets is the same as that of cross-border
information efficient stock markets. Lastly, it documented that the performance of the stock market
in Japan was not related to the US, France, Germany and the UK stock markets during this period [8].
Wattenberg created a new two-dimensional visualization algorithm that can calculate very detailed
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231057
50
information. This display method is based on Shneiderman’s tree graph technology, while utilizing
hierarchical and similarity information, and is highly popular [9]. Zuckerman believed that the value
of corporate stock presents the value of a company’s productive assets, and used the Standard Model
to adjust costs to prove that, and concluded that if assets only include capital goods, not permanent
monopoly franchise, then the value of securities measures the value of capital. Moreover, data from
American companies indicate that they have formed a significant amount of intangible capital,
especially in the past decade [10].
On this basis, the issue in this article is to construct the optimal investment portfolio, which is
beneficial for investors to make decisions. To achieve this goal, this paper will adopt effective cutting-
edge methods to study what issues have been addressed, and this research will contribute to which of
the investment point that investor will choose in the multiple optimal investment portfolio points.
The structure of the remaining of this research is as follows. Section II introduces the background
of each company and the changes in returns and stock prices, along with other data information.
Section III is about the methods and formulas used in the calculation of the article (Efficient Frontier
Model). Section IV explains the calculated results, which are the points of the optimal investment
portfolio. And there is also a discussion in this section. Section V summarizes, including the
advantages and disadvantages of the used method, research issues, and future improvements.
2. Data
2.1. Introduction of 8 Companies and Its Returns Rate
Among the 8 sample companies, AMD is an American multinational semiconductor company
develops computer processors in the field of technology for business and consumer markets. MRVL
is an American chip manufacturer specialized in manufacturing chips for storage, communication and
consumer electronics. NVIDIA is a Fabless manufacturing that mainly designs and sells graphics
processors. LRCX is a Science and technology in the United States company responsible for the
production, design and sales of semiconductor products. INTC is the world’s second largest
semiconductor company. QCOM is a radio communication technology R&D company. AVGO is a
Fabless manufacturing in the United States. TXN is an American multinational technology company
and it is famous for developing, manufacturing and selling semiconductors and computer technology.
This article visualizes the stock return data of 8 companies from 2018 to 2023, as shown in Figure
1, it can be concluded that the most volatile range for AMD’s returns is from August 2018 to February
2019, it experiences the lowest returns amount the 4 companies at around -0.55% in November 2018.
Also, there is a large float between May 2022 to May 2023, and located at 0.01% in the end of the
period. However, other time periods are comparatively gentle. For the MRVL, the period with
significant fluctuations is from February 2022 to June 2023, while the rest of the period is relatively
smooth. It peaked at June 2023 at 0.4%. Finally, it located at 0.2%. NVIDIA has the largest fluctuation
in the graph, except for relatively stable periods from August 2019 to November 2021, with other
periods experiencing significant fluctuations. It has the bottom at -0.4% in May 2022 and float to 0.1%
in June 2023. Moreover, LRCX is the almost the most common line in this graph, as it doesn’t
fluctuate as strongly as other lines. As for the final point it lies, is at around 0% in June 2023. The
returns of the INTC fluctuate heavily from February 2020 to May 2023, and it is shown in a blue line
in the graph. Its peak is at 0.28% in May 2023 and bottom figures at -0.22% at August 2022. Therefore,
QCOM has the largest returns among the 4 companies during 5 years periods at 0.41% in May 2019,
also include the lowest returns at -0.25% in August 2019. In this case, its short-term fluctuations are
significant, which show as a high risk. The AVGO is almost the slightest within the Figure 1, it
fluctuates heavily from November 2021 to June 2023, and peaks at 0.25% in May 2023. At other
times, it is comparatively slight than others. In addition, the return of TXN is peaked at 0.25%, that
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231057
51
there are two time periods reach this percentage, they are May 2020 and August 2022. However, it is
the slightest one whose risk is also lowest. It only has a bottom at -0.15%, which may larger than
other lines in this case.
Figure 1: Returns of 8 companies.
This article conducts descriptive statistics on the stock return indicators of 8 enterprises, and the
results are summarized in Table 1, the information on the average values, median, standard deviation,
minimum and maximum values of each stock shows that, INTC has the smallest mean value of -
0.0047, and the mean values of the other 7 companies are all positive. AMD has the smallest minimum
return of -0.5285, and it is also the most volatile stock as it has the largest standard deviation of 0.1681.
From the maximum return, QCOM is the largest and the value is 0.4123.
Table 1: Returns rate data information.
Mean
Median
Std.
Min
Max
AMD
0.0313
0.0313
0.1681
-0.5285
0.3865
MRVL
0.0176
0.0209
0.1279
-0.3066
0.3931
NVIDIA
0.0324
0.0606
0.1532
-0.3861
0.31
LRCX
0.0195
0.0221
0.1148
-0.201
0.2801
INTC
-0.0047
0.0052
0.0929
-0.2259
0.2704
QCOM
0.0108
0.0252
0.1142
-0.2539
0.4123
AVGO
0.0227
0.0316
0.0881
-0.2353
0.2544
TXN
0.0077
0.0111
0.0719
-0.1448
0.1521
TREASURY BOND
0.0016
0.0014
0.001
0.0002
0.0035
This article conducted a correlation analysis on the stock return of 8 enterprises, and the results
are shown in Table 2. Each of the companies can control their returns in 100%, which can completely
influence their own returns and the coefficient is shown as 1 in the table. Therefore, for the INTC,
apart from its own influence, AVGO has the greatest impact on it (coefficient is 0.5430). As for the
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
8/1/2018
10/1/2018
12/1/2018
2/1/2019
4/1/2019
6/1/2019
8/1/2019
10/1/2019
12/1/2019
2/1/2020
4/1/2020
6/1/2020
8/1/2020
10/1/2020
12/1/2020
2/1/2021
4/1/2021
6/1/2021
8/1/2021
10/1/2021
12/1/2021
2/1/2022
4/1/2022
6/1/2022
8/1/2022
10/1/2022
12/1/2022
2/1/2023
4/1/2023
6/1/2023
INTC QCOM AVGO TXN AMD MRVL NVIDIA LRCX
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231057
52
QCOM, TXN has the greatest impact on it (coefficient is 0.5720). As for VGO, the greatest impact
is brought by MRVL, with the correlation coefficient of 0.7246. For the TXN, the LRCX influences
its returns the most, and the correlation coefficient reach at 0.6704. In this case, the biggest impact
on returns of AMD is from NVIDIA, and the coefficient is 0.6936. The largest positive correlation
coefficient is from AVGO to MRVL, with the coefficient of 0.7246, except MRVL itself. As for the
NVIDIA, the largest positive correlation relationship is with AMD, and the coefficient is 0.6936.
Finally, the largest positive correlation coefficient for the LRCX is 0.6704, which is brought by TXN.
Table 2: Correlation coefficient of returns rate.
INTC
QCO
AVGO
TXN
AMD
MRVL
NVIDIA
LRCX
INTC
1
0.2743
0.5430
0.5380
0.2685
0.3851
0.4171
0.4433
QCOM
0.2743
1
0.5164
0.5720
0.5279
0.5006
0.5381
0.5423
AVGO
0.5430
0.5164
1
0.6504
0.5442
0.7246
0.6086
0.6081
2
TXN
0.5380
0.5720
0.6504
1
0.5266
0.6623
0.5480
0.6704
AMD
0.2685
0.5279
0.5442
0.5266
1
0.5697
0.6936
0.5610
MRVL
0.3851
0.5006
0.7246
0.6623
0.5697
1
0.6278
0.6137
NVIDIA
0.4171
0.5381
0.6086
0.5480
0.6936
0.6278
1
0.5595
LRCX
0.4433
0.5423
0.6081
0.6704
0.5610
0.6137
0.5595
1
3. Method
The effective frontier refers to rational investors who are risk averse and prefers returns. For the given
risk level, the investors will select a combination that can obtain the maximum return. For the given
expected return rate, the investors will select the combination with the minimum risk. The investment
portfolio that can simultaneously meet these two conditions is the effective set (also known as the
effective boundary or effective frontier). The combination on the effective boundary becomes the
effective portfolio [11].
Markowitz portfolio theory refers to a widely used theoretical model in the investment field, aimed
at helping investors optimize the balance between risk and return.
4. Results and Discussion
This paper uses the mean and the variance formulars, which is provided in eq. (1) – (4) to build
functions in order to find the variance in each company. In addition, expected returns are obtained by
using the limits portfolio risk. And further set the portfolio risk from 0.25 to 0.0566, and get the
expected returns as table 3.
!"#!$%&'( ) *+",$%&'
"(
#
$%# (1)
-. /. 0 ) 1* * +"2 +&33456&'
"7 '&(
#
$%#
#
$%# ) 8 (2)
-. /. *+") 9
#
$%# (3)
!"#!:; )'()!*+!!
, (4)
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231057
53
Table 3: Expected return.
Portfolio risk
Expected Return
0.2500
0.0808
0.2250
0.0747
0.2000
0.0682
0.1750
0.0613
0.1500
0.0537
0.1250
0.0450
0.1100
0.0393
0.1000
0.0351
0.0566
0.0060
Thus, the investment portfolio is shown in figure 2. It indicates that the optimal investment
portfolio point is (0.1407,0.0506), where the risk is 0.1407, and the return rate is 0.0506. In this case,
the weight in each company is 0.0502, -0.3867, 0.5289, 0.4367, -0.9987, -0.2804, 1, 0.6502 for AMD,
MRVL, NVIDA, LRCX, INTC, QCOM, AVGO, TXN, respectively. As a result, the investors put
largest percent of money into the AVGO, and do not invest in MRVL, INTC and QCOM. Because
they are negative in the weight. The study also allows mechanisms for buying and selling short, as
negative numbers are equivalent to selling short, and the weight 1 may means buying short. Therefore,
this study has obtained an optimal investment portfolio that is conducive to investors making optimal
investment decisions for the companies selected in this study.
Figure 2: Investment portfolio.
5. Conclusion
This article uses effective frontier methods to study the optimal investment portfolio, the research
finds that the optimal investment portfolio point is (0.1407, 0.0506), with a risk of 0.1407 and a return
of 0.0506. In this case, for AMD, MRVL, NVIDA, LRCX, INTC, QCOM, AVGO, and TXN, the
weights of each company are 0.0502, -0.3867, 0.5289, 0.4367, -0.9987, -0.2805, and 0.6502,
respectively. Therefore, the investors invest the maximum percentage of their funds in AVGO instead
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0 0.05 0.1 0.15 0.2 0.25 0.3
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231057
54
of MRVL, INTC, and QCOM, since the weights of MRVL, INTC, and QCOM are negative. The
study also allows short selling and buying mechanisms, as negative numbers of the weights are
equivalent to short selling, and a weight of 1 may mean short selling. Therefore, this study has
obtained an optimal investment portfolio that is conducive to investors making the optimal investment
decisions for the companies chosen in this research.
The conclusion of this study contributes to help investors better build an investment portfolio and
give them some valuable investment advice. Building an investment portfolio is of great benefits for
investors. Firstly, it is conducive to risk diversification. An investment portfolio involves dispersing
funds into different types of assets, thereby reducing investment risks. Thus, when one asset performs
poorly, losses can be balanced by the appreciation of other assets. Secondly, the diversified
investment portfolios can help the investors to earn the certain returns by investing different assets.
As the financial goal and timing of each investor differs, different investment strategies are required.
The investment portfolios derived from effective frontier method adopted in this study can be tailored
to the needs of different investors.
However, this article has limitations, for example, the paper does not use other models to construct
the portfolio and make the comparisons, including in-depth Mathematical analysis model, Fama-
French factor model, machine learning model. In the future study, those models should be paid more
attention and further discussed.
References
[1] De Bondt, W. F., & Thaler, R. (1985). Does the stock market overreact. The Journal of finance, 40(3), 793-805.
[2] McGrattan, E. R., & Prescott, E. C. (2001). Is the stock market overvalued? (No. w8077). National Bureau of
Economic Research. Retrieved from https://www.nber.org/papers/w8077
[3] Guiso, L., Sapienza, P., & Zingales, L. (2008). Trusting the stock market. the Journal of Finance, 63(6), 2557-2600.
[4] Mazur, M., Dang, M., & Vega, M. (2021). COVID-19 and the march 2020 stock market crash. Evidence from
S&P1500. Finance research letters, 38, 101690.
[5] Jung, J., & Shiller, R. J. (2005). Samuelson’s dictum and the stock market. Economic Inquiry, 43(2), 221-228.
[6] Iqbal, J. (2012). Stock market in Pakistan: An overview. Journal of Emerging Market Finance, 11(1), 61-91.
[7] Blanchard, O., Rhee, C., & Summers, L. (1993). The stock market, profit, and investment. The Quarterly Journal of
Economics, 108(1), 115-136.
[8] Bustos, O., & Pomares-Quimbaya, A. (2020). Stock market movement forecast: A systematic review. Expert Systems
with Applications, 156, 113464.
[9] Wattenberg, M. (1999, May). Visualizing the stock market. In CHI’99 extended abstracts on Human factors in
computing systems (pp. 188-189).
[10] Zuckerman, E. W. (2004). Structural incoherence and stock market activity. American Sociological Review, 69(3),
405-432.
[11] Akhilesh G. (2023). Efficient Frontier: What It Is and How Investors Use It. Retrieved from June 28, 2023. Available
at: https://www.investopedia.com/terms/e/efficientfrontier.asp
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231057
55
The Impact of Systemic Financial Risks on the Shanghai
Composite Index: Evidence from ARIMA Model
Yuqi Li1, a, *
1Columbia Engineering, Columbia University, New York, 28051130, United States
a. 631401100112@mails.cqjtu.edu.cn
*corresponding author
Abstract: Systemic financial risk has always been a topic of concern to financial regulators
and investors, and it often has a profound impact on the securities market. In 2015, China’s
stock market fluctuated dramatically from a bull market to a bear market. The sharp drop in
June of that year was an important turning point, resulting in a huge systemic financial risk,
which led to a subsequent bear market. This paper uses ARIMA model to study the expected
trend of the Shanghai Composite Index after the occurrence of systemic risk. Daily, weekly,
and monthly data are used to analyze separately, and compared with the actual trend in order
to study the impact of systemic financial risk on the Shanghai Composite Index. The study
finds that systemic financial risk has a significant negative effect on the Shanghai Composite
Index. The research in this paper aims to provide investors with a cognition and explanation
of systemic financial risk, and to propose policy recommendations.
Keywords: systemic financial risk, Shanghai Composite Index, ARIMA model, impact
1. Introduction
Systemic financial risks can be defined as: financial risks that may have a significant impact on the
normal development of financial services, and then have a huge negative impact on the real economy.
Xiaopu, Z finds that SFR has different meanings. It represents the whole financial system risk. Also,
it affects most of industries deeply in some cases [1].
Chinese stock market is a policy-dependent stock market. According to the Rong-Cai, H. U.’s
research on policy events in China’s stock market and the Shanghai Composite Index from January
2005 to June 2009, it is found that there were many policy events in Chinese stock market in 2007
and 2008, indicating that the intensity of policy intervention in the market was relatively strong [2].
Using the K-means clustering method, Shu-Lin, W. found that investors were roughly divided into
the following three groups: activist investors, passive investors, and professional investors. Short-
term products or slightly higher-risk products are recommended for activist investors, while long-
term low-risk products are recommended for professional investors and passive investors. Unlike
professional investors, both activist and passive investors hope to make investment decisions through
short-term information [3].
On June 19, 2015, the SSEC fell sharply in a single day. The stock market has shifted from a bull
market to a bear market, resulting in a financial panic. This paper aims to study the impact of SFR on
SSEC, and put forward policy recommendations and investor recommendations. Using ARIMA
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231055
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
56
model to forecast the SSEC, the paper compares forecast data with actual data after June 19, 2015,
and finally draws a conclusion.
2. Literature Review
Existing studies have several researches on the impact of SFR on the stock market, and the research
conclusions are also different.
Al-Qudah A assessed the impact of systemic risk on stock return of industrial companies listed on
the Amman Stock Exchange. Through regression analysis, this paper found no statistically significant
relationship between systemic risk and annual stock returns. [4].
Nugroho M, Arif D, Halik A found that with greater systematic risk, stock price would closely
follow the market index. Investors were likely to predict company stock prices accurately with the
close relationship between stock price and market index, which results in a more transparent market.
Thus, the expected return of investors would be much higher [5].
Maghsoud et al. studied Tehran Stock Exchange. The results showed that different factors
influence stock returns in different situations. Specifically, oil prices and inflation had a greater
impact on stock returns than interest rates and exchange rates. It is recommended to use models to
separate institutional transitions at various levels of risk when predicting stock returns [6].
Zolotoy L studied the relationship of systemic risk and the Fama-French Model and found that
market factor, SMB factor and HML factor showed different status in regard to new information.
Market factor seemed to expand negative shocks to the market portfolio. The paper also found there
was no evidence of the relationship of SMB factor to information shocks. Finally, the paper found
that the HML factor tended to weaken negative shocks to the portfolio [7].
Yang Z H examined Chinese stock market in 2020, and analyzed the risk transmission among
various departments of financial market during the epidemic. The results indicated the event caused
a negative influence on various departments after the Spring Festival. Also, the influence of the event
on financial market was relatively short-lived [8].
Yang, L. measured system liquidity of 50 constituent stocks of Shanghai Stock Exchange from
2003 to 2004, and constructed a Fama-French three-factor model. The research showed that excess
returns contain a premium for system liquidity risk, and system liquidity risk would affect stock prices
and returns [9].
Shengyao, L. discussed the systemic risk in Chinese stock market, and studied particular crash risk
and expected return. The results indicated that there was a positive impact of crash systemic risk on
expected return. This paper verified that investors’ gaming stock preferences and auspicious number
preferences also had a significant impact on the expected return [10].
Existing literatures have roughly studied relationship between SFR and stock returns in various
industries, and found that the relationship differed in different market and situation, while there is a
lack of research on indices. In addition, the SFR during the 2015 stock market crash in China also
need to be studied.
3. Model and Variables
3.1. Data Source
This paper selects SSEC closing price data from January 2010 to August 2015 as the research object,
and uses daily, weekly, and monthly data for analysis. The data source is from WIND database.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231055
57
3.2. Model Specification
This paper chooses a relatively mature stock price forecasting model, that is, the ARIMA model which
is simple and convenient, easy to estimate and has strong short-term forecasting ability. The ARIMA
model is a time-series short-term forecasting method with high accuracy. Therefore, this paper
decides to establish an ARIMA model to make predictions. Finally, this paper compares the predicted
results with actual data to identify and explain the difference.
ARIMA model is a time series forecasting method proposed by Box and Jenkins in the early 1970s
[11]. ARIMA model differentiates the time series data to make the data stationary. The expression of
this model is showed as follows:
𝑤!= 𝑐 + 𝜙1𝑤!#1 + 𝜙2𝑤!#2 + ⋯ + 𝜙%𝑤!#% + 𝜀!+ 𝜃1𝜀!#1 + 𝜃2𝜀!#2 + ⋯ + 𝜃&𝜀!#& (1)
In the formula,the parameter 𝑐 is a constant. {𝑤!} is the variable after the difference, that is, 𝑤!=
𝑧!− 𝑧!#1, 𝑧! is the corresponding logarithm of Shanghai index data. {𝜀!} is a white noise sequence.
-𝜙1, 𝜙2⋯ 𝜙% are autoregressive coefficients. 𝜃1, 𝜃2⋯ 𝜃& are moving average coefficients.
4. Empirical Results and Analysis
4.1. Stationarity Test
In this paper, Augmented Dickey-Fuller test is used to test statistics to determine whether the time
series data is stable. The results of the stationarity test are shown in Table 1. The daily data are stable
after the first-order difference and can be ordered. Weekly and monthly data are stable after the first-
order and second-order differences, while the first-order difference cannot be ordered, and the second-
order difference can be ordered. Therefore, this paper uses the second-order difference to model
weekly and monthly data.
Table 1: Weak stationarity test.
t
p
Daily
Raw
1.310
1.0000
1st order difference
-26.384
0.0000
Weekly
Raw
0.401
0.9966
1st order difference
-10.314
0.0000
2nd order difference
-19.248
0.0000
Monthly
Raw
1.014
1.0000
1st order difference
-5.926
0.0000
2nd order difference
-10.686
0.0000
4.2. Model Identification
In this section of the article, it is first necessary to order the daily, weekly and monthly series using
the PACF and ACF pairs, the results of which are shown below Figure 1.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231055
58
PACF
ACF
Daily
Weekly
Monthly
Figure 1: ARMA (p, q) identification.
Photo credit: Original
Firstly, this paper needs to order the logarithm of the daily SSEC with first-order difference and
the result is showed in the figure above.
According to the result of the two images in the first row in Figure 1, the maximum value of the
significant order less than or equal to 10 of x-axis is 10, so AR(P) is of order 10. MA(q) is also of
order 10. Thus, the value of p and q is 10.
Then this paper needs to order the logarithm of the weekly SSEC with second-order difference and
the result is showed in the figure above.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231055
59
According to the result of the two images in the second row in Figure 1, the maximum value of
the significant order less than or equal to 10 of x-axis is 7, so AR(p) is of order 7. MA(P) is of order
1, so the value of p and q are 7 and 1 respectively.
Finally, this paper needs to order the logarithm of the monthly SSEC with. second-order difference
and the result is showed in the figure above.
According to the result of the two images in the third row in Figure 1, the maximum value of the
significant order less than or equal to 10 of x-axis is 3, so AR(p) is of order 3. MA(P) is of order 1,
so the value of p and q are 3 and 1 respectively.
4.3. Model Testing
This paper uses Portmanteau Q test to check whether the residual sequence is a homogeneous or auto-
correlative sequences. The result is shown in Table 2.
Table 2: Residual test.
Model
Portmanteau (Q) statistic
Prob > chi2
Daily-ARIMA(10,1,10)
144.3737
0.0000
Weekly-ARIMA(7,2,1)
51.4829
0.1054
Monthly-ARIMA(7,2,0)
40.9840
0.4272
It is found from Table 2 that the Q statistics of the daily data are greater than the critical value,
thus the residual sequence is not a white noise process. The Q statistics of the weekly and monthly
are both smaller than the critical value, thus both the data are all white noise processes. This
conclusion shows that the model setting using weekly and monthly data is more reasonable.
4.4. Prediction
The B-J method adopts L-step prediction, that is, according to the known n-time sequence
observations 𝑤', 𝑤(⋯ 𝑤), to estimate the sequence value of n + L times in the future. Linear
minimum mean squared error method is the most used one. The following gives ARIMA (10,1,10),
ARIMA (7,2,1) and ARIMA (3,2,1) to predict the trend of the SSEC after June 18, 2015 respectively,
and compare it with the actual trend, as shown in Figures 2, 3 and 4.
Figure 2: Actual value and fitted value, daily.
Photo credit: Original
3900
4100
4300
4500
4700
4900
5100
5300
2015-06-04
2015-06-06
2015-06-08
2015-06-10
2015-06-12
2015-06-14
2015-06-16
2015-06-18
2015-06-20
2015-06-22
2015-06-24
2015-06-26
2015-06-28
2015-06-30
2015-07-02
Actual value Fitted value
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231055
60
Figure 3: Actual value and fitted value, weekly.
Photo credit: Original
Figure 4: Actual value and fitted value, monthly.
Photo credit: Original
According to above figures, after June 19, 2015, the actual performance of the SSEC differs
obviously from the expected performance. According to the ARIMA model, the SSEC is expected to
show an oscillating upward trend in the next two weeks for daily data, while the actual SSEC has
sharply declined by 18% in next two weeks. For weekly data, the expected index also shows an
upward trend, while the actual SSEC has declined by 10% in a month. And for monthly data, the
expected index tends to show a straight upward trend, while in fact the SSEC has shown a cliff-like
decline by 25% from Jun 2015 to Aug 2015. Specifically, the average difference in weekly data is
15.90%, as shown in Table 3.
3500
3700
3900
4100
4300
4500
4700
4900
5100
5300
42127
42134
42141
42148
42155
42162
42169
42176
42183
42190
42197
42204
42211
42218
Actual value Fitted value
3100
3600
4100
4600
5100
5600
6100
42064 42095 42125 42156 42186 42217
Actual value Fitted value
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231055
61
Table 3: Weekly difference.
Year 2015
6/21
6/28
7/5
7/12
7/19
7/26
8/2
Predicted
value
4565.29
4577.00
4489.17
4625.55
4662.24
4625.27
4791.31
Actual value
4192.87
3686.92
3877.80
3957.35
4070.91
3663.73
3744.20
5. Conclusion
This paper studies the impact of SFR on the SSEC from an empirical perspective. In order to obtain
the results, this paper use the SSEC data from January 2010 to August 2015 with feasible ARIMA
model. Daily, weekly, and monthly data of the SSEC are used to fit the tendency of the ones after
June 19, 2015. Compared with the actual trend, it is found that SFR has a significant negative impact
on the SSEC.
Most of existing literatures take stock return as the research object, and there is not much research
on the index performance. The study in this paper focuses on the SSEC. In addition, the existing
research mainly use regression analysis to identify the relationship between SFR and stock market,
and rarely use the method of predicting the index to study the one. This paper mainly uses ARIMA
model to predict the SSEC, which can reflect the degree to which SFR has affected the SSEC.
SFR have a strong short-term impact on the stock market, and it is necessary to reduce the
likelihood of SFR from a policy perspective. Three policy proposals are put forward as following.
Improve the censorship of local government debt issuance, rationally plan the use of debt funds, and
promote bond market-oriented reform. Improve the financial prevention and control system, steadily
promote the internationalization of RMB cross-border settlement, and construct a domestic and
international double-cycle pattern. Strengthen corporate information disclosure to make financial
market more transparent and avoid financial manipulation.
In a word, the conclusions of this paper may be used to help investors to identify investment risks
and discover investment opportunities. Consideration can be given to combine other forecasting
factors such as the stock market’s own institutional factors, national macroeconomic policies, and the
domestic and foreign market environment. These factors also have certain research value to the stock
market. Using this model to make short-term forecasts of the overall market trend can provide
investors with a basis for investment decisions.
References
[1] Xiaopu, Z. (2010). A study on systemic financial risk:evolution,causes and supervision. Studies of International
Finance.
[2] Rong-Cai, H. U. (2010). New characteristics of policy market in china’s stock market. The Theory and Practice of
Finance and Economics.
[3] Shu-Lin, W., & Zhen-Xi, Z. Design suggestions to the financial products based on the population characteristics of
investors. Journal of Harbin Institute of Technology.
[4] Al-Qudah, A. (2013). The effect of financial leverage & systematic risk on stock returns in the amman stock
exchange (analytical study – industrial sector). Research Journal of Finance & Accounting, vol.4(No.6), 136-145.
[5] Nugroho, M., Arif, D., & Halik, A. (2021). The effect of financial distress on stock returns, through systematic risk
and profitability as mediator variables. Accounting.
[6] MAGHSOUD, Hosein†, RAHNAMAY-ROODPOSHTI, Fraydoon´, & VAKILIFARD. The impact of time-varying
systemic risk on predicting the dynamics of stock return volatility in tehran stock exchange.
[7] Zolotoy, L. Informations Shocks, Systemic Risk and the Fama-French Model: Evidence from the US Stock Market.
[8] Yang Z H, Chen Y T, Zhang P M. (2020). Macroeconomic shock, financial risk transmission and governance
response to major public emergencies. Management World, 36(5), 25.
[9] Yang, L. (2008). The systematic liquidity risk and premium of shanghai stock exchange. Chinese Journal of
Management.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231055
62
[10] Shengyao, L. Yi-Tsung, L. , & Yunhong, Y. . (2016). The systematic crash risk and investor preference in chinese
stock market. Journal of Financial Research.
[11] Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (1976). Time series analysis forecasting and control - rev. ed. Journal
of Time, 31(2), 238-242.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231055
63
Research on the Relationship Between International Trade
and Network
Run Guo1, a, *, Tongxi Wang2, Chengyi Xu3, Yichen Zhou4
1Zhenhai High School International Center, 315099, China
2Penon Education, Chengdu,310011, China
3Tianyuan Public School, Hangzhou, 610000, China
4Nanjing Jinling High School Hexi Campus, Nanjing, 320106, China
a. guorun@seu.edu.mk
*corresponding author
Abstract: In a rapidly globalizing world economy, international trade is undergoing
unprecedented growth, marked by technological advancements, energy demand, and
currency fluctuations. This study explores the intricate interplay between trade networks and
international commerce. As the global economy transforms, this research sheds light on key
trends and characteristics shaping the international trade landscape. Through an extensive
review of relevant literature, this paper investigates the role of trade networks in international
trade dynamics. It delves into the significance of network structures and their impact on trade
patterns, revealing insights about the core-periphery structure where advanced economies
dominate while others form the periphery. The study focuses on the international trade
network's features, including its virtualization, globalization, and intelligent attributes. It
examines how businesses adapt to digitalization, interact through virtual platforms, and
embrace the knowledge-driven nature of modern trade networks. Additionally, the research
explores the implications of trade networks for enterprises, emphasizing market penetration,
supply chain efficiency, brand influence, technological innovation, and resource integration.
The study highlights the works of various scholars who have analyzed international trade
networks. Research findings underscore the importance of understanding network structures,
role centrality, and the impact of digitalization on global trade. By analyzing the evolution of
trade networks and their implications for different economies, this study offers valuable
insights for policymakers and businesses navigating the complex landscape of international
trade. Overall, this thesis contributes to a comprehensive understanding of the relationship
between international trade and network structures, paving the way for more informed
decisions in a rapidly changing global trade environment.
Keywords: trade network, international commerce, core-periphery structure
1. Introduction
1.1. Background
With the development of economic globalization, international trade has rapidly spread to various
corners of the world and entered a new stage of rapid growth.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231051
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
64
The rapid development of international trade often has the following characteristics. When we
focus on economic globalization and international service trade, we can find that industries such as
financial products and telecommunications that provide services to other companies have increased
in the total trade share. In contrast, businesses that promote regional connections have become more
popular, such as tourism and long-distance telephone services, whose share continues to decline;
When we focus on the roles played by different countries in international trade, developing countries
attach greater importance to international trade performance, which also benefits from their
technological achievements.
Research related to international trade has become a global focus. Theories related to international
trade, such as the theory of demand differences, the theory of demand similarity, and the theory of
absolute advantage, have received a lot of attention. Many scholars have also begun to study the
topological structure in international trade, which is the network structure formed by international
trade (scholars use countries as nodes and represent the import and export between trading partners
with lines, forming an international trade network with research value). By analyzing the
characteristics of network structure or using data to analyze the relevant properties of the network,
scholars can not only analyze the roles played by different trading partners in the network and the
market characteristics of different types of trade goods, but also infer the trend of international trade
development and the changes in the network structure over time, and propose improvement
suggestions for the development of international trade in different countries. This is a research with
multiple meanings and is very interesting.
1.2. Related Research
Tajoli et al. investigated the networks of international trade in services by employing the BaTis
database. The study examines global properties of service networks, identifies differences and
similarities among them, and highlights influential countries in specific service trades. Two distinct
sets of service categories with different network structures are discovered. The structure of links in
the networks clarifies countries' roles more effectively than the amount of services traded. While most
countries are connected in service trade, a small group of countries dominates the majority of flows.
The central positions in service networks are predominantly occupied by advanced countries while
emerging and developing nations rarely display high centrality [1]. Sajedianfard et al. addressed the
common issue of missing data in trade network analysis, where some countries do not report trade
flows. The author explores suitable data and methods to deal with missing data, presenting results
using key network measures. The study uses an all-to-all potential connectivity approach,
reconstructing a more complete dataset to study trade relations. The findings show that the imputed
bigger network is robust for analyzing trade relations, providing richer insights, especially for non-
reporting countries. The research emphasizes the importance of considering as many countries as
possible in trade networks for policy-making purposes [2]. Fan et al. used a symmetric link weight
matrix and international trade partner data to conclude that the international trade network has the
characteristics of a central periphery, and there is a fact that a few countries exploit other countries.
In addition, by exploring the microstructure and community structure of trade, it was found that trade
centers such as the United States and China have made the world more interconnected, but the pattern
of uneven economic development will continue to exist. Finally, through the importance of nodes and
Bootstrap infiltration, it is concluded that the European Union holds significant importance in world
trade, while the US trade advantage will gradually weaken [3].
Baskaran et al. estimate a feature parameter that reflects the topological structure of the trading
network for 28 product groups and describes the evolution of the international trade structure.
Afterward, when testing the predictions of the “Herschel Olin” model, the importance of the network
in international trade was demonstrated by clearly explaining the scale characteristics of the network.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231051
65
The model believed that differences in factor endowments determined bilateral trade flows. The
results indicate that differences in factor endowments increase bilateral trade in goods traded in a
“decentralized” network [4].
Maluck and Donner described the complex web of trade relationships in the contemporary
economy, due to the complex supply chains. To understand the complex interrelationship between
different countries and economic sectors, the author develops a comprehensive understanding of the
fundamental structural characteristics of this network. In this paper, the authors use the input-output
data of several regions, decompose 186 national economies into 26 industry sectors, and analyze the
structure of the international trade network from 1990 to 2011 using the method of interdependence
network. At the same time, by studying these complex networks, new insights of global trade are
gained, to identify key elements of the global economy. Through specific studies, the authors found
that specific industrial sectors tend to favor different patterns, such as the selection of geographical
locations, and close countries tend to trade more frequently. As for the evolution of the structure of
trade networks, the paper gives an example of an extreme event, the 2008~2009 financial crisis, which
showed an anomaly compared to the normal annual fluctuations in the structure of the network
leading to a significant reorganization of related trade patterns (tracked and quantified by a widely
available online promotion) [5]. Culture and the network can interact based on their impact on
international trade to mitigate the negative impact of cultural differences. Lee investigated the impact
of the internet and culture on international trade, as well as the possible interactions between the two.
The author uses bilateral datasets from 34 OECD countries to sustain the positive impact of cultural
relationships and networks on trade. This article also explored how various types of networks affect
national trade, such as the “interactive effect” mentioned in the article. Is means that networks such
as foreign direct investment, immigration, and the Internet play an important role in reducing the
deterrent effect of cultural differences on international trade. The author also proves that the
interactive effect of the Internet is the strongest, followed by foreign direct investment and
immigration [6]. International trade is an important form of economic exchange between countries,
and it is deeply influenced by international relations, which in turn affects the evolution of
international relations. Only by establishing and maintaining good and stable international relations
can favorable external conditions be provided for international trade, promoting the prosperity and
development of international trade, and thus greatly promoting the economic growth and social
progress of various countries. The development of international relations not only relies on the leading
role of national governments and news media but also requires the active participation and effective
support of various social forces and civil society organizations. Only in this way can the international
market environment be more just and reasonable, enabling countries to enjoy equal rights and
obligations internationally and engage in fair cooperation and competition.[7]. Baskaran et al. gave
the parameters of the study reflect the changes in the international trade structure between 1980 and
2000, showing the topological structure of the trading network and demonstrating the importance of
the network in international trade. The research results indicate that the donation gap has increased
the bilateral goods trade network that trades in a “decentralized” manner. For goods traded in a
“centralized” network, differences in factor endowments are less important [8].
2. Theory Description
2.1. Adjacency Matrix
The adjacency matrix can clearly show the relationship between the various countries, which is a
dominant concept in graph theory [9]. There is a simple undirected graph with four nodes: A, B, C,
and D. The adjacency matrix for this graph would be a 4x4 matrix, where each row and column
represent a node, and the entries indicate whether there is an edge connecting the nodes or not. The
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231051
66
value “1” is typically used to indicate a connection between nodes, and “0” represents no connection.
The legend for the adjacency matrix is shown in Table 1.
Table 1: Example of adjacency matrix.
A
B
C
D
A
0
1
0
1
B
1
0
1
0
C
0
1
0
0
D
1
0
0
0
From the matrix, it can interpret the following connections:
- There is an edge between nodes A and B.
- There is an edge between nodes A and D.
- There is an edge between nodes B and C.
2.2. Degree and Degree Distribution
In graph theory, the degree of a node (also known as vertex) in a graph is defined as the number of
edges that are incident to that particular node [10]. In a social network graph, the degree distribution
might indicate that most people have a small number of friends (nodes with low degrees), but there
are a few individuals with an extensive network of connections (nodes with high degrees).
Understanding the degree distribution can help identify influential nodes, detect communities, and
infer properties of the network's overall structure.
2.3. Core-Periphery Structure
A core-periphery structure refers to a network or system where there is a central core that is more
developed, economically advanced, and influential, while the surrounding periphery regions are less
developed and have lower levels of economic activity and influence. This concept is often used to
describe economic, social, or political relationships among different regions or countries.
In a trade network context, the core-periphery structure can be observed in how certain countries
or regions dominate in terms of trade and economic activities, while others play more peripheral roles.
For example, in the global trade network, countries like the United States, China, and Germany can
be considered part of the core due to their significant economic activities, high trade volumes, and
advanced industries. On the other hand, smaller and less developed countries may be situated on the
periphery, with limited participation in global trade and a dependency on the core countries for
imports and exports. This structure can lead to imbalances in economic power and development
between core and peripheral regions.
3. Characteristics of the International Trade Network
3.1. Virtualization
The development of ITN is based on modern information technology. Contemporary information
technology, such as computer systems and information processing, allows people to trade in virtual
economic venues, such as virtual commodity markets and virtual financial institutions. In the INT,
both producers and consumers interact through digital information as an intermediary. Both parties
do not need to meet to complete transactions, greatly reducing the time and cost of transactions, and
reducing a series of costs in the operation of international trade. But this kind of virtuality is not
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231051
67
nihilistic, both parties engage and sign contracts through telecommunications networks, and
economic activities are taking place.
3.2. Globalization
ITN has broken through the temporal and spatial limitations of commercial activities, as it uses
digitization and informatization as a carrier for communication and interaction, allowing for
unrestricted trade in both time and place. Various trading countries can quickly find trading partners
through a worldwide computer network, complete trade activities quickly, improve trade efficiency
and reduce trade costs. Of course, this is closely related to the characteristics of the Internet itself, it
is because it has no national boundaries. Just by connecting to the international internet, people from
any country or region can achieve trade communication and cooperation, and global resources can
also be shared.
3.3. Intelligent Features
ITN is a trade activity that utilizes new means based on network technology, which not only has the
characteristics of virtualization and globalization but also includes the characteristics of intelligence.
In this network, people place more emphasis on technology and knowledge, rather than just focusing
on the product itself. Information has become the most important strategic resource. Having advanced
knowledge or technology within the network can provide important resources in the fiercely
competitive international trade market, attract more customers, and ultimately increase internal
revenue.
4. The Precedent of International Trade Networks
4.1. Overview of Macroscopic Research on the Current Situation of International Trade
Networks
Globalization and international relations are two important characteristics of the contemporary world,
which interact and influence each other, jointly shaping the prospects for diversified development of
contemporary international trade. International relations determine the stability and security of the
international trade environment, affect the willingness and ability of countries to participate in
international trade and shape their position and role in international trade.
Under the impact of the digital economy wave, significant changes have taken place in the
transaction targets, transaction objects, transaction methods, and even the production organizations
behind international trade activities. Digital trade has the characteristics of lower production and
transaction costs, more significant economies of scale and scope, higher organic capital composition,
platformization of transactions, and networked supply chains.
4.2. The Current Situation and Influencing Factors of International Trade Networks Faced
by Enterprises
The trade network of enterprises has significantly promoted their export domestic value-added rate.
The more developed a company's trade network is, the higher the proportion of domestic production
factors in its exported products, thereby increasing the domestic value-added rate of exported
products. The trade network of enterprises has a significant impact on the competitiveness of exported
products. Specifically, the following are the impacts:
Market penetration ability: The wider a company's trade network, the stronger its market
penetration ability for exported products. Trade networks can help companies better understand the
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231051
68
needs and trends of target markets, thereby better meeting market demand and improving product
market share and competitiveness.
Supply Chain Efficiency: A company's trade network can help establish a more efficient supply
chain, thereby reducing production costs and improving product price competitiveness.
Brand influence: A company's trade network can help expand its brand influence, enhance brand
awareness and reputation, and thereby increase its market share and competitiveness.
Technological innovation capability: The trade network of enterprises can help them obtain more
technological innovation resources and information, thereby improving their technological
innovation capability, promoting continuous product upgrading and innovation, and enhancing
product competitiveness.
Resource integration capability: The trade network of enterprises can help them integrate various
resources, including talent, technology, capital, etc., thereby improving their resource integration
capability, promoting their continuous development and growth, and improving the competitiveness
of their products.
5. Conclusion
In conclusion, INT becoming increasingly famous around the world, and it has grown significantly
faster and constantly expanding. The analysis results show that services are a key factor in the
globalization process, and In the future, services will become increasingly important. Trade networks
are changing over time, becoming intelligent, globalized, and virtualized. Nowadays, the position of
countries (Naturally, nodes of the INT represent different countries in the world.) is also very
important, as they need to monitor their evolution for a correct understanding of global market trends.
At the same time, monitoring the co-evolution of trade networks in goods and services is crucial, not
only to highlight the similarities between products and networks but also to reflect their differences.
We also need to use technology or data analysis to understand their increasingly intertwined
relationships, so that future trade networks can be more perfect and developed.
Authors Contribution
All the authors contributed equally and their names were listed in alphabetical order.
References
[1] Tajoli L, Airoldi F, Piccardi C. The network of international trade in services[J]. Applied Network Science, 2021,
6(1): 1-25.
[2] Sajedianfard N, Hadian E, Samadi A H, et al. Quantitative analysis of trade networks: data and robustness[J].
Applied Network Science, 2021, 6(1): 46.
[3] Fan Y, Ren S, Cai H, et al. The state's role and position in international trade: A complex network perspective[J].
Economic Modelling, 2014, 39: 71-81.
[4] Baskaran T, Blöchl F, Brück T, et al. The Heckscher–Ohlin model and the network structure of international
trade[J]. International Review of Economics & Finance, 2011, 20(2): 135-145.
[5] Maluck J, Donner R V. A network of networks perspective on global trade[J]. PloS one, 2015, 10(7): e0133310.
[6] Lee J H. Culture, network, and international trade[J]. Journal of International Logistics and Trade, 2015, 13(2):
21-31.
[7] Chen Z .On the Influence of International Relations on International Trade[J].Modern Economics & Management
Forum, 2022, 3(1):30-33.
[8] Baskaran T, Blöchl F, Brück T, et al. The Heckscher–Ohlin model and the network structure of international
trade[J]. International Review of Economics & Finance, 2011, 20(2): 135-145.
[9] Nagoor Gani, A., Latha, S.R. A new algorithm to find fuzzy Hamilton cycle in a fuzzy network using adjacency
matrix and minimum vertex degree. SpringerPlus 5, 1854 (2016).
[10] Avrachenkov, K., Markovich, N.M. & Sreedharan, J.K. Distribution and dependence of extremes in network
sampling processes. Compu Social Networks 2, 12 (2015).
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231051
69
Stock Price Forecasts Based on KNN and LSTM
Zihao Chen1, a, *
1College of the Liberal Arts, Pennsylvania State University, State College, US
a. Zfc5178@psu.edu
*corresponding author
Abstract: Every stock trader wants to successfully predict the price or trend of a stock in order
to make a profit because stock price forecasts provide investors, traders, and financial
professionals with signals about potential price movements, which can help them make more
informed decisions about buying, selling, or holding stocks. This article selects the four
largest stocks in the U.S. stock market by market capitalization: Google, Apple, Microsoft,
and Amazon, and predicts their closing prices from 2013 to 2023. First, K-Nearest Neighbors
(KNN) model is established for the closing price sequence after the first-order difference.
Then a two-layer LSTM model is constructed to visualize the prediction results of the two
models, and RMSE is calculated respectively. Comparing the prediction results of the two
models, LSTM has a better prediction effect on the data set used in this paper. This paper
finds that the LSTM model can capture the crucial time dependencies and relationships in
financial time series data, which are essential for stock price prediction. Therefore, the LSTM
model can often be used when predicting stocks in the future.
Keywords: stock market, KNN, LSTM, time series
1. Introduction
The research on stock price prediction is of great significance to investors and practitioners in the
financial industry. People want to model stock prices correctly so that, as stock buyers, they can rely
on stock price prediction to make wise decisions about buying, selling, or holding stocks. Accurate
forecasting helps optimize a portfolio and may lead to higher returns. At the same time, accurate share
price forecasting helps manage risk by providing insight into potential price movements. Investors
can adjust strategies and positions to mitigate potential losses from market volatility. Therefore, the
issue of stock forecasting is very important. Nevertheless, predictions must be made with a realistic
understanding of their limitations and potential risks, as financial markets are subject to many factors,
many of which are unpredictable or influenced by human behavior. This inherent complexity makes
accurate predictions challenging.
Numerous prior research endeavors have employed machine learning methodologies for the
purpose of forecasting stock values. These methodologies encompass a range of models such as linear
regression [1], autoregressive integral moving average (ARIMA) [2], random forest [3], decision tree
[4], support vector machine (SVM) [5], and others. However, previous investigations still appear to
be limited. Given the rapid development of machine learning methods, this paper selects the historical
data of the four companies with the largest market value in the U.S. stock market from 2013 to 2023
and uses KNN and LSTM models to forecast the data of these four companies.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231064
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
70
2. Data
This article uses data from Yahoo Finance and selects the closing prices of the four largest U.S. stocks
(Google, Apple, Microsoft, Amazon) from 2013 to 2023, a total of 2,517 data pieces. Closing prices
reflect the sentiment and decisions of stock market traders throughout the trading day. It involves a
collective assessment of the value of a stock by buyers and sellers. Some basic information of the
assets is shown in the following Figure 1 and Figure 2.
Figure 1: Ten-year price fluctuations of each stock.
Figure 2: The correlation of each stock.
From Figure 1, we can see how the closing prices of these four companies have fluctuated over
ten years. Furthermore, Figure 2 shows that they have a strong similarity.
3. Methodology
This paper mainly uses KNN and LSTM to forecast the closing prices of Google, Apple, Microsoft,
and Amazon from 2013 to 2023. T The predictive performance is evaluated by computing the RMSE
between the predicted and actual values.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231064
71
3.1. KNN
K-Nearest Neighbors (KNN) model is based on classification and regression. KNN is an algorithm
used in machine learning for classification and regression tasks [6]. The K-nearest Neighbors
algorithm operates under the assumption that a training dataset is provided, containing instances
labeled with their respective categories. During the process of classification, the prediction of a new
instance is determined by a majority voting approach, wherein the category assigned to the instance
is based on the categories of its K Nearest Neighbors. KNN focuses on local patterns in data and can
adapt to changing market conditions and capture changes in trends, making them suitable for specific
dynamic market environments. In financial markets, short-term trends and patterns can often be
effectively captured using local methods.
The KNN algorithm utilizes a distance metric to quantify the similarity between two points in the
sample space. A smaller distance indicates a higher degree of similarity, while a larger distance
indicates a lower degree of similarity. Three distances that are frequently employed in many fields of
study are the Manhattan distance, the Euclidean distance, and the Minkowski distance (See Figure 3).
Figure 3: KNN model.
First, the data set is separated from the feature variables (“ High,” “Low,” “Open,” “Volume”) to
X and the target variable (“ Close “) to y. A TimeSeriesSplit object is then defined, the feature
variables in the training and test sets are standardized using StandardScaler, and two splits perform
time series cross-validation. Hyperparameter tuning is performed using cross-validation, with the
number of neighbors (K) as the hyperparameter, and the errors (mean square errors) for different K
values are plotted to help find the best K value. The selection of the K value in the KNN algorithm,
being the sole one-bit hyperparameter, significantly influences the ultimate prediction outcome of the
algorithm in an intuitive and consequential manner. After that, GridSearchCV is used to find the best
K value from the range of values (1,150). This K value is then used to create the best_knn model and
fit the best_knn model on the scaled training data. The optimal solution prediction function is used to
predict our scaling data. The best_knn model was used to plot the test data’s actual and predicted
closing prices. Calculating the RMSE is part of the performance evaluation of a model.
3.2. LSTM
LSTM is a recurrent neural network (RNN) model. RNNs are an artificial neural network type
expressly designed to process sequential data. Particularly effective in occupations requiring the
manipulation of sequences, such as time series data, voice, and natural language, as well as related
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231064
72
tasks [7]. It is specifically designed to capture long-term dependencies in sequential data. They can
remember patterns and relationships in data over long periods, which is crucial for understanding
complex stock price trends. Stock price data typically exhibit variable time lags between different
events and movements. LSTM can naturally handle these irregular time intervals and capture
dependencies on different time scales. LSTM is widely regarded by professionals as the algorithm
with the highest potential for stock prediction. The system possesses the capability to handle both
discrete data points and intricate data sequences, rendering it suitable for analyzing non-linear time
series data and forecasting significant price movements characterized by high volatility [8].
Creating a sequential model that permits the linear stacking of neural network layers is the first
stage in developing an LSTM model. Then, three LSTM layers are added to the Sequential model.
Each LSTM layer contains 50 neurons. The output of the first two LSTM layers is a suitable sequence
for the subsequent LSTM layer. The output of the third layer LSTM is a single value suitable for
input to the next layer. After each LSTM layer, add the Dropout layer with a Dropout rate of 0.2 to
prevent the model from overfitting. Using the relu activation function, add a dense layer, also known
as the entirely connected layer with 1 neuron. The dense layer serves as the model’s output layer and
outputs the anticipated stock price. After the model’s structure has been defined, the model is
compiled. The loss function was selected as Mean Squared Error, the optimizer as Adam, and the
evaluation index as Mean Absolute Percentage Error. Finally, the model is trained by fitting method,
training data and target are input, validation data is set, 10 training epochs are set, batch size is 32,
and training process is printed (See Figure 4).
Figure 4: LSTM model structure.
3.3. RMSE
RMSE is widely utilized as an evaluation metric for regression tasks because to its ability to assess
the proximity between predictions and actual values on average, while also providing insights into
the impact of significant errors. Significant inaccuracies will affect the outcome of RMSE [9]. When
applied to predicting stocks, RMSE can be used to quantify the degree to which a forecasting model’s
forecast matches a stock’s actual price movement in a specific time period. The performance of its
forecasting model can be evaluated by comparing the predicted stock price with the actual price using
RMSE. A large RMSE shows that the model’s predictions and the real values are very different from
each other. When the RMSE is small, on the other hand, the correlation between the price predicted
by the model and the real price is stronger, which means the model is more accurate. If the model has
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231064
73
an extremely low RMSE on the training data, the model may perform very well. Overfitted models
demonstrate high performance when evaluated on the training data but exhibit poor performance
when tested on fresh or unseen data. Therefore, the constructed model lacks practical utility [10].
RMSE is calculated as follows.
𝑅𝑀𝑆𝐸 =&!"#$%&#'!()*+,-$%+,!."/0/
#
$%! "#$%&#'#/)*+,-$%+,#."
1 (1)
4. Results
4.1. KNN
Using cross-validation to perform hyperparameter tuning, taking the number of neighbors (K) as the
hyperparameter, the errors (mean square errors) of different K values are plotted. GridSearchCV is
used to find the best K value of AAPL is 52, GOOG is 51, and AMZN is 60 from the range of values
(1,150). The optimal K value of MSFT is 147. Then, the best_knn model is created using these K
values, and the best_knn model is fitted on the scaled training data and the actual and predicted closing
prices of the test data are plotted using the best_knn model. Lastly, the efficacy of the model is
evaluated by calculating the RMSE. Based on this KNN model, the RMSE of AAPL is 21.2507, the
RMSE of GOOG is 5.4037, the RMSE of AMZN is 6.3037, and the RMSE of MSFT is 44.6457
(Details are shown in the following Figures 5-8).
Figure 5: KNN is used to predict the price of AAPL stock.
Figure 6: KNN is used to predict the price of AMZN stock.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231064
74
Figure 7: KNN is used to predict the price of GOOG stock.
Figure 8: KNN is used to predict the price of MSFT stock.
4.2. LSTM
A sequential model is created that allows for linear stacking of neural network layers. Add three
LSTM layers to the Sequential model. Each LSTM layer contains 50 neurons. The output of the first
two LSTM layers is a sequence suitable for input into the next LSTM layer. The output of the third
layer LSTM is a solitary value that is suitable for use as input to the subsequent layer. After each
LSTM layer, add the Dropout layer with a Dropout rate of 0.2 to prevent the model from overfitting.
Then, add a dense layer, which has one neuron. The dense layer serves as the final layer of the model
and generates the forecasted stock price. Once the structure of the model has been established, the
model undergoes the process of compilation. The loss function was selected as Mean Squared Error,
the optimizer as Adam, and the evaluation index as Mean Absolute Percentage Error. Finally, the
model is trained by fitting method, training data and target are input, validation data is set, ten training
epochs are set, the batch size is 32, and the training process is printed. Lastly, the trained model is
then used to forecast the time series data, and the accuracy of the prediction is determined by
calculating the RMSE. The RMSE of AAPL is 9.7359 of GOOG, is 4.8965 of AMZN is 6.6956, and
of MSFT is 10.3862 (Details are shown in the following Figures 9-12).
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231064
75
Figure 9: LSTM is used to predict the price of AAPL stock.
Figure 10: LSTM is used to predict the price of GOOG stock.
Figure 11: LSTM is used to predict the price of AMZN stock.
Figure 12: LSTM is used to predict the price of MSFT stock.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231064
76
4.3. RMSE
As demonstrated in the Table 1 below, the RMSE of the LSTM model is very stable because it
predicts AAPL, GOOG, and MSFT with fewer errors than the KNN model. However, the KNN model
is only marginally more accurate than the LSTM model when predicting AMZN price.
Table 1.
Table 1: RMSE of KNN and LSTM for the four assets.
RMSE
KNN
LSTM
AAPL
21.2507
9.7359
GOOG
5.4037
4.8965
AMZN
6.3037
6.6956
MSFT
44.6457
10.3862
5. Conclusion
Stock prediction plays a very important role for stock traders. Based on the actual stock data of AAPL,
GOOG, AMZN, and MSFT from 2013 to 2023, this paper uses KNN and LSTM models to predict
the exact closing price of stocks in the future, which can be used to further evaluate the rise and fall
of stocks. The prediction effect is good, and all four companies show an upward trend. Upon doing a
comparative analysis between the two models, it has been observed that the LSTM model exhibits a
smaller and more consistent error in comparison to the KNN model. This paper argues that LSTM
can capture the crucial time dependencies and relationships in financial time series data. Financial
markets are influenced by historical trends, news events, and other time-sensitive factors that KNN
may have difficulty capturing effectively, which makes the LSTM model very much a stock price
forecasting model.
References
[1] Rishi, T. (2022). Stock Market Analysis Using Linear Regression. In Proceedings of the Jepson Undergraduate
Conference on International Economics, 4.
[2] Loukas, S. (2020). Time-Series Forecasting: Predicting Stock Prices Using An LSTM Model. Retrieved from
https://towardsdatascience. com/lstm-time-series-forecasting-predicting-stock-prices-using-an-lstm-model-
6223e9644a2f.
[3] Sadorsky, P. (2021). A random forests approach to predicting clean energy stock prices. Journal of Risk and
Financial Management, 14(2), 48.
[4] Li, R., Ma, M., and Tang, N. (2023). Stock Price Prediction Based on Decision Trees, CNN and LSTM. In
Proceedings of the 4th International Conference on Economic Management and Model Engineering, ICEMME
2022, November 18-20, 2022, Nanjing, China.
[5] Nti, I. K., Adekoya, A. F., and Weyori, B. A. (2020). Efficient Stock-Market Prediction using ensemble support
Vector machine. Open Computer Science, 10(1), 153–163.
[6] Sonkavde, G., Dharrao, D. S., Bongale, A. M., Deokate, S. T., Doreswamy, D., and Bhat, S. K. (2023). Forecasting
Stock Market Prices Using Machine Learning and Deep Learning Models: A Systematic Review, Performance
Analysis and Discussion of Implications. International Journal of Financial Studies, 11(3), 94.
[7] Biswal, A. (2023). Power of Recurrent Neural Networks (RNN): revolutionizing AI. Retrieved from
https://www.simplilearn.com/tutorials/deep-learning-tutorial/rnn
[8] Gülen, K. (2023). Stock Prediction In Machine Learning Explained - Dataconomy. Retrieved from
https://dataconomy.com/2023/01/11/stock-prediction-machine-learning/
[9] Ogunbiyi, I. A. (2022). Top evaluation Metrics for regression problems in Machine Learning. Retrieved from
https://www.freecodecamp.org/news/evaluation-metrics-for-regression-problems-machine-learning/
[10] Shah, R. (2022). Performance comparison of regularized and unregularized regression models. Retrieved from
https://www.analyticsvidhya.com/blog/2021/08/performance-comparision-of-regularized-and-unregularized-
regression-models/
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231064
77
The Impact of Corporate Social Responsibility on Corporate
Financial Performance
--An Empirical Study of Chinese Energy Firms
Zhuoqun Xu1, a, *
1College of LSA, University of Michigan, Ann Arbor, the United States of America
a. zhuoqunx@umich.edu
*corresponding author
Abstract: As the impacts of environmental pollution and climate change on humanity have
become increasingly severe, achieving sustainable development has heightened global
concerns in recent years. An increasing number of companies have recognized the importance
of corporate social responsibility (CSR) and are actively investing in enhancing their
performance. Moreover, an increasing number of investors are considering social factors
when making investment decisions. Although the potential impact of CSR on corporate
financial performance (CFP) is a growing interest in related research filed, there still needs
to be consensus on their relationship. This study seeks to address this gap in the context of
the Chinese energy sector. By conducting fixed effect panel regression on the CSR scores,
accounting financial indicators, and fundamental panel data of 876 listed companies in the
Chinese energy sector over the past decade, this paper reveals a nonlinear positive impact of
CSR on CFP. Further threshold regression is applied to identify distinct threshold values that
affect different financial indicators. The findings of this paper complement the existing
related research and provide valuable support and guidance for companies in the energy
sector to undertake social responsibility and engage in social responsibility investments
effectively.
Keywords: corporate social responsibility, corporate financial performance, Chinese energy
industry, panel analysis
1. Introduction
Global attention to sustainable development has grown fast in recent years. Governments and global
organizations all over the world are collaborating to develop strategies for a greener world. When it
comes to the action of international institutions, the United Nations approved the 2030 Agenda for
Sustainable Development in September 2015, and the Paris Agreement also aims to control global
warming to less than 2oC. As the world's second-largest economic unit, China plays an essential role
in contributing to global sustainable development.
With the link between the concept of sustainability and CSR, enterprises play a significant role in
achieving sustainable development goals as the main microeconomic. Their stakeholders' needs and
desires influence businesses' motivations and goals. To achieve profitability and growth, companies
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231065
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
78
must prioritize their stakeholders' social acceptance and involvement. The impact of CSR on CFP
sets a company's attitude toward and level of investment in social responsibility activities. Clarifying
the relationship between CSR and CFP is thus critical for corporate strategic decision-making and
promoting corporate and national social responsibility efforts.
Despite the fact that the existing literature has paid close attention to the relationship between CSR
and CFP, they have not yet reached a consensus on whether social responsibility can enhance
corporate financial performance or not. Furthermore, empirical research on the relationship between
CSR and the financial performance of Chinese companies, particularly in specific industries, still
needs to be conducted. This paper investigates the relationship between CSR and CFP by addressing
the relationship and threshold effect between CSR and CFP using Chinese-listed companies in the
energy sector as the research sample.
2. Literature Review
2.1. Corporate Social Responsibility Concept
CSR was first proposed in the nineteenth century by American economist Howard Bowen in his book
“Social Responsibilities of the Businessman” in 1953 [1]. It is defined as the responsibility that
corporations must accept in order to profit while also maintaining and increasing the wealth of society
as a whole [2]. Not limited to beneficial, Drucker [3] claimed that CSR does not mean doing good
things or not doing harm, but the conversion from social problems into profit opportunities for
corporations. As one of the important international institutions, World Business Council For
Sustainable Development [4] think company shows their CSR by acting ethically with an ongoing
commitment. After a long period of development and expansion, its concept has been widely
recognized and has become a successful management tool [5]. However, there is no one standard
model can be reached, and it is difficult to give a concept that can be accepted by everyone [6].
There have been a large number of publications about CSR over the last few decades, but they are
highly fragmented. On the one hand, this is due to the numerous research directions of CSR extension,
and on the other, to the high heterogeneity of CSR concept application subjects [7].
2.2. Relationship Between Corporate Social Responsibility and Corporate Financial
Performance
Plenty of studies on the relationship between CSR and CFP have been conducted in the past, but no
consensus has been reached. Griffin and Mahon [8] carried out a comprehensive review of 51 articles,
encompassing 62 studies over the past 25 years, and found that 33 of them showed a significant
positive correlation, 20 studies showed a significant negative correlation, and 9 studies yielded
inconclusive results. Similarly, Margolis and Walsh [9] summarized 127 related empirical studies
from 109 articles spanning 1972 to 2002 and found that almost half of them showed a positive
relationship, 7 showed a negative relationship, 28 did not give a strong relation, and the rest of 20 did
not show an obvious result. Besides the linear relationship that many studies and papers discussed,
the non-linear relationship between CSR and CFP also exists [10], [11].
The divergent findings in the literature can be attributed to various factors. One of them is about
the inappropriate measurement approaches due to the complexity of CSR theory [12]. Also, the direct
and indirect effects of CSR need to be disentangled [13]. Another possibility is the missing moderator
variable and mediator variable, which can help to capture the key complexity precisely [14].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231065
79
2.3. The Impact of Developing Countries and the Energy Sector
Local CSR takes many forms, and the difference is becoming more visible in developed as well as
developing nations [15]. However, China, as a prominent player among developing nations, has not
received adequate attention in this context. Besides the difference among countries, the fact that
different relationship results are shown by different sectors suggests the need of choosing a specific
focal one [16], [17]. Consequently, this study centers on the energy industry in China. The energy
sector holds particular significance as its contribution to global greenhouse gas emissions and its
importance of being the key to fight against environmental issues, according to IEA and IRENA [18].
The importance of the energy sector in decarbonization and achieving sustainability goals cannot be
underestimated. Therefore, this study focuses on publicly traded companies in the energy sector as a
means to comprehend the underlying dynamics and provide a model for other industries.
3. Methodology
3.1. Data Collection
This study uses CSR scores and selected financial indexes of listed companies in the Chinese energy
industry over the past 10 years. This study's CSR scores were obtained from Hexun.com, and needed
financial data was obtained from the CSMAR Database. Due to their abnormal financial status,
samples with missing data and companies with the marks ST, *ST, and PT were removed to ensure
the reliability and validity of the results.
3.2. Hypothesis
According to the literature review, the relationship between CSR and CFP may vary from study to
study. In fact, rather than being straightforwardly positive or negative, the relationship between CSR
and CFP may be nonlinear. On the one hand, excessive CSR commitment can lead to increased
marginal costs and decreased marginal benefits, resulting in a decline in CFP [19]. Companies that
do not actively invest in CSR activities, on the other hand, may gain an edge over their competitors
by reducing related costs, and potentially improving their CFP [20]. On this basis, this paper puts
forward the following hypothesis:
Hypothesis: The impact of CSR on CFP is nonlinear.
3.3. Model Setup
3.3.1. Panal Regression Model
To estimate how CSR influences CFP, this paper constructs the following panel regression model:
𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒!"# = 𝛽$+ 𝛽%𝐶𝑆𝑅!"# + 𝛽𝐶𝑆𝑅!"#
&+∑𝐶𝑜𝑛𝑡𝑟𝑜𝑙!"# + 𝜀!"# (1)
In the model, the independent variable Performance is CFP. The dependent variable is CSR. CSR2
is the quadratic term of CSR, which is set to capture the non-linear connection between CSR and CFP.
Control variables are those that are used to adjust for the possible influence of others. To address
potential time-lagged effects of financial indicators, a fixed-effect panel regression model including
lagged variables is used in the analysis.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231065
80
3.3.2. Threshold Effect Test
This study employs a dynamic panel threshold model inspired by Hansen [21] to investigate the
possible threshold effect of CSR on CFP. The model is configured as follows:
𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒!"# = 𝑈!+ 𝛼%𝐶𝑆𝑅!"#𝐼7𝐶𝑆𝑅!"# ≤ 𝛾: + 𝛼&𝐶𝑆𝑅!"#𝐼7𝐶𝑆𝑅!"# > 𝛾: + 𝛼'∑𝐶𝑜𝑛𝑡𝑟𝑜𝑙!"# + 𝜀!"#
(2)
Where i indicates the firm and t indicates the year. γ establishes the scale of CSR where the impact
on CFP can change, and I(*) is the indicator function. When 𝐶𝑆𝑅!"# ≤ 𝛾 , I(*)=1, and I(*)=0
otherwise.
The threshold test is done by the following steps:
1) The null hypothesis below is proposed to test the existence of the threshold effect:
H0: α% = α& (3)
2) The bootstrap method is employed to model the progressive distribution of the likelihood ratio,
which is calculated using a formula as follows; for each threshold, 300 bootstrap simulations are
performed:
𝐹 = ()!*)"+
,
-# (4)
In this formula, S0 is the sum of squares of residuals of the linear model without threshold effect,
and S1 is the sum of squares of residuals of the estimated model with the single threshold effect.
3) calculatetheAestimateAofAthresholdAvalueA𝛾 by minimizing the sum squares of residuals:
𝛾M = 𝑎𝑟𝑔.𝑚𝑖𝑛𝑆%(𝛾) (5)
4) Calculate the LR ratio, which is calculated as below, is a statistic to form γ’s confidence intervals
and the “no-rejection region”:
𝐿𝑅%(𝛾)=()"(.+*)"(.
/++
,
-# (6)
5) With a given confidence level 𝛼, the null rejection cannot be rejected when:
𝐿𝑅(𝛾)> −2 𝑙𝑜𝑔(1 − 1 − 𝛼) (7)
3.4. Meaning of the Variables
3.4.1. Dependent Variable: Corporate Financial Performance
Marketing measures, accounting measures, and perceptual measures are mainly used as financial
performance measures in the previous literature [22]. McWilliams and Siegel [23] claimed that
accounting measures can reflect the effectiveness of the corporate internal process for decision-
making and the performance of the managers. And when it comes to the goal of figuring out the
relationship, using accounting measures is better than an index based on the market [24]. Therefore,
this study adapted ROA and ROIC as measures of financial performance.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231065
81
3.4.2. Independent Variable: Corporate Social Responsibility
The CSR scores utilized in this study were obtained from the Hexun database. Hexun.com has
developed a comprehensive system for assessing the CSR of listed companies. The evaluation system
considers five key aspects: shareholder responsibility, supplier, customer, and consumer
responsibility, environmental responsibility, employee responsibility, and social responsibility.
3.4.3. Control Variables
Other factors may influence the relationship between CSR and CFP, so control variables are required
[25]. This study used firm size and leverage, as in previous papers [26-28].
Size: The firm’s size is calculated by the natural logarithm (LN) of total assets. Both excessively
large and small sizes, according to Marshall's theory of economies of scale, are detrimental to business
development.
Leverage: The percentage of indebtedness is used to measure leverage, which is the proportion of
total debt to total assets specifically. The likelihood of financial distress can be captured by the
leverage [28].
4. Results
4.1. Descriptive Statistics and Correlation Analysis
The descriptive statistics for the variables in this study are shown in Table 1. For the index of CFP,
the mean value of ROA and ROIC for Chinese listed energy firms are 1.89% and 1.92%, respectively.
The standard deviations of them are 2.45 and 3.39, indicating relatively significant fluctuation in the
sample data with regard to financial performance. The average value of the firms’ CSR scores is
26.29, with a standard deviation of 17.72, which also demonstrates that there is considerable variation
in CSR performance among sampled firms.
Table 1: Descriptive statistics of key variables.
Statistics
N
Mean
St.Dev
Min
Max
ROA
957
1.89
2.45
-6.75
17.72
ROIC
957
1.92
3.39
-18.74
63.78
Size
957
22.98
1.59
16.85
26.82
LEV
957
56.62
26.79
1.28
437.24
CSR
957
26.29
17.72
-11.30
79.96
Shareholder
957
13.35
6.55
-10.43
26.72
Employee
957
2.57
2.92
0.00
15.00
SCC
957
2.40
5.8
0.00
20.00
Environment
957
2.04
5.02
0.00
30.00
Social
957
5.93
4.49
-15
18.50
Note: The variables are: ROA=return on assets, ROIC=return on investment capital, Size= LN (Total Assets), LEV=leverage of the
firm, CSR=CSR total scores, Shareholder=shareholder responsibility scores, Employee=employee responsibility scores,
SCC=suppliers, customers, and consumers responsibility scores, Environment=environment responsibility scores, Social=social
responsibility scores.
Table 2 displays the correlations between variables. ROA and ROIC have a strong correlation
because they both represent the company's financial situation. The different components of the CSR
score also show a strong correlation among themselves, but since we only include the total CSR
scores in the model, the specific correlations between its constituent factors will not impact the
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231065
82
robustness of the model. Notably, the correlations between critical dependent and independent
variables are relatively low, indicating that there is no significant collinearity.
Table 2: Correlations between variables.
Size
LEV
ROA
CSR
Shrhlder
Emply
SCC
Env
Sci
ROIC
Size
1
LEV
0.196
1
ROA
-0.054
-0.214
1
CSR
0.199
-0.049
0.223
1
Shrhlder
0.276
0.192
0.370
0.591
1
Emply
0.036
0.029
0.015
0.784
0.140
1
SCC
0.094
0.026
0.069
0.861
0.201
0.843
1
Env
0.066
0.032
0.033
0.815
0.120
0.898
0.908
1
Sci
0.164
-0.002
0.204
0.550
0.388
0.145
0.250
0.165
1
ROIC
-0.091
0.027
0.783
0.156
0.266
0.020
0.043
0.016
0.143
1
Note: ROA=return on assets, ROIC=return on investment capital, Size= LN (Total Assets), LEV=leverage of the firm, CSR=CSR
total scores, Shrhld=shareholder responsibility scores, Emply=employee responsibility scores, SCC=suppliers, customers, and
consumers responsibility scores, Env=environment responsibility scores, Sci=social responsibility scores.
4.2. Relationship Between CSR and CFP
Table 3, which reports the results of the fixed effect panel data regression, summarizes the main
findings of this paper. The empirical analysis result shows that, under the control of individual and
time effects, the regression coefficients of the independent variable, which is CSR, are 0.070 and
0.116, which are statistically significant given the 1% level. Similarly, the coefficients of the squared
term (CSR2) are both -0.001 and are significant at the 1% level as well.
Based on the short review of the regression result above, key findings emerge that the influence of
CSR on CFP is nonlinear, thereby supporting the preliminary hypothesis 1 Furthermore, based on the
fitted curves and the sample distribution range, it is discovered that the non-linear relationship
between CSR and CFP follows an “inverted U-shaped curve”, which is supported via dynamic panel
threshold model later.
Table 3: Correlations between variables.
Dependent Variable
ROA
ROIC
CSR
0.070***
(0.011)
0.116***
(0.015)
CSR2
-0.001***
(0.0001)
-0.001***
(0.0002)
Size
0.152*
(0.091)
-0.049
(0.127)
LEV
-1.449***
(0.235)
1.705***
(0.330)
Lagged_ROA
0.073***
(0.027)
Lagged_ROIC
8.824***
(2.105)
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231065
83
Table 3: (continued).
Observations
876
876
R2
0.137
0.126
Adjusted R2
-0.006
-0.019
F Statistic (df = 5; 751)
23.837***
21.565***
Note: The variables are: ROA=return on assets, ROIC=return on investment capital, Size= LN (Total Assets), LEV=leverage of the
firm, CSR=CSR total scores, Lagged_ROA=ROA in lagged time period, Lagged_ROIC=ROIC in lagged time period. *p<0.1,
**p<0.05, ***p<0.01
4.3. Threshold Effect Analysis
This study ran a series of threshold tests to see if there were any threshold effects in the relationship
between CSR and CFP. A single threshold test was performed firstly to detect the existence of
threshold effect, and continuous examinations about double and triple threshold tests are made to find
out more details. Figure 1 represents the LR ratio graph, and the threshold test results are shown in
Table 4. For the model with ROA as the dependent variable, both the p-values of its first and second
thresholds are smaller than 0.05, indicating the existence of a second threshold effect. For the model
with ROIC as the dependent variable, only the p-value of its first threshold is smaller than 0.05, which
approves that the first threshold effect exists. As a result, while the two models with different
dependent variables show different levels of threshold effects, the analysis results can confirm the
presence of a nonlinearity relationship between the dependent and independent variable, which is a
segmented function divided by CSR performance.
Figure 1: LR ratio graph of ROA (left) and ROIC (right).
Table 4: Threshold effect test result.
Model
Dependent
Variable
Threshold Value
P-value
Single Threshold
ROA
34.580
0.000
ROIC
36.240
0.000
Double Threshold
ROA
[26.980, 34.580]
0.027
ROIC
[20.890, 36.240]
0.490
Triple Threshold
ROA
[22.710, 26.980, 34.580]
0.723
ROIC
[17.150, 20.890, 36.240]
0.747
020 40 60 80 100
LR Statistics
020 40 60 80
First Threshold
0 5 10 15 20
LR Statistics
020 40 60 80
Second Threshold
020 40 60
LR Statistics
020 40 60 80
First Threshold
0 2 4 6 8
LR Statistics
020 40 60 80
Second Threshold
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231065
84
Table 5 shows the results of threshold effect regression, where includes 𝛾% and 𝛾& are the
threshold values. When the total CSR scores are below 𝛾%, the coefficients estimate of CSR are
0.0054 and 0.0925, which means that for every 1% improvement in CSR performance, the CFP
performance (ROA and ROIC) is going to increase by 0.54% and 9.25% respectively, on average.
When the CSR scores are between 𝛾% andA𝛾&, a 1% increase in CSR scores can lead to a 7.42% and
2.89% increase in ROA and ROIC, on average. For CSR scores that are greater than 𝛾&, with every
1% increase in scores, ROA is going to increase by 2.32%.
The study has confirmed that CSR has a positive influence on CFP, but the positive relationship
is not linearly constant. For ROA, this influence first increases and then decreases; while for ROIC,
it decreases with the improvement of CSR. The above empirical findings support the notion that the
effect of CSR on CFP is non-linear, further validating the original research hypothesis.
Table 5: Threshold effect of CSR on CFP.
Variable
ROA
ROIC
𝛾%
26.980
36.340
𝛾&
34.580
-
CSR (csr<𝛾%)
0.0054***
(0.009)
0.0925***
(0.026)
CSR(𝛾%<csr<𝛾&)
0.0742***
(0.009)
0.0289***
(0.009)
CSR (csr>𝛾&)
0.0232***
(0.004)
-
N
657
657
R2
0.2097
0.1507
Note: *p<0.1, **p<0.05, ***p<0.01
5. Discussion
This study found a relationship between corporate social responsibility (CSR) and corporate financial
performance (CFP) in listed Chinese energy companies. In line with the hypothesis, the findings
demonstrate that CSR can impact CFP positively and nonlinearly, with the existence of a CSR
threshold that leads to varying degrees of positive impact on CFP. This research not only
complements the existing research on the impact of CSR on CFP in China but also provides
theoretical support and guidance for companies, especially those in the energy sector, to proactively
undertake social responsibilities. It should be noted, however, that due to the source of data limitations,
this study concentrates solely on the research of publicly traded companies in China's energy sector.
As a result, there may be a lack of explanatory power for those smaller and less financially robust
enterprises. Additionally, since the goal of this paper is to verify the non-linear relationship between
CSR and CFP, the aspects of exploring the threshold effects manifested by different financial
indicators, and the different threshold effects manifested by different financial indicators are not
explained. This leaves room for future investigation.
6. Conclusions
The purpose of this study is to investigate the relationship between CSR and CFP among listed
Chinese energy companies. By utilizing fixed effect panel data regression and threshold tests on CSR
scores and typical financial indicators, it can be discovered that CSR can affect CFP positively and
nonlinearly, which is in accordance with the research hypothesis. Also, the threshold test approves
that the presence of a CSR threshold causes variations in the degree of this positive impact on CFP.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231065
85
This study offers theoretical insights and guidance to companies by supplementing previous research
on the impact of CSR on CFP in China. Yet the limitations of sample choice need to be noted and a
more general conclusion and threshold effect across industries with a focus on different financial
measures are needed to be researched by those practitioners further.
Corporate social responsibility (CSR) has grown into an increasingly popular subject across
various sectors in recent years. Many companies use CSR to improve their reputation and
competitiveness and thus leading to better financial performance. Nonetheless, the perceived
additional costs and potential lack of short-term benefits make CSR activities appear less profitable
in the immediate future. The review of previous literature does not show a unanimous conclusion
about the causal effect of CSR on CFP, particularly when taking Chinese energy sectors as research
samples. This study goes beyond prior research by proposing a non-fixed connection between CSR
and CFP, highlighting the potential of an optimal CSR threshold.
References
[1] Howard Rothmann Bowen, Peter Geoffrey Bowen, & Jean-Pascal Gond. (2013). Social responsibilities of the
businessman. University Of Iowa Press. (Original work published 1953)
[2] Davis, K., & Blomstrom, R. L. (1975). Business and society : environment and responsibility. Mcgraw-Hill.
[3] Drucker, P. F. (1984). Converting Social Problems into Business Opportunities: The New Meaning of Corporate
Social Responsibility. California Management Review, 26(2), 53–63. https://doi.org/10.2307/41165066
[4] WBCSD.(2000).World Business Council For Sustainable Development (WBCSD). Wbcsd.org.
https://www.wbcsd.org/
[5] Isaksson, L., Kiessling, T., & Harvey, M. (2014). Corporate social responsibility: Why bother? Organizational
Dynamics, 43(1), 64–72. https://doi.org/10.1016/j.orgdyn.2013.10.008
[6] Carroll, A. B. (2000). Ethical Challenges for Business in the New Millennium: Corporate Social Responsibility and
Models of Management Morality. Business Ethics Quarterly, 10(1), 33–42. https://doi.org/10.2307/3857692
[7] Belas, J., Škare, M., Gavurova, B., Dvorsky, J., & Kotaskova, A. (2022). The impact of ethical and CSR factors on
engineers’ attitudes towards SMEs sustainability. Journal of Business Research, 149, 589–598.
https://doi.org/10.1016/j.jbusres.2022.05.056
[8] Griffin, J. J., & Mahon, J. F. (1997). The Corporate Social Performance and Corporate Financial Performance
Debate. Business & Society, 36(1), 5–31. https://doi.org/10.1177/000765039703600102
[9] Margolis, J. D., & Walsh, J. P. (2003). Misery Loves Companies: Rethinking Social Initiatives by Business.
Administrative Science Quarterly, 48(2), 268–305. https://doi.org/10.2307/3556659
[10] Barnett, M. L., & Salomon, R. M. (2012). Does it pay to be really good? Addressing the shape of the relationship
between social and financial performance. Strategic Management Journal, 33(11), 1304–1320.
[11] Nollet, J., Filis, G., & Mitrokostas, E. (2016). Corporate social responsibility and financial performance: A non-
linear and disaggregated approach. Economic Modelling, 52, 400–407.
https://doi.org/10.1016/j.econmod.2015.09.019
[12] Galant, A., & Cadez, S. (2017). Corporate social responsibility and financial performance relationship: a review
of measurement approaches. Economic Research-Ekonomska Istraživanja, 30(1), 676–693.
[13] Lioui, A., & Sharma, Z. (2012). Environmental corporate social responsibility and financial performance:
Disentangling direct and indirect effects. Ecological Economics, 78, 100–111.
https://doi.org/10.1016/j.ecolecon.2012.04.004
[14] Namazi, M., & Namazi, N.-R. (2016). Conceptual Analysis of Moderator and Mediator Variables in Business
Research. Procedia Economics and Finance, 36, 540–554. https://doi.org/10.1016/s2212-5671(16)30064-8
[15] Jamali, D., & Karam, C. (2016). Corporate Social Responsibility in Developing Countries as an Emerging Field of
Study. International Journal of Management Reviews, 20(1), 32–61. https://doi.org/10.1111/ijmr.12112
[16] Baird, P. L., Geylani, P. C., & Roberts, J. A. (2011). Corporate Social and Financial Performance Re-Examined:
Industry Effects in a Linear Mixed Model Analysis. Journal of Business Ethics, 109(3), 367–388.
https://doi.org/10.1007/s10551-011-1135-z
[17] Lin, C.-S., Chang, R.-Y., & Dang, V. (2015). An Integrated Model to Explain How Corporate Social Responsibility
Affects Corporate Financial Performance. Sustainability, 7(7), 8292–8311. https://doi.org/10.3390/su7078292
[18] IEA, & IRENA. (2017). PERSPECTIVES FOR THE ENERGY TRANSITION Investment Needs for a Low-Carbon
Energy System. International Renewable Energy Agency.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231065
86
[19] Adegbite, E., Guney, Y., Kwabi, F., & Tahir, S. (2018). Financial and corporate social performance in the UK listed
firms: the relevance of non-linearity and lag effects. Review of Quantitative Finance and Accounting, 52(1), 105–
158. https://doi.org/10.1007/s11156-018-0705-x
[20] Bhattacharya, C. B., & Sen, S. (2004). Doing Better at Doing Good: When, Why, and How Consumers Respond to
Corporate Social Initiatives. California Management Review, 47(1), 9–24. https://doi.org/10.2307/41166284
[21] Hansen, B. E. (1999). Threshold effects in non-dynamic panels: Estimation, testing, and inference. Journal of
Econometrics, 93(2), 345–368. https://doi.org/10.1016/s0304-4076(99)00025-1
[22] Orlitzky, M., & Benjamin, J. D. (2001). Corporate Social Performance and Firm Risk: A Meta-Analytic Review.
Business & Society, 40(4), 369–396. https://doi.org/10.1177/000765030104000402
[23] McWilliams, A., & Siegel, D. (2000). Corporate social responsibility and financial performance: correlation or
misspecification? Strategic Management Journal, 21(5), 603–609. https://doi.org/10.1002/(sici)1097-
0266(200005)21:5%3C603::aid-smj101%3E3.0.co;2-3
[24] Moore, G. (2001). Corporate Social and Financial Performance: An Investigation in the U.K. Supermarket Industry.
Journal of Business Ethics, 34(3/4), 299–315. https://doi.org/10.1023/a:1012537016969
[25] Huang, K., Sim, N., & Zhao, H. (2020). Corporate social responsibility, corporate financial performance and the
confounding effects of economic fluctuations: A meta-analysis. International Review of Financial Analysis, 70,
101504. https://doi.org/10.1016/j.irfa.2020.101504
[26] Hasan, I., Kobeissi, N., Liu, L., & Wang, H. (2016). Corporate Social Responsibility and Firm Financial
Performance: The Mediating Role of Productivity. SSRN Electronic Journal, 149(3).
https://doi.org/10.2139/ssrn.2768097
[27] Orazalin, N., & Mahmood, M. (2021). Toward sustainable development: Board characteristics, country governance
quality, and environmental performance. Business Strategy and the Environment, 30(8).
https://doi.org/10.1002/bse.2820
[28] Bose, S., Shams, S., Ali, M. J., & Mihret, D. (2021). COVID‐19 impact, sustainability performance and firm value:
international evidence. Accounting & Finance, 62(1). https://doi.org/10.1111/acfi.12801
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231065
87
Three Applications of the Anchoring Effect
Yu Liu1,a,*
1Beijing No.2 Middle School, Beijing, China, 100062
a. liu1101652005@163.com
*corresponding author
Abstract: To find out the relationship between psychology and economics, many people try
to study the reasons that lead people to make choices. This paper aims to explain the influence
of the “anchoring effect” in behavioural economics on consumption, investment, and food
intake. Through reading the previous literature review and the experiments, and based on the
definition of the anchoring effect in Thinking Fast and Slow by Daniel Kahneman, three real-
life applications of the anchoring effect were identified. In the anchoring effect in
consumption, investment, and food intake, the anchoring price is proportional to people’s
price estimates. Anchors can be divided into high anchor conditions and low anchor
conditions in those cases. However, possible biases in some of the experiments mentioned in
some of the citations and related solutions are also mentioned. In a complete decision-making
process, people’s choices will be influenced by multiple anchors, and those anchors are
always related to each other.
Keywords: anchoring, consumption, investment, food intake
1. Introduction
In recent years, with the in-depth study of behavioural economics, more and more people have studied
consumers’ choices. The study of the anchoring effect is mostly focused on psychology, economics,
law, and medicine. Because anchor information exists, decision-makers are more susceptible to bias
than in more standard marketing circumstances. Therefore, while creating sales tactics, entrepreneurs
frequently take the impact of anchoring into account [1]. The anchoring effect allows people to walk
into a store and measure whether the price of these items is fair. It allows people to decide how much
to eat when they see a serving of food. It is also used as a sales strategy to buy stocks and health
insurance. The influence of the anchoring effect in real life is ubiquitous. Through case studies, this
paper explains the effects of anchoring effect on consumption, investment and food intake in order to
demonstrate its wide application in daily life. As economist Daniel Kahneman defined in Thinking
Fast and Slow, the anchoring effect occurs when people consider a particular value for an unknown
quantity before estimating that quantity [2]. According to Tversky and Kahneman, Science, 1974,
different starting points yield different estimates, which are biased toward the initial values.
2. Consumption
The anchoring effect affects consumers’ choices. In the decision-making process, people are easily
influenced by previously provided information. In one experiment, students were shown an expensive
chocolate and asked if they would pay an amount equal to the last two digits of their ID card. If they
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231068
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
88
were willing to pay, they were divided into two groups with the last two digits of their ID less than
50 and more than 50, and they were asked to decide the maximum they were willing to pay for the
chocolate. The data showed that students with the last two ID numbers greater than 50 will pay an
average of R$25, and those with the last two ID numbers less than 50 will pay an average of R$16
[3]. The experiment shows that the last two digits of their ID are the anchor price, and the anchor
price is proportional to the amount of money people are willing to pay. This experiment has certain
limitations. Non-price factors in the above experiments have an impact on people’s decision-making.
Also, since this experiment was conducted in a university and the participants were all management
students of the university, self-selection bias, conviction bias, and observer bias are difficult to avoid.
The next study is based on the whole country, to avoid the bias of the above experiment as much as
possible, and also show that there is more than one factor (price) affecting the anchor in a transaction.
For instance, consumers’ first-hand information about the product itself will also affect their
consumption decisions. The purchase of organic versus non-organic food has long been debated.
According to data from the Certification and Accreditation Administration of China in 2014, organic
food is still in its initial stage of development in China and is not well known to the public. At the
same time, the price of machine food is usually 3-5 times that of ordinary food, and some even as
high as 8-10 times. Therefore, because people do not have enough knowledge about organic food,
they use the price of other commodities, such as non-organic food, as an anchor. When they find that
the price of organic food is higher than their anchor point, they will consider organic food to be a bad
deal, leading to less consumption [4]. Additionally, the anchoring effect is also often considered by
retailers as a marketing strategy. For example, in discounts, retailers always write the original price
on the label, because the original price will automatically become an anchor in the mind of consumers.
When consumers see the current price will be compared with the original price, and then they think
that they are the profit from the transaction. However, many retailers often raise the price of the
product first to turn the anchor into a high anchor to attract more consumers to spend. In addition,
purchase restriction is another principle that uses the anchoring effect to attract consumers. One
supermarket conducted such a survey, and the retailer put a “limit of 12 cans per person” sign on the
$0.79 can of Campbell Soup. When the sign was not on display, the average purchase per customer
was 3.3 cans per person, but when the limit sign was displayed. The average number purchased per
person jumped to 7 cans per person. It can be seen that 12 cans are used as anchors to stimulate
consumption [4]. The anchoring effect is also taken into account in sales promotion, especially for
luxury goods. For example, some merchants will always use $9.99 instead of $10 for their sales.
Because $9 is the anchor, in people’s minds, the price of the item is $9 instead of $10. According to
one study, this technique increased sales by an average of 24% [5].
3. Investment
Both individual and institutional investors are disturbed by anchoring prices. Investors often use the
historical performance of stocks to judge the future development of the stock market. Barberis,
Shleifer, and Vishny used the classic BSV model to explain that the past price and stability of a stock
will directly affect investors’ judgment on the future value of a stock, that is, when the stock price is
high and stable in the past, investors will use the past performance as an anchor and believe that the
stock will perform well in the future. They then choose to continue buying, causing the stock price to
continue to rise. When the stock price is low or unstable in the past, investors will also be pessimistic
about the future performance of the stock, reluctant to purchase the stock, or even want to sell the
stock, which will cause the stock price to fall further [6]. Therefore, the anchoring effect directly
affects the investors and indirectly affects the development of the stock market. The anchoring effect
not only affects the investment in the stock market but also affects the investment in medical insurance.
Nowadays, many people guard against low-probability disasters. People’s investment in health
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231068
89
insurance may be affected by related factors, such as risk perception, premium, and welfare. Therefore,
sales are usually made by asking buyers about the amount of insurance they are willing to pay and
offering a sum twice as high as the price given by the buyers as a claim fee [7]. In this case, people
use the price they are willing to pay as an anchor, so when the insurance company offers a higher
price as a claim fee, the buyers think that they will get more money when they encounter a serious
illness or disaster, so the buyers will think that this is a good deal. As more and more people regard
auctions as an investment, the role of anchoring effect in auctions cannot be ignored. “Anchors affect
the judgments of both individuals and groups, with three-member negotiating teams displaying biased
judgments similar to those of single negotiators” [8]. However, no matter how different conditions
people put forward in the auction and negotiation, the final judgment is often assimilated to the
previously considered anchor value. According to the research by Galinsky, Leonardelli, Okhuysen,
and Mussweiler in 2005, the results are always determined by the first price, and the correlation
coefficient is between 0.72 and 0.93 [9]. From this, it can be explained that in the negotiation, the
starting price of one party will affect the negotiation process of both parties, so the final negotiation
result will often fluctuate around 50% of the original anchor price. Therefore, the situation in the
negotiation will also happen in the auction, and the starting price will greatly affect the final
transaction price. Because the starting price is an anchor, consumers will make decisions based on
this anchor, and the fluctuation of the upper and lower prices will not be too great.
4. Food Intake
The amount of food served as an anchor can also influence people’s eating decisions. In an experiment,
in order to verify the influence of food quantity on people’s intake, the experimenter divided the
participants into three groups. The first group got a lot of food (high anchoring condition), and the
second group got little food (low anchoring condition). The third group as the control group did not
set the amount [8]. The results show that the anchoring effect has a great influence on people’s food
intake. Regardless of the types of food, the larger the amount given (high anchoring condition), the
more the subjects ate, and the smaller the amount given (low anchoring condition), the less they ate,
showing that the anchoring effect is proportional to people’s food intake. Low estimates of unknowns
are shown in another study. This tendency may explain why people eat more when there is more food
available. In this study, participants with a 2- pound box will eat 30% more than participants with a
1- pound box. Meanwhile, 23% more oil will be used from a 32-ounce bottle than from a 16-ounce
bottle [10]. These data also illustrate the direct effect of anchoring on decision-making, that is,
positive correlation.
5. Conclusion
The anchoring effect plays an indispensable role in the decision-making process. In terms of
consumption, this paper verifies that the level of the anchor is proportional to the amount of money
people are willing to pay. Also, the anchor is not only a consideration of the price of the commodity
itself but is also affected by its substitutes. In addition, some retail businesses will use the anchoring
effect to formulate some marketing policies, such as discounts, purchase restrictions, and sales
promotions. In the case of investment, multiple factors can be called anchors and have an impact on
people’s decisions. In the final food intake survey, experimental data showed that the size of the
anchor is directly proportional to people’s penetration. However, there are some limitations in this
paper. The experiments are common to find that the conclusion cannot be generalized due to the small
amount of data. Therefore, simple random sampling and Systematic sampling can be used to expand
the sample data to make the conclusion more universal. In addition, observer bias is inevitable.
Therefore, people can make the experimental results more accurate by inviting multiple observers to
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231068
90
participate in the experiment and double-blind experiment. The first part of the consumption section
of the chocolate experiment also has high results. Because the people excluded in the first part of the
experiment were also affected by the anchoring effect, people overpriced the chocolate. Therefore,
the first step can be deleted to increase the accuracy of the test results. In terms of content, the lack
of relevant economic models and calculations in this paper leads to insufficient proof content. In
addition, some literature data appeared earlier and may not be consistent with the current situation.
The most widespread application of the anchoring effect is to influence pricing and decision-making.
However, the influence of anchoring effect is ubiquitous, so it is possible to focus on the influence of
anchoring effect on other fields, such as law and medicine, in the future.
Acknowledgements
I sincerely thank my dissertation teacher; your selfless help and careful guidance played a vital role
in the process of writing the paper. Thank you for your patience and support during the whole writing
process. Your encouragement and trust make me firmly go on, full of enthusiasm and motivation for
research. Thank you for your guidance on the structure and language of the paper. Your professional
guidance has made my paper more accurate, clear and academically standardized, and I have learned
a lot of writing skills and methods from it. Finally, I would like to sincerely thank my dissertation
teacher for his selfless dedication and hard work. Your guidance and help have not only enabled me
to make great progress in thesis writing but also enabled me to grow into a scholar of independent
thinking and research.
References
[1] Zong, Yi, and Xiaojie Guo. “An Experimental Study on Anchoring Effect of Consumers’ Price Judgment Based on
Consumers’ Experiencing Scenes.” Frontiers, Frontiers, 7 Jan. 2022,
www.frontiersin.org/articles/10.3389/fpsyg.2022.794135/full.
[2] Kahneman, D(25 October 2011). Anchor. Thinking fast and slow.p.119.
[3] Neto,J.C., Filipe.J.A.,Ramalheiro. and B. “Neuromarketing: Consumers and the Anchoring Effect”.Int. J Latest
Trends Fin.Eco. Sc.Vol-1 No.4 December, 2011 Available at: https://repositorio.iscte-
iul.pt/bitstream/10071/14171/1/Neuromarketing%20Consumers%20and%20the%20Anchoring%20Effect.pdf
[4] Shan, L., Diao, H., & Wu, L. (2020). Influence of the framing effect, anchoring effect, and knowledge on consumers’
attitude and purchase intention of organic food. Frontiers in Psychology, 11. Available at:
https://doi.org/10.3389/fpsyg.2020.02022
[5] Kaur, G. (1 June 2022). 14 anchoring examples in real life. StudiousGuy. Available
at:https://studiousguy.com/anchoring-examples-in-real-life/
[6] Zhu, S. (2022). Anchoring effect in capital market. Highlights in Business, Economics and Management, 1, 66–70.
Available at: https://doi.org/10.54097/hbem.v1i.2320
[7] Metcalfe, R., & Dolan, P. (2012). Behavioural economics and its implications for transport. Journal of Transport
Geography, 24, 503–511. Available at: https://doi.org/10.1016/j.jtrangeo.2012.01.019
[8] Ku, G., Galinsky, A. D., & Murnighan, J. K. (2006). Starting low but ending high: A reversal of the anchoring effect
in auctions. Journal of Personality and Social Psychology,90(6), 975–986. Available at:
https://doi.org/10.1037/0022-3514.90.6.975
[9] Marchiori, D., et, al. “The portion size effect on food intake. An anchoring and adjustment process?”Appetite, 2014,
p.108. Available at: https://healthycognitionlab.org/wp-
content/uploads/2016/02/2014_Marchiori_Papies_Klein_Appetite_PSE_anchoring.pdf
[10] Roberto, C. A., & Kawachi, I. (2014). Use of psychology and behavioral economics to promote healthy eating.
American Journal of Preventive Medicine, 47(6), 832–837. Available at:
https://doi.org/10.1016/j.amepre.2014.08.002
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231068
91
A Case Study of Bright Food’s Acquisition of GNC in 2011
Guangmei Zhou1,a,*
1JiNan Foriegn Language School, JiNan, ShanDong Province, 25000, China
a. zhou20060614@126.com
*corresponding author
Abstract: With the development of the capital market, Mergers and Acquisitions has
become an important way to enhance market competitiveness. As the most popular dairy
industry in China, Bright Food tries to expand its scale through M&A with GNC
Corporation. The Bright Food uses their specific method and characteristics to M&A the
GNC. Although it ended in failure, we can still learn a lot from this event business. Based
on this situation, this article will use the case study method to analyze the theory behind
mergers and acquisitions. It will also focus on the reasons why Bright Food failed in M&A.
It will, from the financial angle and industry angle, conduct an in-depth analysis and put
forward the corresponding countermeasures and suggestions. The final result is that Bright
Food has failed on M&A at GNC. The main reasons that caused failure were poor financial
capacity and a slow speed to make a decision, as well as a lack of international recognition.
In order to countermeasure the problem that caused the failure of M&A, the GNC can
borrow from the bank to expand their financial capacity and do more advertising abroad to
increase their international recognition.
Keywords: bright food corporation, mergers and acquisition, GNC, financial ability
1. Introduction
Bright Food Group is a large state-owned comprehensive food industry group integrating modern
agriculture, food manufacturing, and commercial circulation. The group is committed to building
the whole food industry chain “from field to table”, and has many well-known brands such as
Guangming, White Rabbit, Guansheng Yuan, Merlin, Shikumen, and wine, as well as nearly 5,000
retail stores such as agricultural, industrial, and commercial stores and the first Food Store. Also,
M&A is important to a corporation because mergers and acquisitions can bring economies of scale
to enterprises. Mergers and acquisitions can bring market power to enterprises. Mergers and
acquisitions can reduce transaction costs for enterprises. Mergers and acquisitions can enable
enterprises to achieve diversified development at the lowest cost. The background of the topic of
why Bright Food had a failure in M&A with GNC has been thoroughly analyzed [1][2]. However,
it’s difficult to search online about analyzing the failure business deeply. The reason is that people
are more likely to focus on the case of Bright Food’s successful M&A with GNC. This article will
focus more on the reasons why M&A is a failure and the corresponding countermeasures and
suggestions. This part is seldom seen in the public article. This article will use the event study
method to conclude the result. The meaning behind this research is to help some merchants and
students who want to know about the basic reasons and knowledge behind this failure event. It will
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231076
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
92
bring a lot of benefits for the whole social cognition of the method of M&A and practice people’s
abilities in M&A.
2. Background
2.1. Background of M&A and GNC
As the main dairy industry in China, Bright Food has always enjoyed high national popularity and
international recognition [1]. As the famous economic theorist George said, ‘All of America’s big
companies have come through mergers and acquisitions in some way or another. There is hardly a
large business that has grown primarily through internal expansion.’ So in order to expand the scale
of corporations, Bright Food has merged with other famous companies in the world.
At that time, it’s a really tough thing to M&A overseas companies. But why do we say it is a
tough thing? The reason behind this is that China’s dairy industry seems less reliable than those
overseas. Some famous scandals, like the melamine incident’ and hormone gate, and even some
toxic milk powder events, were published and exposed by the news. These things really are
drawbacks of China’s dairy industry. Not only did the dairy industry itself suffer a setback, but the
whole reputation of China has also plummeted internationally. Even on some overseas market
goods shelves, it says china free. So it’s not easy to let the mainland corporations go abroad.
GNC was founded in 1935 by David Shakarian. Shakalien opened the first health food chain
called General Nutrition in Pittsburgh. GNC has grown into one of the world’s leading dietary
supplement brands, with operations in more than 50 countries and regions around the world. It is
now considered the world’s first health care brand and is exported in to several countries [2].
2.2. The Reason of Bright Food’s Acquisition of GNC
The Bright Food Group is a large domestic food group that has four listed companies, covering a
wide range of major industries and other industries, including the dairy industry, sugar industry,
wine industry, integrated food manufacturing industry, brand agency industry, chain retail industry,
modern agriculture, urban logistics, real estate, and tourism hotels. At the beginning of the
establishment of the group company, the business was overlapping and scattered, the quality was
uneven, the resulting management efficiency was questioned, and the criticism of “big but not
strong” was widely discussed. Further more, the bright food’s ROE is lower than YiLi dairy and
MengNiu dairy. The ROE (return on equity) refers to the ratio between the net profit and the net
assets of the listed company, reflecting the size of the profitability of the listed company [3]. In
order to compete with the other major dairy industries like MengNiu dairy and YiLi dairy in China,
Bright Food decided to expand their company in order to make a higher ROE and receive more
benefits. Why does Bright Food want to acquire GNC? According to the Pengmin, there are several
reasons that can support it. The first reason is that bright foods need retail consolidation. At that
point, what is the retail industry? How can we define the retail industry? According to the common
definition, retail is any individual or company engaged in the marketing of products from producer
to consumer, buying goods from wholesalers, middlemen, or manufacturers and selling them
directly to consumers. The benefit of consolidation is that it can really help a company reduce the
cost of buying goods from whole suppliers, middlemen,or products from producer to consumer.
Since 2010, with the development of China’s whole economy, salaries have risen, which means that
the Chinese living standard has also risen. People are more likely to pay attention to their own
health instead of just thinking about how to survive. Thus, the demand for some imported
health-care products has increased. As a result, the increase in demand for health-care products
means that the development of the health-care industry is rising. As the biggest health-care industry
in the world, the GC has a variety of health care products. The final reason why Bright Food wants
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231076
93
to M&A with GNC is because GNC has a unique supply base. As the biggest health-care industry in
the world, the GNC has an integrated product line and product machine, and the GNC has already
accumulated a lot of customer reliance. Thus, if Bright Food can M&A GNC successfully, it can
receive an amount of benefit directly instead of building the product line or hiring some
professional workers. So it’s very convenient to make a profit.
2.3. The Timeline of the Bright Food’s M&A’s
Table 1: The timeline of the bright food’s M&A [4].
Date
Acquired
Enterprise
Description
Amount
Stock
Right
Profession
Result
August
2010
CSR
It holds 45% of
Australia’s raw sugar
capacity
A $1.68
billion
N/A
Sugar
Industry
Lose
January
2011
GNC
Ranked first in the global
health products industry
$2.5 billion
N/A
Snack food
Lose
March
2011
Yoplait
The world’s second
largest yogurt maker
1.75 billion
euros
51%
Dairy
product
Lose
May
2012
Weetabix
The second largest
breakfast cereal brand
manufacturer in the UK
12.2 billion
RMB
60%
Snack food
Successful
June
2012
Diva
Foreign trade deals all
over the world, mainly
serving markets in
Europe and Asia
Markets in
Europe and
Asia
70%
Food and
beverage
Successful
January
2014
Mundella
foods
Australia 40 years of
history of dairy
enterprises
Products
are mainly
cheese,
fresh
yogurt
N/A
Dairy food
Successful
Table 1 is a timeline graph showing the bright food M&A since the 2010 year, and it can be seen
that there are several businesses that are failing: CSR, GNC, and Yoplait. It really shows that at that
time, it was so difficult to let us domestic companies go abroad.
3. The Process of the Acquisition
On December 7, 2010, foreign media reported that Bright Group was close to reaching a deal to buy
the United States vitamin retail chain Jian ‘an for $2.5 billion to $3 billion. This is the first
large-scale overseas merger and acquisition of domestic food enterprises. Then on January 26, Cao
Shumin, president of Bright Group, told this reporter, “The reason for not talking is that GNC is
ready to go public and fund companies. (GNC shareholders) want to exit all, and Bright Group
wants to control the fund. Prior to this, in 2010, Bright Group had negotiated the acquisition of
Australia’s CSR Company and United Biscuits Company in the UK. CSR Corporation United
Biscuits, which was seized by Wilmar International Group, was “abandoned” because of the
simultaneous acquisition of GNC. However, according to Cao Shumin, although there have been no
successful overseas mergers and acquisitions, Bright Group’s internationalisation will go out sooner
or later”.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231076
94
4. The Failure
Why Bright Food is a failure on the M&A of GNC? There are several reasons that can explain it.
But according to the public report, the major reason is that the GNC Worried that the approval time
of China’s mergers and acquisitions is too long and the pace of mergers and acquisitions is too slow,
they gave up cooperation with Bright and decided to exit by IPO. (An initial public offering is when
a business sells its shares to the public for the first time.) [5] However, except for this, I think there
are several other reasons that led to the failure of M&A at GNC.
4.1. Financial Stress
The GNC acquisition may have put Bright Food under significant financial stress. After an
acquisition, it may be necessary to pay high debt or balance sheets, which may negatively affect the
company’s financial condition [6].
Table 2: The main accounting data grafh [7].
Item
Amount
Operating profit
208621589
Gross profit
240962588
Net profit attributable to shareholders of listed companies
194375554
Net profit attributable to shareholders of listed companies
after deducting non-recurring gains and losses
166999179
Net cash flow from operating activities
534244918
Table 3: Whole year financial statement [7].
Main Accounting Data
2010
2009
Increase or decrease
(%) over the same
period last year
2008
Operating Income
9,572,111,030
7,943,169,7
11
20.51
7,358,544,3
48
Gross Profit
240,962,588
189,821,12
5
26.94
-319,543,91
1
Net profit attributable to
shareholders of listed
companies
194,375,554
122,470,12
8
58.71
-285,994,37
4
Companies after
deducting non-recurring
gains and losses
166,999,179
97,583,841
71.13
-420,331,95
7
Table 4: Analysis of financial position during the reporting period [7].
Item
2010
2009
Amount of increase
and decrease
Income tax expense
13267403
61265099
-78.38
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231076
95
Table 5: Analysis of normal financial asset gragh [7].
Item
2010
2009
Amount of increase and decrease
Long-term equity investment
10896132
23138226
-52.91
Tables 2–5 show Bright Food’s financial situation during 2010. In the first graph, it shows the
Bright Food Corporation’s whole-year financial situation in 2010, including some operating profits
and some gross profits. We can clearly see that there are hundreds of millions of dollars in profit
that the Bright Food Corporation earns. Although there are a lot of profits that Bright Food earns in
2010, according to Bright Food’s official financial situation in 2009, we can see that there are
several items whose amounts are less than in 2009. Like the income tax expense and the long-term
equity investment. These two items are less than 2009 in 2010. The income tax revenue means that
Bright Food pays back a large proportion of debt. However, the long-term equity investment is less
than in 2009 and 2010, which means that the investment of other corporations is decreasing. It may
be that during that time, the bright food overseas M&A is not so bright at that time. So many
companies choose to take back their investment. Thus, the bright food hasn’t enough money, also
called the financial ability to M&A at the GNC. So that’s the major reason why Bright Food failed
in M&A at GNC. Invest.
4.2. A Challenging Market Environment
Why it’s a challenging market environment? because GNC operates in the health and nutrition retail
industry, which faces changing consumer needs and fierce competition, making the industry
challenging. International manufacturers also have some challenges because GNC operates globally,
involving regulatory and cultural differences in different countries.
4.3. Limitations of Brand Recognition
Although GNC has a high brand recognition in the international market, its visibility in China is
relatively low. According to the research, China’s economy actually increased during the 2010 and
2011 years. However, there are many people who can’t maintain their normal lives, including
making sure they feel full and wearing warm clothes on cold days. Thus, it’s impossible to let many
people make sure their minds are clear and their bodies are healthy. Most people are unlikely to
recognise the brand GNC. As a result, Bright Food may have failed to effectively enhance GNC’s
brand awareness in the Chinese market, affecting its business expansion. In China the market
conditions may have changed after the acquisition, and there may have been problems that were not
anticipated. If Bright Food does not make timely market adjustments and strategic changes, it may
lead to a decline in performance. Bright Food may face the challenge of a lack of management
capacity after the acquisition of GNC, especially in handling international operations and
diversifying businesses [8][9]. Based on these reasons, that’s why Bright Food is a failure at M&A
GNC.
5. Discussion
Based on these reasons, we can conclude that there are some main reasons that will affect the M&A.
All in all, the first component that can affect mergers and acquisitions is the environment of the
whole market. If a company is in a competitive market, they need to have some special points that
can attract an M&A company. Financial ability will also be an important point in M&A. The speed
of the M&A is also very important. According to the important angles that can affect the M&A, the
author has some suggestions on it. For the environment of the merger, a good leader who leads the
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231076
96
company to make the decision and decides the specific details of the merger is useful to the merger.
For financial ability, a corporation can actually borrow some money from the bank if they don’t
have enough financial ability [10]. But the point forward is that they need to ensure the
development of the M&A corporation. If they can make sure the company’s development is raised.
They can borrow it for a certain amount of benefit.
6. Conclusion
As a result, the main reason that affects the bright food failure M&A on GNC is their too slow
M&A speed and indecisiveness in M&A decision-making. Also, according to the case study of
bright food failure M&A on GNC, it shows that several reasons that cause the M&A failure of GNC
are the inadequate financial capacity and lack of recognition in the international market. The GNC
can borrow from the bank to boost their financial capacity and increase their international
recognition in order to address the issue that led to the failure of M&A. There are still some
drawbacks in this article that need to be amended, like some specific data analysis and calculating
and using the cap to run the data. So future research will focus on calculating and analyzing the
data.
Acknowledgement
In this article, I want to show my deepest gratitude to my professor Sun, who has taught me and
provided me with tons o of professional knowledge and the method to analyse the case study. Then,
thanks to my assistant teacher, she always answers my questions with a patient disposition and
offers me the amount of material I need. At the end, I’d like to be thankful to my sister Carol, who
is older and more experienced than me. When I have some questions that the assistant teacher can’t
answer on time,I always ask Carol for help, and she always gives me some useful information.
References
[1] Wenming, Song. (2010). Guangming Dairy’s overseas acquisition of the entire industry chain model is similar to
that of COFCO. China Business Journal.
[2] About the introduce to the GNC. (2006.10/4). https://www.gnc.com/
[3] Xu, Zhang. (2010). the Bright Food M&A GNC, 017. https://www.21 CBN.com/
[4] The timeline of the bright food M&A. (2015.6/14). https://Bright/4004716
[5] Xia, Liu, (2023). A study on the spillover effect of issuance price-earnings ratio control — Based on the
perspective of IPO price suppression and premium. Journal of Shandong University of Finance and Economics
(04), 12-23.
[6] Shanshan Jiang. (2023). The role and limitations of financial statement analysis — a case study of Bright Dairy
Company. Bohai Rim Economic Outlook(6).
[7] The financial asset gragh of 2010. (2010). https://doc.rongdasoft.com
[8] Xinyu, He. (2022). Research on the Merger and Acquisition Integration of Bright Dairy Industry from the
Perspective of Value Chain (Master’s Thesis, Beijing Jiaotong University)
[9] Jian, Liu. (2022). Research on the Influencing Factors of Goodwill of Mergers and Acquisitions of Listed
Companies in China (Master’s thesis, University of Electronic Science and Technology).
[10] Yang, Han. (2021). Influencing factors and economic consequences of auditor selection in M&A transactions
(Doctoral dissertation, Central University of Finance and Economics).
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231076
97
An Analysis of the Applicability of Pricing Models in the
Chinese Stock Market
Tianshu Wang1,a,*
1Capital University of Economics and Business, International School of Economics and
Management, Beijing, 100070, China
a. terry_wang1016@163.com
*corresponding author
Abstract: The stock market in China has an important position in the world, but it has not
been able to find a pricing model suitable for the market. This paper reviews the history and
development of various pricing models, analyzes relevant research on their application in the
Chinese stock market, and discusses the applicability of various pricing models. It is found
that the applicability of the capital asset pricing model (CAPM) in China is low; the
applicability of the Fama-French three-factor model and the Fama-French five-factor model
is better than CAPM, but there is significant controversy over the advantages and
disadvantages of the two. It is also found that the applicability of the five-factor model may
have regional differences in different markets, and there may be differences in redundant
factors in different markets. It is proposed that future research should obtain more accurate
market data, explore new influencing factors, and build a pricing model more suitable for the
stock market in China by combining the characteristics based on more advanced data analysis
technology.
Keywords: CAPM, three-factor model, five-factor model, Chinese stock market
1. Introduction
In 1952, portfolio theory and the concept of risk were put forward for the first time, which laid the
foundation for modern investment theory [1]. By optimizing investment portfolios, it enables
investors to achieve the optimal balance between risk and return to a certain extent. In the following
decades, different pricing models have emerged to try to explain the pricing of assets and the return
on investment. In the 1960s, the capital asset pricing model based on the portfolio theory was
developed [2]. CAPM is a linear single-factor model composed of market risk factors. CAPM
establishes the relationship between expected return and risk and puts forward that return and risk are
consistent, which is also widely recognised in modern investment theory. In 1993, the Fama-French
three-factor model was put forward. On the basis of the CAPM model, market value factor and book-
to-market ratio factor were added as compensation for the factors that were not reflected in the market
risk factor in CAPM [3]. In 2015, they further expanded the three-factor model and put forward the
Fama-French five-factor model, which has become a relatively mainstream pricing model in the
world [4].
With the acceleration of China’s market opening to the outside world and the promotion of capital
market reform, pricing models widely used in the world have been gradually applied. Many scholars
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231077
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
98
have conducted empirical analyses of the applicability of various pricing models. Due to the
particularity of the financial market, the practice of pricing models in China may be different from
those in other markets. China implemented the new regulations on asset management in 2022, and
the income of wealth management products has been poor in the past year, so it is particularly
important to find a pricing model suitable for the market in China. Due to the unique characteristics
and challenges, the application of different pricing models is facing limitations and challenges, so it
is necessary and of practical significance to explore the applicability of various pricing models in
China. This paper summarizes the development history of pricing models, analyses the typical models
and their applicability in China, and looks forward to the future trend of pricing models so as to
provide readers with a more comprehensive perspective and way of thinking.
2. Historical Development of Pricing Model
2.1. Portfolio Theory
Markowitz portfolio theory links returns to risks. The expected return represents the return, the mean
represents the expected return rate, and the variance and covariance represent the risks of a single
asset and a portfolio respectively. Therefore, a mean-variance (covariance) model is proposed. Based
on the theory that investors are risk averse and prefer returns, Markowitz put forward Efficient
Frontier, a portfolio set that can satisfy two conditions: maximum expected returns and minimum
expected returns. Due to its idealistic assumptions, portfolio theory itself is greatly limited in practice,
and its assumptions of market efficiency are often not valid in real life. However, this theory is still
of great significance to modern investment theory.
2.2. CAPM
Markowitz put forward the concept of risk but did not quantify the relationship between risk and
return. Sharpe and Lintner put forward the CAPM. They believed that when the market reaches
equilibrium, risk determines the price of assets. The model considers that the expected return rate of
assets is directly proportional to the market risk premium. On the basis of Markowitz, Sharpe and
others introduced risk-free assets and recombined it with portfolios on the Efficient Frontier to form
the Capital Allocation Line, thus obtaining the optimal portfolio. The model contains a market risk
factor, and the relationship between factors is linear. The equation is as follows:
(1)
Where () represents the expected return rate of assets, is the risk-free rate of return, and
() is the expected rate of return of the market.
measures the systemic risk of assets, which represents the relationship between assets and
markets. The calculation equation of is as follows:
(2)
CAPM puts asset pricing theory into practical application, which has had a far-reaching impact on
finance, but it still has many limitations. CAPM model is based on many assumptions, but
assumptions are often difficult to hold in the real world. Secondly, the stock market cannot be
effectively explained by a single market risk factor. In the follow-up empirical research, the
effectiveness of CAPM has also been questioned. Some studies claim that CAPM cannot predict the
accurate expected rate of return. On the basis of CAPM, some scholars have also made extensions
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231077
99
and improvements. ICPAM and CCAPM proposed by Merton and Lucas are also extensions of
CAPM. Ross proposed Arbitrage Pricing Theory (APT) in 1976. APT needs fewer assumptions than
CAPM, and its fundamental assumption is that there is no arbitrage opportunity in the securities
market. It is proposed that asset prices are driven by multiple factors, rather than a single market risk
factor [5]. APT has been recognized by many scholars since it came out, and factor model has
gradually become the most significant research method in the asset pricing field.
2.3. Fama-French Three-factor Model
With the deepening of research, scholars found that β in CAPM cannot interpret the difference
between different stock returns, and β should not have a linear relationship with returns. Fama and
French selected the stock data from the NYSE, Amex, and NASDAQ from 1963 to 1990 as samples
and found that market value and book-to-market ratio can explain the differences that CAPM cannot
explain [6]. In 1993, Fama and French tested stock returns by time series regression method and put
forward the Fama-French Three-factor model, which explained excess returns jointly by market risk
factor, market value factor, and book-to-market ratio factor. The equation of the model is as follows:
(3)
Among them, (small minus big) represents the market value factor, which measures the
difference between the small market value portfolio and the large market value portfolio. (high
minus low) represents the book-to-market ratio factor that measures the difference between a high
book-to-market ratio portfolio and a low book-to-market ratio portfolio. (−) is a market risk
factor, which is the same as the factor in CAPM and is used to measure market changes. − is
the excess return of the portfolio.
In the follow-up research results, the Three-factor model can better explain the rate of return than
the CAPM model. Market risk factors can reflect the systemic risk of stock; market value factors and
book-to-market ratio factors reflect the unique non-systematic risk of the company. Fama-French
Three factor model is a widening of the traditional CAPM single-factor model, and it is also a
relatively mainstream pricing model now. After that, scholars also try to add new factors to better
explain the benefits. For example, in 1997, Carhart also expanded the Three-factor model and added
momentum factors to form the Carhart four-factor model [7].
2.4. Fama-French Five-factor model
The emergence of the Fama-French Three-factor model makes the revenue better explained, but there
are still problems that cannot be explained. Fama and French improved the original three-factor model
by adding a profitability factor and an investment factor to the model and creating a new Fama-French
Five-factor model. The equation of the model is as follows:
(4)
Among them, (robust minus weak) stands for profitability factor, which is the difference
between high profitability portfolio and low profitability portfolio. (conservative minus
aggressive) represents the investment factor, which is the difference between the portfolio of
conservative investment style companies and the portfolio of aggressive investment style companies.
The follow-up research of scholars found that the applicability of the Fama-French Five-factor
model has great regional differences, and its performance in different regional markets is quite
different. There may be redundant factors in the model. Scholars are also exploring new factors
through practice in order to have a better explanation for the benefits. As the second largest economy
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231077
100
in the world, market in China often has a significant impact on the global market, and its importance
is obvious. Therefore, it is very important for the market to find a pricing model with high
applicability.
3. Applicability in China
China’s stock market developed late, and it is dominated by a public-owned economy. Markets tend
to be more volatile, and there is strong state interference. Shi Donghui used CAPM to test the
Shanghai market and found that the Shanghai market has high systemic risk and that there is a
nonlinear relationship between systemic risk and return. There are non-systematic risks that cannot
be eliminated, which shows that CAPM cannot explain the benefits well [8]. Sun Gang tested the
constituent stocks of the SSE 30 Index as samples and also found that there are irrevocable non-
systematic risks, which also proves that CAPM has poor applicability to China’s market [9].
Since the advent of the Fama-French Three-factor model, Chinese scholars have tried to verify it
in China’s market. Huang Xingwang et al. used it to test the applicability of China’s A-share market
and the result showed that, compared with the American market, there was a significant scale effect
in China’s market, but its value effect was not significant [10]. Yang Kun and others studied the
volatility of stock returns in the Chinese A-share market. After studying, they found that the scale
effect and value effect of the A-share market were significant. This significance is particularly evident
in small companies and high-value investment portfolios. It also verified that the three-factor model
has the same explanatory power under different grouping situations [11].
According to Liu Yuanyuan’s research, the model has better explanatory power than the
conventional CAPM. The author found that the return rate of large-scale companies (the top 30% of
market value) is higher than that of small-scale companies (the bottom 30% of market value), which
is contrary to the scale effect. In February, small companies have higher yields than big companies.
The author thinks that the reason for this phenomenon may be the unique Lunar New Year in China.
Investors may choose to sell stocks for consumption before the New Year, and then buy investment
positions after the New Year, which causes the upward trend of the market [12]. Li Hui et al. studied
the stock markets of China and the United States between July 1994 and June 2013 to examine how
the three-factor model applied differently in the two markets. In contrast to the US market, the China
A-share market’s market risk factor has a higher impact, the scale effect is more explicative in small-
cap firms, and the book-to-market ratio factor has a little impact. he author thinks that this may be
because Chinese citizens prefer to believe in the development prospects of national policy orientation
rather than individual companies [13]. In conclusion, the three-factor model has stronger applicability
in China compared to CAPM, and has been recognized by scholars. Among them, market risk factors
and market value factors have stronger explanations.
In 2016, Zhao Shengmin and others found that the three-factor model is more appropriate for
China’s stock market than the five-factor model with profitability factor and investment factor, and
that profitability factor and investment factor may be redundant factors, using monthly A-share yield
data from Shanghai and Shenzhen stock markets from January 1995 to December 2014 [14]. Liu
Lanlan et al. tested the Fama-Macbeth cross-sectional regression method. The investment factor and
market value factor were found to be significant when the five-factor model was verified, however
the other three factors were weakly significant and had just little explanatory power [15]. Li Zhibing
et al. verified that the five-factor model has greater explanatory power than CAPM, the three-factor
model, and the Carhart four-factor using monthly data from the China A-share market from July 1994
to August 2015 [16]. Guo et al. tested China’s stock market with a five-factor model and concluded
that market risk factors, market value factors, book-to-market ratio factors, and profitability factors
all have good explanatory power for the excess returns of China’s stock market, but the explanatory
power of investment factors is weak [17]. Zhang Xindong et al. made an empirical analysis of the
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231077
101
monthly earnings data of A shares from 1999 to 2017 and found that adding profitability factors to
the three-factor model can greatly improve the explanatory power of the model. The investment factor
cannot contribute to the explanatory power, and it is considered that the investment factor is a
redundant factor [18]. Ouyang Hongbing uses turnover rate to measure liquidity based on the five
factor model. He added liquidity factors and established a six factor model. It was found that the six
factor model is more effective in explaining income than the three factor and five factor model [19].
4. Discussion
CAPM’s applicability has been proven to be low in China’s market, while the applicability of three-
factor and five-factor models is still inconclusive. However, some Chinese scholars believe that
investment factors are redundant factors in China’s market, which is different from the conclusions
obtained by Fama and French in the global market. China’s market is complex and huge, and scholars
also try to add new factors according to the characteristics of China’s market in order to achieve better
explanatory power. China’s market is highly volatile, and there are strong irrational factors in the
behaviour of market participants, which may lead to the inability of existing pricing models to
accurately predict asset price fluctuations. The institutional environment of China’s market is
different from that of the international market, with relatively high government intervention and a
relatively low opening degree of the capital market, which may have an impact on the asset price
formation mechanism. Future research needs to consider new influencing factors such as China’s
market situation, stock liquidity, fixed asset ratio, irrational factors, etc., and also consider many
factors such as the market institutional environment in China, etc., and use more data sources and
more advanced data analysis techniques to obtain more accurate market data and build a pricing
model that is more suitable for the market in China based on these data.
5. Conclusion
Fama-French five-factor model is still a relatively new research field, but its advantages over the
three-factor model are not obvious in both the global market and China’s market, and the redundancy
of investment factors found in China’s market is also worthy of scholars’ consideration. Because of
its uniqueness, the explanatory power of various models in China’s market is bound to be different
from that in the global market. The future research direction should be to develop new pricing models
according to their characteristics. With the growth of computer power in the future, scholars will be
able to make more effective use of stock market data for research and find out more effective factors
for China’s market so as to achieve a better interpretation effect. There may be some limitations
because this paper cannot collect all the research on China’s market and there is no empirical analysis.
With the increase of related research in the future, after considering other influencing factors, it may
get better results and explore a model with higher applicability.
Acknowledgment
First of all, I would like to express my gratitude to my mentor Professor Honeyberg. She provided
me with valuable guidance and support throughout the entire process of researching and writing this
article. Her expertise, patience have played an important role in improving the quality of this work.
Additionally, my thanks are given to the faculty and staff at Capital University of Economics and
Business for providing a conducive environment for conducting this research. Their resources,
facilities, and assistance have been instrumental in the successful completion of this article.
References
[1] Markowitz HM. Portfolio selection. The Journal of Finance, 1952, 7(1), 77.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231077
102
[2] Sharpe WF. CAPITAL ASSET PRICES: A THEORY OF MARKET EQUILIBRIUM UNDER CONDITIONS OF
RISK*. Journal of Finance, 1964, 19(3), 425–442.
[3] Fama EF, French KR. Common risk factors in the returns on stocks and bonds [J]. Journal of Financial Economics,
1993, 33(93): 3~56.
[4] Fama EF, French KR. Dissecting Anomalies with a Five-Factor Model, Review of Financial Studies, 2016, Vol.
29(1):69-103.
[5] Ross, SA. “The Arbitrage Theory of Capital Asset Pricing.” Journal of Economic Theory, vol. 13, no. 3, 1976, pp.
341–360.
[6] Fama EF, French KR. The Cross
‐
Section of Expected Stock Returns. Journal of Finance, 1992, 47(2), 427
–
465.
[7] Carhart Market M. On Persistence in Mutual Fund Performance [J]. The Journal of Finance, 1997, 52(1):57-82.
[8] Donghui Shi. Empirical study on risk of Shanghai stock market [J]. Economic Research, 1996(10):44-48.
[9] Gang Sun. Empirical Study on Capital Asset Pricing Models in the Chinese Stock Market [J]. Journal of Statistics
and Information, 2003 (05): 91-94
[10] Xingwang Huang, Sixiu Hu, Jun Guo. Two-factor model of Chinese stock market [J]. Contemporary economic
science, 2002(05):50-57+95.
[11] Xin Yang, Zhanhui Chen. An empirical study of three-factor asset pricing model in Chinese stock market [J].
Quantitative economy and technical economy research, 2003(12):137-141.
[12] Yuanyuan Liu. Empirical research on the efficiency of Chinese stock market: based on Fama-French three-factor
model [D]. Southwestern University of Finance and Economics, 2012:1-65.
[13] Lihui, Guanying Wang, Wei zhang. Three-factor model Pricing: How China is different from the United States [J].
Global Finance. 2014(7):37-45.
[14] Shengmin Zhao, Honglei Yan, Kai Zhang. Is the Fama- French five-factor model better than the three-factor model
— from the Chinese A-share market. Empirical evidence. Nankai Economic Research, 2016(2): 41 - 59 .
[15] Lanlan Liu, Shujie Yao. Empirical analysis of asset growth effect and market efficiency based on risk factors [J].
Journal of Chongqing University (Social Science Edition), 2016(5):34-42.
[16] Zhibing Li, Guangyi Yang, Yongchang Feng, Liang Jing. An empirical test of the ama-French five-factor model in
the Chinese stock market [J]. Financial research, 2017(6):191-206.
[17] Bin Guo, Wei Zhang, Yongjie Zhang, Han Zhang. The Five-Factor Asset Pricing Model Tests for the Chinese Stock
Market [J]. Pacific-Basin Finance Journal, 2017, 43:84-106.
[18] Yimeng Ma. An Empirical Study of Fama-French Five-Factor Model in Valuation of Internet Listed Firms [D].
Capital University of Economics and Business, 2018: 1-45.
[19] Hongbing Ouyang, Jingqiong Yu. Research on the Effect of Liquidity on the Asset Pricing Based on the Factor
Model [J]. Journal of Financial Development Research, 2020, (07):13-22.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231077
103
The Study of Risk Management of Financial Derivatives
-Taking the Archegos Liquidation Event as Example
Fengyi Sun1,a,*
1College of Business and Economics, The Australian National University, Canberra, 2601,
Australia
a. fengyi.sun@anu.edu.au
*corresponding author
Abstract: In recent years, the market economy has declined due to the impact of COVID-19.
Whether the person is a consumer or an investor, they both want to get more out of it.
However, people often may be blinded by this idea of quick success, and then lose a lot.
Taking the Archegos liquidation event as an example, this article analyzes the risk
management of financial derivatives and their effectiveness when the financial market
fluctuates through case studies. Through analysis, it can know that when buying or investing
in an aggressive investment company, it may bring them high returns and may also bring
them huge risks. Because they obtain returns through some of the characteristics of financial
derivatives, it is very risky. The impact of this event on Nomura was analyzed by using the
method of event research. Furthermore, through this incident, it found some loopholes and
problems in financial management and gave rationalization suggestions.
Keywords: financial derivatives, study event, Archegos, financial institution
1. Introduction
With the development of the financial industry, the scale of financial institutions continues to expand.
Risk management is essential to optimize portfolio management and, in turn, the management of
financial derivatives [1]. As an important tool for risk aversion and effective investment, the potential
risks of financial derivatives are also revealed [2]. Losses caused by the financial derivatives business
are not few. In every economic crisis, one or two large financial institutions fail. Moreover, some
large financial events such as the Archegos liquidation event will affect more or less financial
institutions. Through this incident, it can also be seen that some risk problems and loopholes in the
supervision of financial derivatives and financial institutions. Therefore, this article analyzes the
Archegos liquidation event as an example, hoping to enlighten the current standardized risk
management of financial institutions and financial derivatives.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231082
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
104
2. Case Description
2.1. Background
Archegos Capital Management is a Family Office-style hedge fund. Its predecessor was Tiger Asia.
In 2012, Tiger Asia was involved in an alleged count of insider trading, resulting in a settlement with
the U.S. Securities and Exchange Commission and payment of USD$44 million in damages. Tiger
Asia then returned its external capital to investors and changed its name to Archegos [3].
Its founder is South Korean Bill Hwang, who was the apprentice of renowned hedge fund pioneer
Julian Robertson. He is a bold investment trader with ambition and a desire to capture more wealth.
His investment style is also more radical [4]. He has always focused on and excelled in Asian equities.
2.2. Introduction event-Archegos
At the end of March 2021, Archegos exploded due to the sharp decline in the share price of Viacom,
which was heavily invested in high leverage [5]. It collapsed in an instant, losing USD$20 billion in
2 days [6]. For many years, Archegos has focused on investing in long positions in tech sector stocks.
It uses total return swaps, which allows the company to gain leverage through synthetic prime
brokerage [3]. Archegos buys CFD derivatives with high leverage and trades with major investment
banks to gain profits from rising prices. Due to a series of unfortunate events, some of its basket of
stocks began to fall. As share prices fell, Archegos faced unmet margin calls that led its major brokers
to sell shares, estimated at over USD$30 billion [7]. When the margin could not be replenished, it
triggered the forced liquidation of its CFD contracts by traders, and at the same time, investment
banks began to sell their shares in large quantities, passively reducing their asset exposure, and
resulting in a sharp decline in the company's stock price. At the same time, it also depressed the value
of Hwang's portfolio, which was unable to repay debts owed by creditors or pay the bank under the
swap agreement. For banks, exposure is essentially unsecured due to high initial margin requirements
[8]. When ACM events occur, some of the world's largest banks can have spillovers, with some facing
billions of dollars in losses [7]. For example, Nomura lost USD$2.9 billion and Credit Suisse lost
about USD$4.7 billion.
2.3. Event study
2.3.1. Determine the event window and estimate window
This article examines the impact of one of the companies that suffered the most in this event. For
event studies, this paper needs to clearly the date of the event, and the event window. On March 26,
2021, this liquidation occurred at Archegos Capital Management. Therefore, this article chooses this
date as the date on which the event occurred. Further, 30 days before and after the event date is
selected as the event window. So, the event window is 61 days, from t-30 days to t+30 days. The
event window for this study was from 24 February 2021 to 25 April 2021. The estimated window is
generally considered to range from 120 to 255 working days [9]. The estimation window is mainly
used to determine the relationship between individual stocks and the market. The estimation window
for this study was from 28 July 2020 to 23 February 2021. From t-210 to t-31, the estimated window
is 210 days. This work only considers the analysis of the trading day.
2.3.2. Calculate expected return, abnormal return (AR) and cumulative abnormal return
(CARs)
By fitting a linear regression using the least squares method to expect a rate of return, the calculation
formula is the formula below.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231082
105
(1)
In this formula, ERt is the expected return on the t trading day, Yt is the company's stock yield on
the t trading day, It is the market yield on the t trading day, according to the data in the Wind financial
terminal data, Nomura is listed on the New York Stock Exchange in the United States and belongs to
one of the top ten securities companies in the United States. The data used for the analysis are all
from Wind Financial Terminal data, and since it is a US stock listing, the market yield is using the
Dow Jones Industrial Index. Fitted regression is performed by Excel data analysis.
Figure 1: Chart of Nomura's individual stock yields and Dow Jones Industrial Index market returns.
Then, this function is obtained.
(2)
Secondly, the formula for the abnormal return (AR) is the calculation formula below.
(3)
Moreover, the formula for the cumulative abnormal return (CARs) is the calculation formula
below.
(4)
According to the above two formulas, the Nomura window period abnormal return and cumulative
abnormal return can be calculated. Since there are weekends in the middle when the market is closed
and no trading is held, the earnings do not change on weekends.
To determine whether the abnormal return is caused by random market fluctuations in the stock or
by the Archegos liquidation event, a one-sample T-test is required for the cumulative abnormal return.
H0: CAR=0, the Archegos liquidation event does not affect Nomura's stock price changes.
H1: CAR0
y = 0.3741x + 0.0576
R² = 0.0326
-20.0000
-15.0000
-10.0000
-5.0000
0.0000
5.0000
10.0000
15.0000
-5.0000 -4.0000 -3.0000 -2.0000 -1.0000 0.0000 1.0000 2.0000 3.0000
Y(R|t)
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231082
106
By using STATA statistical analysis software for this one-sample T-test, the following results can
be obtained.
Table 1: One-sample T test results.
Variable
t
degrees of
freedom
Obs
Mean
Std. Err.
Std. Dev.
[95% Conf. Interval]
CAR_1
-
5.5331
41
42
-6.96521
1.258821
8.158091
-9.50745
-4.42297
By calculation, P=0.00. It shows that the test results are very statistically significant, that is the
test results are very valid, thus indicating that the Archegos liquidation event has a very large impact
on the movement of Nomura's stock price.
2.3.3. Result analysis
Figure 2: The trend of abnormal return and cumulative abnormal return during the event window.
According to Figure 2, it can be seen that the abnormal return throughout the event window period
fluctuates around the zero point except for a sharp decline within 3 days of the event, in the range of
-3.2328% and 2.8562%. When the incident occurred, whether it was from an abnormal return or a
cumulative abnormal return, it was clear that Nomura was greatly affected. Within 2~3 days of the
event, Nomura gave a very rapid and sharp downward response. Within 3 days of the event date, the
abnormal return reached a minimum of -12.728%, and the cumulative abnormal return dropped from
3.5670% to -9.1610. On the tenth day, the abnormal return gradually recovered and approached
1.4845%. However, it still fluctuates around 0 as a whole. The cumulative abnormal return has been
on a downward trend, and the downward trend only began to slow down around day 13. Therefore,
the Archegos liquidation event had a very significant negative impact on Nomura, and it can be seen
that it has brought significant losses.
-20
-15
-10
-5
0
5
t=-30
t=-28
t=-26
t=-24
t=-22
t=-20
t=-18
t=-16
t=-14
t=-12
t=-10
t=-8
t=-6
t=-4
t=-2
t=0
t=2
t=4
t=6
t=8
t=10
t=12
t=14
t=16
t=18
t=20
t=22
t=24
t=26
t=28
t=30
AR CAR
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231082
107
3. Analysis On The Problems
3.1. Financing leverage is too high
When the stock price rises, high leverage can make a huge profit. However, when the stock price fell,
the high leverage also brought him huge losses. Just as Archegos liquidation, it faced huge losses
when most of the stocks in its basket were trending down.
At the same time, for banks, high leverage is also a state of coexistence of advantages and
disadvantages. For example, Credit Suisse offers up to 300 times leverage with Archegos. It resulted
in a loss of 4.4 billion Swiss francs (USD$4.7 billion).
In the event with a low margin ratio required by the bank. The total yield swap deal structure for
Archegos has provided him with great convenience. Not only did Archegos secure a lower margin
ratio requirement, but it also allowed it to continue to increase its leverage. It has led to other hedge
funds mainly using 2 to 3 times leverage, while Archegos usually reaches 5 to 6 times leverage [10].
3.2. High concentration, low liquidity, and high volatiliy
Archegos' exposure is highly concentrated in technology stocks, with individual stocks holding 5 to
10 times the average daily trading volume of the stock [4]. Therefore, it also has a higher historical
rate of return. Highly concentrated funds have more divergence in performance, and their average
volatility is higher. So, it needs to take greater risks.
3.3. Regulatory loopholes
Because Archegos is a Family Office-style hedge fund, despite its size, it is not required to disclose
its portfolio to regulators. It leads to insufficient disclosure. In addition, Archegos indirectly holds
shares through over-the-counter transactions with several investment banks through derivatives,
circumventing the reporting requirements of US securities regulatory authorities [11]. Since there is
no unified industry standard for CFD derivatives trading, the risks arising from them are difficult to
be identified by the regulator promptly.
4. Suggestions
Through the analysis and research of this paper, the following three suggestions are provided for
financial institutions. First of all, for the selection of customers, it is necessary to fully understand
and rationally and objectively analyze the customer's history information [4]. Especially, the illegal
and bad record it has appeared. People can't invest heavily just because it has good data and potential
trends. Funds with high concentration do not necessarily have good long-term performance. While
they achieve high short-term returns, they also hide huge risks. Therefore, when selecting the type of
fund, it is necessary to analyze it according to the market phase. Secondly, financial institutions
should quantitatively estimate the customer's financing leverage ratio and its potential losses on time,
and make timely and corresponding adjustments to the financing leverage ratio. It allows some large
losses to be avoided. Finally, financial institutions should improve their risk supervision capabilities,
formulate good risk management policies to avoid loopholes and optimize their systematic analysis
capabilities of risk exposure.
5. Conclusion
This article specifically analyzes the huge losses caused by a company that has been addicted to high-
risk and high-yield operations through the radicalized use of financial derivatives. At the same time,
from the partners with Archegos, through the method of event research, it understands how significant
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231082
108
the impact of this event is. Furthermore, it was found that financial derivatives have the problems of
high leverage, low liquidity, and high volatility. Financial institutions have problems and loopholes
in some reporting requirements, which enable some companies to obtain benefits through this
problem to carry out non-standardized operations. As a result, regulators are unable to identify and
mitigate their risks promptly There are some drawbacks to this study, such as the selection of only
one company as a representative for the study. In future research, more companies can be selected to
jointly study to make their case analysis more in-depth and perfect.
References
[1] Hammoudeh, S., & McAleer, M. (2013) Risk management and financial derivatives: An overview. The North
American Journal of Economics and Finance, 25, 109-115.
[2] Ge, Z. (2022) Research on risk management of financial derivatives of commercial banks in China. Chinese market,
30, 56-58.
[3] Jeong, Y. S. (2023) Portfolio Investment: A Driving Structural Factor behind Recent Financial Turmoil. KIEP
Research Paper, KIEP Opinions, 260.
[4] Qi, J. (2021) Inspiration of Archegos Defaults on counterparty credit risk management Banking Management.
International finance, 10, 64-69.
[5] Wang, Y. (2021) Strictly control highly leveraged financial speculation. [J]. Caixin Weekly. 13, 9-9.
[6] Bouveret, A., & Haferkorn, M. (2022) Leverage and derivatives-the case of Archegos., Retrieved from
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4165572
[7] Patrick, M., & Kosaku N. (2021) Archegos Blowout Heaps Pressure on Credit Suisse. Wall Street Journal. 2021.
March 30. Retrieved from https://www.wsj.com/articles/archegos-blowout-heaps-pressure-on-credit-suisse-
11617108216?mod=series_archegos
[8] Ita, A. (2023) Counterparty credit risk: Lessons from recent events. Journal of Risk Management in Financial
Institutions., 3, 256-272.
[9] Andoh-baidoo, F. K., & Osei-bryson K. M. (2007) Exploring the Characteristics of Internet Security Breaches that
Impact the Market Value of Breached Firms[J]. Expert Systems with Applications, 3, 703-725.
[10] Li, N. (2021) Warning of Greenhill and Archegos Events. Financial Market. 2021, 94-95.
[11] He, Y. (2021) The transregulatory arbitrage of credit risk to market risk: A case study of Archegos implosion event.
International Finance. 94, 101-104.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231082
109
The Influence of Sino-US Trade War on Chinese Students
Studying in the United States
Xuemei Wang1,a,*
1Business school, University of Leeds, Leeds, LS2 9JT, United Kingdom
a. xuemeiwang@ldy.edu.rs
*corresponding author
Abstract: China has promulgated a series of policies on trade, immigration, and culture,
which will undoubtedly greatly impact Chinese students preparing to study in the United
States. This paper focuses on the influence of American policies and a series of chain
reactions on Chinese students studying in the United States since the Sino-US trade war in
2018. It analyzes the factors that affect Chinese students’ decision to study in the United
States. This paper conducted a quantitative study by issuing online questionnaires, and 323
valid questionnaires were received. The basic information, educational experience, and
factors affecting the respondents’ going to the United States were investigated, and a series
of statistical methods were applied to analyze them. Finally, the following conclusion is
drawn: the Sino-US trade war greatly influences Chinese students’ decision on whether to go
to the United States, mainly including the influence of visa and professional restrictions. The
factors that affect Chinese students’ decision to study abroad are family and policy factors,
teacher factors, and intermediary factors.
Keywords: Sino-US trade war, Chinese students, United States, international students
1. Introduction
Chinese students have studied in the United States for over forty years. According to statistics, in
2022, Chinese students went to the United States to receive education at different stages, from primary
school to doctoral degrees. China is the largest source of international students in the United States.
Since March 2018, the Trump administration has launched a 301 investigation on China and
promulgated a series of policies on trade, immigration, and culture, which will undoubtedly greatly
impact Chinese students preparing to study in the United States. This article will discuss these effects
in detail.
2. Literature Review
2.1. Sino-US Trade War
The trade dispute between China and the United States was further escalated in March 2018 when the
United States launched a series of sanctions against China in accordance with Article 301 of the Trade
Act of 1974. Therefore, the tariffs of both sides were raised, which led to the development of trade
disputes into trade wars [1]. In response, China’s Ministry of Commerce released a list of 232
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231086
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
110
measures on March 23rd, including increasing tariffs on products imported from the United States of
about 21 billion RMB and suspending concessions on steel and aluminum products imported from
the United States.
The main reason for the Sino-US trade war is the imbalance of international payments. That is,
China’s exports to the United States are much larger than those of the United States to China. In 2017
alone, China’s trade surplus with the United States reached 2,726.5 billion yuan [2]. China’s perennial
trade surplus with the United States and the United States’ restrictions on products exported from
China to China, such as some high-tech products, made the United States want to change this situation
and protect its trade. However, in the past 40 years, China’s GDP growth has been three times that of
the United States, which worries American stakeholders because the dominant position of
international trade may further deteriorate [3].
2.2. The History and Present Situation of Chinese Students Studying in the United States
As early as 1847, Rong Hong and his classmates Huang Sheng and Huang Kuan went to the United
States to study. Rong Hong returned from his studies 7 years later and devoted himself to introducing
Western learning to the East, creating the earliest Chinese education in the United States [4]. Statistics
show 2,710 Chinese students in American universities in 1948, distributed in 45 states [5]. The
following year, the number of students studying in the United States increased by 40%, with 3797
students. Some researchers believe that these post-war students studying in the US have some obvious
characteristics compared with previous students studying in the US: the level of education has been
greatly improved, the number of graduate students accounted for more than half of all students
studying in the US; the number of students studying in the United States choosing liberal arts
increased [6]. Since 2010, Chinese students in US universities have exceeded 150,000, accounting
for 18.5% of all US students studying abroad [7].
2.3. Reasons for Chinese Students Studying in the United States
After 30 years of rapid education development in the United States, Chinese people’s understanding
of studying there has become relatively mature. One of the manifestations is that Chinese students
studying in the US now show a more diversified trend in majors and development choices, showing
rationality and pragmatism.
The affirmative action policy was formulated by President Obama during his term of office, which
enabled American universities to adhere to diversified admission during the Obama administration.
It mainly ensures that students of every gender, race, and belief can have a fair chance to enter
American universities. According to a study in 2012, the SAT scores of Asian students in the top 30
universities in the United States are, on average, 160 points higher than whites, 210 points higher
than those of brown people and 450 points higher than black people. In addition, yellow people are
not worse than other races in other aspects. The promulgation of the equal rights law undoubtedly
supports protecting Asian students.
The change in China’s economic development is the main reason for the upsurge of studying in
the United States in the past decade. Since 1980, the people’s living standard in China has constantly
improved, and the middle class has expanded. The high cost of studying abroad is no longer an
obstacle for overseas students in China. America has the largest number of high-level universities in
the world. In the rankings of universities worldwide, American universities are among the top, and
the United States is naturally more popular with Chinese university students.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231086
111
2.4. The Influence of the Sino-US Trade War on Chinese Students
Since 2018, the Trump administration has reissued the document of the Affirmative Action Bill. They
decided to return to the policy of “racial neutrality” during the Bush administration, and American
universities will no longer consider racial factors when enrolling. What followed t was that the Trump
administration called on American universities to close the application channels for international
students in sensitive majors, especially Chinese students.
The Trump regime influences Chinese students studying in the United States. As the world’s
largest recipient of international students, the number of new international students in the United
States has continued to increase yearly since its statistics. Still, this number has dropped by more than
10,000 in one year for the first time since Trump took office, and it is much more difficult for 38
countries that are not long-term allies of the United States to obtain American visas.
The current situation in the United States worries many students and parents who want to study
there. This can be seen from the decrease in college applicants in the United States and inside
HigherEd reported that a recent survey of more than 250 universities in the United States showed that
4 out of 10 universities indicated that the number of applications from international students
decreased [8]. Among them, 25% of the schools surveyed said that the number of undergraduate
applicants in China decreased, and 32% said that the number of graduate applicants from China
decreased. In this regard, college freshmen admissions officers generally attribute it to the changes in
the 2016 US election and visa policy. 77% of the international admissions officers of the schools
surveyed expressed concern about the future.
3. Method
3.1. Introduction
This research uses the design of quantitative research. Quantitative research is a way to process
relevant and statistical data using mathematical methods, reflecting the regular relationship between
related variables and calculating various numerical values of the research object. Because this study
adopts a survey method and collects data through an online questionnaire, the purpose is to analyze
a large number of data and to study the influencing factors of Chinese students studying in the United
States. Hence, the quantitative design is the most suitable.
3.2. Questionnaire
This research uses the design of quantitative research. Quantitative research is a way to process
relevant and statistical data using mathematical methods, reflecting the regular relationship between
related variables and calculating various numerical values of the research object. Because this study
adopts a survey method and collects data through an online questionnaire, the purpose is to make a
statistical analysis of a large number of data and to study the influencing factors of Chinese students
studying in the United States, so the quantitative design is the most suitable.
The research method of a questionnaire survey can collect the relevant opinions of the respondents
as comprehensively as possible and collect information systematically. In addition, the questionnaire
has a larger sample size to improve the accuracy of the data. Other survey tools, such as focus groups
or interviews, can be biased due to the smaller sample they typically collect data from [9,10]. In this
study, in order to ensure that the respondents are international students or interested groups in
studying abroad, questionnaires were distributed on the websites of the International Student Forum
and WeChat friends circle of the top 100 universities in the United States.
The questions and options of the questionnaire are determined according to the literature review
results and the problems studied in this paper. This questionnaire investigates the relevant situation
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231086
112
of the respondents from three aspects: first, the basic information of the respondents; second, their
intention to study abroad, such as the country of intention to study abroad, changes in the country of
intention, etc., and finally, the factors that affect the respondents’ decision-making about studying
abroad.
3.3. Ethical
Considering the protection of respondents’ privacy, the questionnaire did not collect information
about their specific identities, such as their names and addresses. When launching the questionnaire,
the respondents will be told that the questionnaire is not mandatory and they are free to choose
whether to participate in the survey or not.
3.4. Limitations
Due to the time factor, the questionnaire can only set a certain sample size. Generally speaking, the
larger the sample size, the more universal the data conclusions are. In theory, more than 300 samples
can reach 95% confidence. Because of the way of issuing questionnaires, it is impossible to rule out
repeated questionnaires, although considering this limitation, and explaining the instruction of
“Please read carefully when answering questions, and don’t answer them at will”. At the same time,
the use of convenient sampling limits the external validity of the data because the respondents are not
randomly selected from a wider population, so, as mentioned above, it will produce biased samples
[11].
4. Results
A total of 330 questionnaires were received in this survey, of which 323 were valid. More than half
of the respondents were women, accounting for 61.6%, reaching 199; Men accounted for 38.4%,
reaching 124. Among these 323 people, 89.4% of the respondents have plans to study abroad in the
next three years. Table 1 shows the basic information of the participants and their study abroad.
Table 1: Sample Characteristics.
General Information
Sample number
%
Gender
Male
124
38.4
Female
199
61.6
Previous Study Abroad Experience
Yes
124
38.4
No
199
61.6
Plan to Study Abroad in the Next 5 Years
Yes
289
89.4
No
34
10.6
4.1. Multiple Response Analysis
The questionnaire selects the six most relevant factors that affect Chinese students to study in the
United States, namely, restricting Chinese students from studying technology that may be used for
national security (factor 1), visa restrictions introduced by the Trump administration (factor 2), the
essence of Sino-US trade war is technology competition (factor 3), the speed of China students going
to the United States is slow due to visas and other reasons (factor 4), China students have more
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231086
113
restrictions on studying in the United States (factor 5), and Sino-US relations (factor 6). The survey
results are shown in the figure below.
As can be seen from the data, the biggest influence on the respondents to study in the United States
is that it is more difficult for Chinese students majoring in national security to apply, accounting for
more than 70%. The visa restrictions accounted for 69%, followed by the Sino-US trade war, in which
technology competition was the essence, accounting for 67%. China students’ speed of going to the
United States slowed down due to visas and other reasons, accounting for 66%, and international
students’ restrictions on going to the United States increased, accounting for 57%. Sino-US relations
influenced 33% of the respondents. Generally speaking, the higher the correlation, the better the
reliability of the data. The α coefficient in this survey is the internal correlation test. Generally, there
are two ways. First, add questions. According to the theoretical basis and own theoretical evidence,
add questions that match the concept of the questionnaire: The more same questions, the better the
reliability. Second, delete topics that differ from other items and increase the alpha coefficient. As to
which one to delete, we can see the operation of calculating the α coefficient according to SPSS. In
addition, it can also be carried out on a theoretical basis, but the title is deleted. In short, it is to add
and delete the same questions with different qualities. Reliability is a necessary condition for validity.
According to the data, the reliability of the questionnaire is acceptable.
4.2. Correlation Analysis
The analysis was pre-tested before analyzing the factors that affect Chinese students to study in the
United States. The test result data is much higher than the standard value, indicating that the data is
suitable for sampling and factor analysis. This paper makes a correlation analysis of the reasons that
affect Chinese students’ decision to go to the United States. Generally speaking, the stronger the
correlation, the closer the correlation value is to 1. The correlation values of teachers, parents,
persuasion of study abroad agents, peer competition, and personal attitude towards current affairs are
all between 0.67 and 0.88, which shows that these six factors have a positive influence. Teachers’
opinions and peer competition have the strongest influence, and the correlation values are 0.87 and
0.83, respectively.
4.3. Multiple Regression Analysis
Table 2 is an analysis of variance table to test the reliability of survey data and the significance of
multivariate linear equations. The f value of this data is much greater than 1, indicating that the
difference between different mean values is statistically significant. Multiple linear regression can
explore the influence of different independent variables on dependent variables [11]. This survey
refers to the influence of different elements on Chinese students’ decision to study in the United States.
Multiple linear regression can get quantitative results to explore the research problems better.
Table 2: Variance analysis.
ANOVA
df
ss
ms
f
Significance f
Regression analysis
5
228.4485
45.6897
502.5349
1.8E-147
Reidual
314
28.54839
0.090918
Total
329
256.9969
Table 3 is the T-test table, which is tied with the F-test to test the reliability of the data. The number
of samples in this survey is 323, which is relatively large, so the t distribution is (0,1). When α is set
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231086
114
to 0.05, the quantile of α on n is 1.96, and the t Stat in the table far exceeds 1.96, which proves that
the linear regression coefficient is reliable.
Table 3: T-test and linear regression.
Coeffici
ents
SE
T Stat
P-value
Lower
95%
Upper
95%
Lower
limit
95.0%
Upper
Limit
95.0%
Intercept
-
0.13882
0.04858
2
-
2.8788
6
0.005597
-
0.4326
6
-
0.0191
7
-
0.1125
5
-0.02916
X
Variable 1
0.28378
2
0.01286
11.610
27
1.46E-26
0.2226
78
0.2765
82
0.2226
78
0.276582
X
Variable 2
0.22717
3
0.01238
3
5.5073
84
6.49E-09
0.0870
24
0.2705
23
0.0870
24
0.
270523
X
Variable 3
0.25274
8
0.01358
3
7.2699
16
2E-08
0.2043
98
0.1002
98
0.2043
98
0.100298
X
Variable 4
0.11678
2
0.01891
9.1561
78
4.2E-25
0.2845
18
0.1710
82
0.2845
18
0.
171082
X
Variable 5
0.25251
8
0.01123
8
7.8900
27
5.987E-
11
0.2085
57
0.2855
01
0.2085
57
0.
285501
5. Discussion
5.1. Introduction
In this chapter, the research results of this paper will be compared and discussed with the existing
research, that is, the second part of this paper. The scope of application and inapplicability of this
conclusion will also be discussed, and finally, some relevant suggestions will be put forward.
5.2. Findings
This study aims to investigate the following questions: Does the Sino-U.S. trade war impact Chinese
students studying in the United States? If so, how is it affected? The following key findings have shed
light on this question from a review of relevant literature and primary research. First of all, the Sino-
US trade war seems to have had an impact on students going to study in the United States. Due to a
series of policy changes, such as increasing the difficulty of applying for sensitive majors and
reducing the number of H1-B and F-1 visas, Chinese students studying in the United States decreased
significantly in 2018. Students who originally planned to study in the United States changed their
first choice to EU countries or Britain.
5.2.1. Policy Influence
The results of this study show that 68% of the questionnaire participants do not choose the United
States as their intended country for studying abroad because of visa restrictions, and the most
important aspects of visa restrictions are the validity period and application difficulty. The visa
problem caused by the policy has affected the decision of Chinese students to study in the United
States, which is consistent with the results of the literature review in the second part of this paper.
H-1B visa is a kind of non-immigrant visa in the United States and a work visa. This visa is mainly
issued to non-American employees with special professional skills. People with H1B visas can get
legal residency for up to six years. After that, if they can’t get other types of visas, they must leave
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231086
115
the United States. According to the relevant policies, it is more difficult to obtain an H1-B visa, which
also greatly affects the intention of Chinese students to study in the United States. In this study, there
was no special option for an H1-B visa, so comparison with previous research with respect to this
aspect is not possible.
Another change caused by policy reasons is the increase in tuition fees, especially related majors
in high-tech fields. In this paper, 25% of the respondents gave up studying in the United States
because of the high increase in tuition fees.
5.2.2. Teacher Influence
According to the literature in the past, Chinese students are mainly influenced by their parents when
deciding to study abroad. However, according to the survey in this paper, the teachers in middle
schools or universities have a greater influence on their decision-making.
5.2.3. Parents Influence
In China, parents are also an important factor influencing students’ decision to study abroad, mainly
divided into two aspects. The first is the income of parents or families. Data show that more than 85%
of the funds for Chinese students to study abroad come from their families, and funds are the key to
deciding whether students can study abroad. According to the data, in 2022, Chinese students spent
an average of 100,000 dollars a year in the United States. Therefore, the high cost of studying abroad
determines that parents are an important factor affecting students’ studies in the United States.
Secondly is the parents’ education level. If parents have overseas experience or received higher
education, their willingness to let their children study abroad is stronger than parents who have only
received secondary vocational education. Parents who have received higher education are more
willing to create conditions for their children to study abroad, including financial conditions.
Therefore, the influence of parents’ spirit and financial support on children’s decision to study in the
United States can not be ignored.
5.2.4. Agence Influence
The investigation of this paper also considers the influence of intermediaries on students. The advice
of intermediaries in the United States or China impacts students’ decision to study in the United States,
including the choice of study country, preparation for different periods before studying abroad, and
choice of schools and majors. The influence of an intermediary is rarely considered in previous studies.
5.2.5. Competitive Influence
Conformity psychology is also a major factor affecting students’ decision-making. Some students say
they want to study abroad because all their classmates or friends of the same age have gone to study,
so they want to study abroad themselves. Or listen to classmates say that the current employment
form is not good and is the best choice to study abroad and improve yourself, so blindly follow
without considering other circumstances. The research of Xu and Tu shows that students who have
been engaged in consistent activities for a long time are more likely to correlate strongly with each
other’s behavior. Obviously, this kind of student’s willingness to study abroad is not clear, and there
is a high possibility of giving up due to various factors in the future. Therefore, students’ values and
feelings of competition with peers are also important factors that affect students’ decision to study
abroad, which is consistent with the conclusions of other related documents.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231086
116
5.3. Research Implications
The advice of teachers and parents was the most significant factor influencing whether Chinese
students would decide to study in the United States. In addition, this study’s other four strongly
correlated influencing factors were all positively correlated. This can provide suggestions for students
and families who want to study abroad in the future and can guide them to make relevant decisions
about studying abroad from more angles. At the same time, this finding provides support for related
studies. Therefore, results relating to the factors influencing Chinese students study in the US during
the U.S.-China trade war proposed in this paper may provide new ideas for future researchers in view
of the diverse choices of different students.
5.4. Limitations
The sample surveyed in this study was restricted to the region, and more than half of the respondents
are residents of Beijing. In addition, although the questionnaire has tried to collect the factors that
may affect students’ decision-making, it is a comprehensive decision to influence whether students
study abroad and where to study, so the factors designed in the questionnaire are limited and not
comprehensive. Compared with other research techniques in the survey method, such as interviews,
the restriction imposed on the questionnaire design by the need to collect quantitative data did not
consider the wide range of decision-making factors. Finally, the COVID-19 epidemic broke out at
the end of 2019, which greatly reduced the number of round-trip flights between China and the United
States, changed the original flight routes, and made the entry procedures in China and the United
States cumbersome, which would also have an impact on the decision-making of China students.
However, this survey did not consider this sudden factor.
6. Conclusion
According to all the survey results in this paper, Chinese students have changed when making relevant
decisions to study in the United States. Respondents in this article believe that the deterioration of
Sino-U.S. relations caused by the Sino-U.S. trade war is the biggest reason why respondents no longer
choose the United States as a country to study abroad. Secondly, the very important reason is that the
United States has tightened the visa for Chinese students and the immigration policy due to the
deterioration of Sino-US relations, which has greatly affected the willingness of Chinese students to
study in the United States.
This paper also analyzes the factors that affect the decision-making of studying abroad. Based on
the family and policy factors in previous studies, it also analyzes the factors of teachers and
institutions studying abroad. This study also shows that the influence of teachers is deepening day by
day.
References
[1] Zhao, K. (2018). Causes, Development and Countermeasures of Sino-US Trade War. The science of leadership
forum. 10(1)3-13.
[2] He, M. (2022). Research on the Trade Effect of Sino-US Trade War on China’s Export Commodities. Donghua
University. Retrieved from: https://oss-cnki-net.sjlib.cn.ade28b3498069
[3] Yu, Y. D. (2019). The deep roots and future trend of the US trade war. Research on financial issues,8(01)3-12.
[4] Yufa, Z. (2002). Returned Chinese Students from America and the Chinese Leadership (1846-1949). Chinese Studies
in History, 35(3), 52-86.
[5] Zhang, W. (2009). Do “Sea Turtles” creep faster than “Soft-shell Turtles”: A quantitative study of academic
performance of law faculty in premier Chinese law schools. Guang Hua Law Review, 4, 1. Ren, C. A Study on the
American Government’s Policy and Influence on China Students from the Perspective of “Immigration and
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231086
117
Diplomacy” (1949-1957). Northeast Teachers University. Retrieved from: https://kns-cnki-net-
443.wvpn.sjlib.cn/kns8/DefaultResult/Index
[6] Tao, W. Z. (1999). History of Sino-US Relations 1949-1972.Shanghai: Shanghai People’s Publishing House, pp.
12.
[7] Yang, B. (2019). An Analysis of the International Student Exchange between China and the United States under the
Background of the High-level Consultation Mechanism of Humanities Exchange between China and the United
States (2010-2018). Foreign Affairs College. Retrieved from: https://oss-cnki-net.sjlib.cn. ad06.
[8] Redden, E. (2019). International student numbers in US decline. Inside Higher Ed.
[9] Yang, W. M., Yang, Q. Y. and Kang, N. (2022). Suggestions on using second-hand data in multi-method research-
taking consumer behavior research as an example. Lanzhou University. Retrieved from: https: https://kns-cnki-net-
443.wvpn.sjlib.cn/kns8/Detail/Read/
[10] Luo, N. (2021). Literature review on the application of questionnaire survey research methods. Regional
governance, 6(01), 286-287.
[11] Wang, B. S. (2009) Medical Statistics and SAS Application. Shanghai Jiaotong University Press: Shanghai,
2009:95:135. Luo, N. (2021). Literature review on the application of questionnaire survey research methods.
Regional governance, 6(01), 286-287.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231086
118
Research on the Influencing Factors of Chinese Gasoline
Price Fluctuation During the COVID-19 Epidemic
Chuhan Zhang1, a, *
1Ranney School, Tinton Fall, 07724, New Jersey, USA
a. zchvvv7@gmail.com
*corresponding author
Abstract: Gasoline is an important energy source in China, and the stability of its price lays
the foundation for the country’s transportation and industrial development. Therefore,
fluctuations in gasoline prices are directly related to the normal operation of the country’s
transportation system and industrial production. Since 2020, during the COVID-19 outbreak,
China’s gasoline prices have experienced huge abnormal fluctuations, which have affected
the national economy. The purpose of this article is to study what factors have affected
gasoline price fluctuations in China during the COVID-19 pandemic and how to solve the
problem of excessive fluctuations. This article analyzes the fluctuation characteristics of
gasoline prices in China from 2019 to 2023 and lists four factors that affected gasoline price
fluctuations through qualitative research. For unexpected fluctuations like this, this article
proposes two feasible suggestions, including formulating a stronger environmental protection
policy and strengthening international cooperation based on oil exports to stabilize the price
of gasoline.
Keywords: gasoline, COVID, price fluctuations, influencing factors
1. Introduction
Gasoline is a petroleum product and an important secondary energy source in China. It is of great
significance in transportation and industry in modern society. Gasoline is the most commonly used
fuel for automobiles. It is not only an essential energy source for transportation but is also used for
the operation of many mechanical devices. It can be said that gasoline laid the foundation for the
transportation and industrial development of China. The stability and security of gasoline supply are
important guarantees for national economic development, so it is necessary to stabilize gasoline prices.
However, in the past four years, Chinese gasoline price fluctuations have shown very unusual
trends. Gasoline prices have seen a trough and a peak in just a few years, going from one extreme to
the other due to the COVID and the Russia-Ukraine conflict one after another. This article is to study,
during the COVID period, what factors have affected the fluctuation of gasoline prices in China, and
how to solve the problems of these factors. The purpose of this study is to propose possible
countermeasures to stabilize gasoline prices, prevent huge price fluctuations that may occur in the
future, and thus promote stable economic development.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231052
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
119
2. Characteristics of Chinese Gasoline Price Trend
In China, gasoline No. 92 and No. 95 are the gasoline most people use. In the following price analysis,
NO. 92 gasoline in Beijing is used as an example. The price trend of No. 95 gasoline is basically the
same as that of No. 92 gasoline, but the average price is 0.5 CNY/L higher. according to figure 1.
Figure 1: No.92 gasoline price trend in China from 2019 to 2023.
In 2019, the price of gasoline in China has been fluctuating slightly in the range of 6.44 CNY/L to
7.13 CNY/L. The beginning of 2019 was the lowest point in the whole year of gasoline prices. It
showed an upward trend from January to May, then fell after June, and finally began to rise in
November. At the beginning of the following year, gasoline prices recovered to 6.99 CNY/L before
falling to 5.5 CNY/L in just three months. From March to November 2020, the price of gasoline in
China ushered in the lowest price in many years. It fell to 5.43 CNY/L in early November, and then
began to gradually recover. Until the end of 2020, the price of gasoline was still 5.95 CNY/L. It is
very unusual for the price of gasoline to be below 6.0 CNY/L for most of the year. In 2021, gasoline
prices continued to rise, reaching 7.56 CNY/L at the end of October after three consecutive increases.
This was the first time that Chinese gasoline prices have exceeded 7.5 CNY/L since 2018. After this
brief peak, gasoline prices fell back to 7.14 CNY/L at the end of the year. Immediately afterwards, in
2022, gasoline prices experienced an unprecedented rise, climbing all the way to 8.74 CNY/L in early
April, and reaching 9.33 CNY/L in mid-June. Although the price dropped in the second half of the
year, the gasoline price basically fluctuated between 8.0 CNY/L and 9.0 CNY/L throughout the year,
and even broke through 9.0 CNY/L in the second quarter. In 2023, gasoline prices still fluctuated
stably between 7.5 CNY/L and 8.0 CNY/L, and there is currently no downward trend [1].
Under normal circumstances, gasoline prices will fluctuate slightly within a price gap of 1.0 CNY
every year, but it is clear that gasoline prices have not fluctuated in accordance with the general law
since COVID broke out. During these five years, gasoline prices in China have experienced a sharp
drop and then a sharp rise, which is very rare in the history of gasoline prices. It has been less than a
year and a half since the price of gasoline rose from 5.43 CNY/L in 2020 to 9.33 CNY/L in 2022.
There are many factors that lead to this unusual price fluctuation.
3. Factors Influencing Gasoline Price Fluctuations
Since 2020, gasoline prices in China have suddenly plummeted and then risen. The price fluctuation
range exceeded 2.5 CNY/L, which is an abnormal fluctuation. The outbreak of COVID and the sharp
drop in the price of crude oil are the reasons gasoline prices fell in 2020, while the Russia-Ukraine
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231052
120
conflict and national environmental protection policies were the factors that caused gasoline prices to
surge in 2022.
3.1. COVID
In December 2019, the coronavirus epidemic broke out in Wuhan, China. With the arrival of the
Spring Festival, population mobility has increased dramatically, leading to the rapid spread of the
epidemic across the country. In the face of the increasingly serious epidemic, China has implemented
blockade measures, and people are restricted from hanging out, so the demand for gasoline has
dropped sharply [2]. With the shrinking demand, the supply price of refined oil has also fallen all the
way, ushering in the lowest point in many years from April to September 2020.
In the first quarter of 2020, domestic gasoline sales in China were dismal due to the blockade
policy, so domestic companies increased exports to ease the increased supply pressure in the domestic
market. China’s gasoline exports are mainly to Southeast Asia. Due to the increase in domestic
gasoline inventories compared with previous years, the export volume also increased significantly in
the early stages of the epidemic, reaching its peak in April, 2020 [3]. As the COVID epidemic spreads
around the world, the export demand for gasoline has also decreased, falling by 60% in May. The
shrunk of global demand for gasoline has exacerbated the pressure of a domestic gasoline glut.
3.2. Price of Crude Oil
The price of gasoline depends largely on the price of crude oil. The COVID epidemic was not only a
huge blow to China, but also to all countries in the world. With the continuous spread of the epidemic,
airlines in various countries around the world have begun to reduce their flights, resulting in a
significant reduction in the demand for transportation oil. The sharp drop in global demand for crude
oil and excess crude oil reserves in oil-producing countries have led to a dramatic decline in global
oil prices [4]. Therefore, the cost of refining gasoline also became lower, and the price of gasoline
fell again.
3.3. The Political Situation
As one of the most important energy sources in the world, oil supply and price are deeply affected by
the international situation. In February 2022, the conflict between Russia and Ukraine continued to
intensify, culminating in a war. The initial stage of the outbreak of the war caused panic in the energy
market, and international oil prices began to rise, so gasoline prices also rose. As an oil-producing
country, Russia’s oil exports have been affected by ongoing wars and international sanctions [5]. The
global supply of oil has decreased, so the price of gasoline has soared since the beginning of the war,
and reached its peak in June 2022.
3.4. Environmental Protection Policies
In recent years, China has continuously promoted environmental protection policies, advocating
energy conservation and emission reduction, and using green new energy. For traditional energy,
which currently occupies a large share of the market, the government imposed high taxes on
companies that produce gasoline, which led to higher production costs, so gasoline prices rose
accordingly [6]. This is the reason that even though the Russia-Ukraine conflict has ended, the
Chinese gasoline price remains at 7.5 CNY/L and continues to fluctuate without falling nowadays.
As the Chinese government pays more attention to the development and application of new energy
sources, gasoline prices will only rise.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231052
121
4. Suggestions on Stabilizing Gasoline Prices
4.1. Strengthen Environmental Policies
When there is a high demand for gasoline, the price of gasoline will rise accordingly. But once people
find alternatives to gasoline and reduce the demand, the price of gasoline will fall and become
manageable. To achieve this goal, the government can formulate policies to support the development
of renewable energy, such as providing tax incentives, subsidies, and loan guarantees to companies
that develop new energy. This will help promote the technical research and market application of
renewable energy sources such as solar energy, wind energy, and water energy [6]. Moreover, the
government can invest in the construction of green infrastructure, such as renewable energy power
generation facilities, increasing the proportion of renewable energy in the energy supply and reducing
dependence on traditional energy.
In order to reduce people’s dependence on gasoline and reduce the carbon emissions of
transportation, the promotion of electric vehicles is a good solution. The government can provide
support, such as car purchase subsidies and charging facility construction, to encourage consumers
and businesses to adopt electric transportation vehicles. In addition, they can establish partnerships
with new energy vehicle manufacturers to jointly promote the development and popularization of
electric vehicles. When electric vehicles become popular, gasoline will no longer be the dominant
energy source in the transportation market, so the price will therefore go down.
4.2. Strengthen International Cooperation
Since rising gasoline prices are a serious global problem, countries can cooperate internationally to
control rising prices. Countries can promote import and export trade to ensure that each country’s
demand for oil is met to the greatest extent, which can solve the problem of a certain area’s oil being
in short supply and a certain area’s oil reserves being too large at the same time. This approach ensures
that each country imports enough oil to meet demand, thereby preventing the price of oil storage from
skyrocketing due to shortages. Moreover, countries can establish a shared strategic oil reserve to deal
with market tensions when prices fluctuate sharply. Keeping oil prices under control ensures that
there will be no sudden spikes in gasoline prices.
Conflicts and wars between countries will lead to sudden increases in oil prices, as the conflict
between Russia and Ukraine last year was a good example. In order to avoid the recurrence of this
situation as much as possible, countries can promote diplomacy, consultation and cooperation, and
actively negotiate when they have different opinions on international events, so as to avoid further
escalation of conflicts. This can minimize the sharp rise of oil prices caused by political events and
keep the prices as stable as possible, which has a great effect on stabilizing gasoline prices within a
certain fluctuation range.
5. Conclusion
This article studies the factors causing the volatility of gasoline prices in China. After the above
research, it is concluded that the COVID epidemic was the biggest reason for the decline in gasoline
prices in 2020. Not only was the domestic demand tightening, but the global spread of the epidemic
has led to a decline in global demand. The global oil supply far exceeded demand, so the cost of
gasoline refining has dropped, and the price has also fallen to the lowest point in many years. The
Russia-Ukraine conflict in 2022 was the biggest factor leading to the rise of gasoline prices. The war
brought panic to the energy market, and Russia was subject to international sanctions as an oil
exporter, which led to a reduction in exports. The shortage of supply led to an increase in oil prices,
so gasoline prices increased as well. In addition, China’s policy of continuously promoting the
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231052
122
development and application of green new energy and imposing higher taxes on gasoline companies
has kept gasoline prices at a relatively high level and will not fall back. These are the factors that led
to the bottom and peak of gasoline prices in the past four years.
In order to stabilize gasoline price fluctuations, this article proposes two feasible solutions. One is
to strengthen environmental protection policies by promoting new energy technologies and
subsidizing new energy vehicle companies, thereby reducing the demand for gasoline. The second is
to strengthen international cooperation by encouraging oil imports and exports, stabilizing
international relations to avoid conflicts, and reducing sudden increases in oil prices. Both of these
proposals would lead to more stable gasoline prices, and stable gasoline prices would reduce
economic panic and promote national economic development.
This article has some deficiencies in the source of data, because the gasoline price data released
by the Chinese government in recent years has not been found, so the data can only be extracted from
unofficial websites. And because there are slight differences in gasoline prices in each region of
China, it is impossible to make all the statistics on the gasoline prices of the 31 inland provinces, so
the price of No. 92 gasoline in Beijing is selected as an example. Although the price fluctuation trends
of No. 92 gasoline, No. 89 gasoline, and No. 95 gasoline are exactly the same, there is still a difference
of about 0.5 CNY/L between the prices. If the data on the three gasolines can be summarized and
discussed together, it will make this article more convincing.
The volatility in gasoline prices that characterizes these extraordinary times of COVID is not
universal. At present, the world has come out of the epidemic period, and the future trend of gasoline
prices is the most worthy of attention and research.
References
[1] National Oil Price Data - Data Center , Oriental Fortune.com, data.eastmoney.com/cjsj/oil_city.html. Accessed 4
Sept. 2023.
[2] Norouzi, Nima, et al. “When pandemics impact economies and climate change: Exploring the impacts of covid-19
on oil and electricity demand in China.” Energy Research & Social Science, vol. 68, 2020, p. 101654,
https://doi.org/10.1016/j.erss.2020.101654.
[3] Albulescu, Claudiu. “Coronavirus and oil price crash.” SSRN Electronic Journal, 2020,
https://doi.org/10.2139/ssrn.3553452.
[4] Wu Lei China (2020) “Analysis of the Oil Crisis and Its Impact Under COVID”, 10.19422/j.cnki.ddsj.2020.06.003
[5] Zhang, Qi, et al. “Unveiling the impact of geopolitical conflict on oil prices: A case study of the russia-ukraine war
and its channels.” Energy Economics, vol. 126, 2023, p. 106956, https://doi.org/10.1016/j.eneco.2023.106956.
[6] Yuan, Xueliang, et al. “The development of New Energy Vehicles for a sustainable future: A Review.” Renewable
and Sustainable Energy Reviews, vol. 42, 2015, pp. 298–305, https://doi.org/10.1016/j.rser.2014.10.016.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231052
123
A Case Study of Bright Dairy’s Acquisition of Israel’s Tnuva
Foods
Rui Zhong1,a,*
1Finance, Shanghai University of Finance and Economics, Shanghai, 200010, China
a. 2169348825@qq.com
*corresponding author
Abstract: In recent years, with the development of China’s economy and the policy
encouragement of the Chinese government, more and more Chinese enterprises have
participated in cross-border acquisitions and mergers and acquisitions. Studying the cases of
cross-border mergers and acquisitions of large Chinese enterprises can help reveal the impact
of similar mergers and acquisitions on the short-term stock price trend of enterprises and their
financial indicators. This paper chooses Bright Dairy Food’s acquisition of Tnuva as the main
body of the case study. By constructing the FAMA-French Three-Factor Model, this paper
studies the short-term effect the acquisition activity exerted on the stock price of Bright Diary
Food. By studying the financial indicators, this paper studies the influence of M&A activities
on the profitability and debt-paying ability of Bright Dairy Food. The research shows that
Bright Dairy’s decision to acquire Tnuva aroused widespread concern from shareholders, and
shareholders were optimistic about this decision, so the abnormal return of Bright Dairy’s
stock has increased in the short term. In addition, the acquisition of Tnuva has improved the
profitability of Bright Dairy but reduced its debt-paying ability.
Keywords: Bright Dairy Food, merge and acquisition, abnormal return, profitability, debt-
paying ability
1. Introduction
Foreign scholars have conducted mature theoretical research on the motivations of transnational
acquisitions, believing that companies are driven by various interest factors when making acquisition
decisions, including acquiring new resources for strategic integration, purchasing new technologies
to enhance strength, and increasing brand awareness. Lee conducted a study on the motives of
corporate acquisitions, building a cross-border acquisition model based on data from corporate
acquisition cases from 1985 to 2007. He then selected data from companies in developed countries
around the world for acquisition activities to test the model. Ultimately, the study found that if
companies make cross-border acquisitions to expand the market, companies in developed countries
are the preferred acquisition targets [1]. Ramsin et al. believe that cross-border mergers and
acquisitions can help both parties strategically restructure their corporate resources [2]. For the study
of how to evaluate acquisition performance, scholars have explored various value evaluation
application methods. Different methods can obtain different research results under the limitations of
research methods and backgrounds [3][4][5]. Therefore, the evaluation methods should be selected
according to the actual situation of companies in different case studies. At present, most of the
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231078
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
124
research on Chinese enterprises’ overseas M&A focuses on the risks and synergies of M&A, and the
industries studied are mainly the energy industry and traditional manufacturing industries, while there
are few studies on the dairy industry [6][7][8]. In 2015, Bright Dairy completed the acquisition of
Israel’s Tnuva Food Company, which is the largest cross-border merger and acquisition in China’s
dairy industry. The study on the performance of Bright Dairy’s acquisition of Tnuva can help predict
the impact of this type of transnational acquisition on the company’s stock price and profitability. At
the same time, this case study can provide a reference for Chinese dairy enterprises to conduct
overseas mergers and acquisitions, helping them to prevent financial risks [9][10].
2. Background
2.1. Background Information of Bright Dairy Food and Tnuva
Bright Dairy Food was established in 1996. In 2000, Bright Dairy completed its shareholding reform
and was listed on the Shanghai Stock Exchange. Enterprise capital is composed of state-owned capital
and social capital. As one of the giants in China’s dairy industry, Bright Dairy has diversified products,
rich marketing channels, and an excellent brand reputation. The core business of the group consists
of modern agriculture, food manufacturing, chain business, and trade, with the food industry chain as
the core. Tnuva Foods is the largest food company in Israel. It was originally founded as an
agricultural cooperative and has been established for more than 85 years. Tnuva Foods is an Israeli
integrated food company with a monopoly advantage, focusing on milk and dairy products. At present,
the group’s business includes the production, processing, and sale of a variety of food products,
including dairy products, meat products, pastries, etc. The company has a local market share of more
than 50% in Israel and is also widely involved in the Middle East, Europe, and the United States.
2.2. Motivations of the Acquisition
Transnational merger and acquisition is an important step in Bright Dairy’s long-term strategic
development. In 2015, Bright Dairy set its sights on the Middle East market, choosing one of Israel’s
oldest and most iconic food companies as its target. Israel has a very high degree of agricultural
modernization and has unique advantages in the development of agriculture and animal husbandry.
Tnuva Food, as an excellent supplier of milk powder, can provide high quality and reliable raw
materials for Bright Dairy to develop the powder market. At the same time, Tnuva Food has a unique
agricultural and animal husbandry industry, which is reflected in mature breeding technology and
high-quality dairy cows. In addition, Tnuva also has world-leading technology for producing organic
yogurts, cheeses and butters. Bright Dairy takes this opportunity to obtain advanced production
technology and upgrade the industrial chain.
3. Performance Analysis
On November 12, 2014, Bright Group established Bright Singapore Company in Singapore. Its main
business is to assist Bright Food Israel Limited Partnership in completing the acquisition of 100%
equity in AP.MS.TN and 0.01% equity in T.A.M. Milk. On March 31, 2015, Bright Dairy planned to
acquire 100% equity in Singapore Holdings through the non-public issuance of A-shares in order to
indirectly own the equity of Tnuva Group. In this acquisition, 56% of the shares held by Apax Partner,
a shareholder of Tnuva Food Company, and 21% of the shares held by Mivtach Shamir were acquired
by Bright Dairy Food. The market value of the acquired Tnuva reached 2.5 billion US dollars, or
approximately 15.3 billion RMB. In order to analyze the short-term and long-term effects of the
acquisition activity on Bright Diary Food, this paper tries to use the FAMA-French three-factor model
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231078
125
for short-term performance analysis and the financial indicators method for long-term performance
analysis [11][12].
3.1. Short-term Performance Analysis
Usually, after a listed company releases a major event announcement in accordance with the
requirements of the stock exchange, market investors will quickly judge the possible impact of the
announcement event, and the reaction of investors will be reflected in the recent stock price
fluctuations of the company, thus affecting the stock return rate. Event study is a commonly used
empirical method to evaluate enterprise performance. Based on the market efficiency theory, the
market can quickly and effectively reflect the value of the company. Therefore, the short-term
performance evaluation of the acquisition plan is carried out through the event study method, and the
investors’ attitude toward the acquisition event is illustrated according to the reaction of the stock
market to the share purchase arrangement of Bright Dairy Food.
3.1.1. Determining the Event Day and the Window Period
The first step in applying the event study method is to determine the event day and period of the study.
The event day does not necessarily refer to the day on which the event occurred, but the day on which
the event was definitively informed to the public. On March 30, 2015, Bright Israel Partnership paid
all the equity transfer payments to the original equity holders of AP.MS.TN and T.A.MMilk and
completed the necessary property transfer procedures. Bright Israel Partnership actually controlled
the financial and operating policies of Tnuva Group, enjoyed the corresponding interests, and
assumed the corresponding risks. Bright Diary Food was suspended from March 7, 2015, to June 8,
2015, because of restructuring matters related to the acquisition of Tnuva, and resumed trading on
June 9, 2015. In order to ensure the accuracy and reliability of the results, the window period excluded
the market trading day during the suspension period without any trading data. The event day is
determined to be June 9, 2015.
Secondly, the estimation window period generally selects a period of time before the event occurs
and is used to calculate the expected normal rate of return of the company when the event does not
occur. This paper selects February 24, 2014, to March 6, 2015 as the estimated window period of the
study, including 253 trading days. In addition, the event window is between June 9, 2015 and July 7,
2015.
3.1.2. Calculation of Expected rate of Return
If the market is efficient, the event is unexpected, and the occurrence of the event is related to the
stock value of a specific company, the abnormal stock return can be calculated by subtracting the
estimated stock return rate from the actual stock return rate. The result of this calculation can also
help to analyze whether stock holders pay much attention to this M&A event.
Stock prices are not only affected by the overall market economy but also by many other exposures,
so this paper uses the FAMA-French three-factor model for long-term performance analysis. Besides
considering market risk, the FAMA-French three-factor model also analyzes the impact of the scale
factor and the book market value factor. Therefore, the formula used to estimate the expected rate of
return by the market model is expressed as:
(1)
RiskPremiumt is the market risk premium factor on day t; SMBt is the Scale factor on day t; HMLt
is the Book value factor on day t.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231078
126
Collect the data on individual stock return rate during the estimation window period. According
to the parameters listed above, the relevant data for Bright Dairy Food stock during the estimation
window period from February 24, 2014, to March 6, 2015, were collected through the single stock
return rate (day) of the Csmar database and the index (day) of the Fama-French three-factor model.
Three indexes — the market risk premium factor, scale factor, and book market value factor — were
selected as independent variables of the regression model, and the expected rate of return was selected
as a dependent variable to carry out linear regression work.
Table1: The regression result of the Fama-French three-factor model.
Variable
Coefficient
Std.Error
t-Statistic
Prob
RiskPremium
0.712228
0.106253
6.703117
0.0000
HML
-0.980885
0.233936
-4.192968
0.0000
SMB
-0.273436
0.184399
-1.482850
0.1394
C
-0.000809
0.001233
-0.655788
0.5126
As shown in Table1, the regression coefficient of RiskPremium is about 0.712 (t=6.703,
p=0.000<0.01), which means that RiskPremium has a significant positive influence on the stock
return rate. The regression coefficient of HML was approximately -0.981 (t=-4.193, p=0.000<0.01),
indicating that HML had a significant negative effect on Rt. However, the regression coefficient of
SMB is -0.273 (t=-1.483, p=0.139>0.05), which means that SMB has an insignificant influence on
Rt. Therefore, according to the results of the significance level, the SMB factor is removed and the
expected return rate is obtained, as shown in the equation.
3.1.3. Calculation of the Abnormal Return
According to the formula in 3.1.2, the estimated daily return of the stock of Bright Dairy Food in the
event window period can be calculated. Then, the author gets the actual daily return of Bright Dairy
Food on these days from the Csmar database. The abnormal return (AR) of each day is the difference
between the actual return and the estimated return. Accumulate the value of the abnormal return to
obtain CAR. If the value of CAR is positive, it indicates that acquisition activities have a positive
effect on the company’s stock price.
3.1.4. Analysis of Short-term Performance
As shown in Figure1, in the 20 days of the time window, the CAR value was always greater than zero
and showed an overall upward trend, from 0.105 on day 0 to 0.364 on day 20. The news that Bright
Dairy Food is about to successfully acquire Tnuva has brought positive abnormal returns to Bright
Group’s stock in the short term. The acquisition event increased the stock price in the short term. This
result shows that the shareholders of Bright Group were very concerned about the acquisition event
and generally held a positive attitude toward it.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231078
127
Figure1: CAR of the stock of BDF in the 20 days after the event day.
3.2. Long-term Performance Analysis
This paper uses the financial index method to analyze the long-term impact of the acquisition event
on Bright Dairy Food. This trinational acquisition event had a long time span, and the impact of it on
the performance of the company could be reflected over a long period of time. This paper selected
the period of two years before and three years after the completion of the acquisition events, a total
of five years, and comprehensively analyzed the impact of changes in various financial indicators on
the long-term financial performance of Bright Dairy Food.
3.2.1. Profitability
Table 2: Financial indicators of the profitability of BDF between 2013 and 2017.
Year
2013
2014
2015
2016
2017
EPS
0.33
0.46
0.34
0.46
0.50
Net profit margin on
sales
2.91%
2.86%
2.56%
3.34%
3.77%
Return on equity
9.45%
11.21%
9.43%
10.96%
12.24%
Return on total assets
4.10%
4.53%
3.21%
4.20%
4.95%
As shown in Table 2, the earnings per share of Bright Dairy had a relatively obvious decline in the
year when the merger and acquisition occurred from 0.46 in 2014 to 0.34 in 2015, and showed a
gradual upward trend in the two years after the completion of the merger. The EPS of BDF reached
0.50 in 2017, exceeding the EPS of the year before the acquisition. The decline in earnings per share
in 2015 was due to the fact that the acquisition was paid for in the form of a private offering of stock,
which somewhat diluted earnings for existing shareholders. Since then, the annual earnings per share
have gradually increased, indicating that the merger and acquisition have brought good profitability
to the enterprise over a long period of time.
The net profit rate on sales, return on equity, and return on total assets of Bright Dairy all
experienced a decline in 2015, indicating that the acquisition activities had a certain degree of impact
on the cost and profit of Bright Dairy, resulting in a decline in the profitability of Bright Dairy’s assets
in that year. In 2015, the entire dairy industry’s raw milk price fluctuations, competition in the same
industry, and other external environmental factors led the overall cost of China’s dairy industry
enterprises to rise. From this point of view, merger and acquisition activities are the main but not the
only reason for the reduction in profitability at Bright Dairy. In 2016, after the merger, all three values
began to rise, indicating that the negative impact of the merger had dissipated. For instance, the net
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231078
128
profit margin on sales of BDF increased from 2.56% in 2015 to 3.77% in 2017. In addition, the return
on equity reached 12.24% in 2017, the highest in the five years before and after the M&A study.
From the overall data of the past five years, the profitability of Bright Dairy was affected by merger
and acquisition activities and some external factors, which led to a temporal decrease of some
financial indicators in the year of the merger, but the indicators increased in the following year. On
the whole, all indicators were stable and improved, indicating that the merger and acquisition activity
had a positive effect on the profitability of the company.
3.2.2. Debt-paying Ability
Table 3: Financial indicators of the debt-paying ability of BDF between 2013 and 2017.
Year
2013
2014
2015
2016
2017
Asset-liability ratio
56.57%
59.64%
65.93%
61.69%
59.60%
Equity ratio
43.43%
40.36%
34.07%
38.31%
40.40%
Long-term debt-paying ability mainly refers to debt repayment ability with a payment term of one
year or more. The higher the asset-liability ratio, the more assets the enterprise obtains by borrowing,
which may lead to certain financial risks. It is generally believed that the asset-liability ratio of
enterprises had batter to be between 40% and 60%. As can be seen from Table 3, the asset-liability
ratio of Bright Dairy showed an upward trend from 2013 to 2015, increasing from 56.57% in 2013 to
65.93% in 2015. Although the value decreased to 61.69% in 2016, the year after the acquisition, it
was still higher than the asset-liability ratio before in 2013 and 2014. The equity ratio is a measure of
how much of a business’s assets are invested by its owners. According to table 3, the overall
shareholder equity ratio of Bright Dairy decreased to 34.07% in 2015, indicating that the amount of
debt borrowed by the enterprise in the process of using financial leverage to expand business activities
has increased compared with that before the acquisition. From the asset-liability ratio and
shareholders’ equity ratio, the long-term debt-paying ability of Bright Dairy Food has decreased
compared with that before the acquisition.
4. Conclusion
Bright Dairy and Tnuva are both large dairy production companies, and many of their products and
production lines are similar. Therefore, Bright Dairy’s decision to acquire Tnuva will help the
company obtain high-quality raw materials, expand overseas markets, and solve the problem of
overcapacity. In addition, as one of the largest food production companies in Israel, Tnuva has leading
technology patents in the field of milk powder production, so the acquisition of Tnuva can provide
strong technical support for Bright Dairy, helping it enter the middle-to-high-end milk powder market.
From a short-term perspective, the stock of Bright Dairy Food had a positive abnormal return during
the event window, which shows that the acquisition event increased the stock price in the short term
and that the shareholders of Bright Group generally held a positive attitude toward the acquisition
event. From a long-term perspective, on the one hand, merger and acquisition activities have
improved the profitability of Bright Dairy, the owner of the merger. Whether the information of
earnings per share or return on assets shows an upward trend, it indicates that the sales channels of
products are broadened and the overall sales and profits of enterprises are steadily improving. On the
other hand, after the merger, the debt-paying capacity of Bright Dairy was slightly weakened. In
conclusion, shareholders were optimistic about Bright Dairy’s decision to acquire Tnuva, and Bright
Dairy’s profitability indeed improved after the acquisition. The data used in this article is financial
data disclosed in domestic and foreign public markets. In data analysis, this paper uses the average
method to ensure the objectivity of the data, but the quality of the data is inevitably flawed. In addition,
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231078
129
there is a problem of insufficient sample size in the linear regression. In order to further improve the
research, the sample size can be expanded to select more time points that affect M&A activities and
shareholders’ reactions.
References
[1] Lee D. Cross-Border Mergers and Acquisitions with Heterogeneous Firms: Technology vs. Market Motives [J]. The
North American Journal of Economics and Finance, 2017, 42: 20–37.
[2] Ramsin Y, Richard N, Patrik S. Chinese foreign acquisitions aimed for strategic asset-creation and innovation
upgrading: The case of Geely and Volvo Cars [J]. Technovation, 2018, 36: 70–71.
[3] (Katherine) K M,Jorge F. The impact of political freedoms on cross-border M&A abandonment likelihood [J].
Quarterly Review of Economics and Finance, 2023, 91.
[4] Ping L,Boqiang L,Mengting L, et al. Empirical Study of Factors Influencing Performance of Chinese Enterprises
in Overseas Mergers and Acquisitions in Context of Belt and Road Initiative — A Perspective Based on Political
Connections [J]. Emerging Markets Finance and Trade, 2019, 56(7).
[5] Yao C. Integration and technology innovation in technology-sourcing M&As — A comparative study on overseas
and domestic M&As of Chinese enterprises [J]. Asian Journal of Technology Innovation, 2022, 30(3).
[6] You T. Research on Motivation and Performance of Chinese Enterprises’ Overseas M&A under the background of
“One Belt and One Road” [D]. Central University of Finance and Economics, 2022.
[7] Bing L. Case study of Harbin Pharmaceutical Group’s acquisition of GNC (Jiananxi) in the United States [D].
Shenyang University of Technology, 2021. DOI:10.27322/d.cnki.gsgyu.2021.001007.
[8] Xiangxiang H,Yue S,Yunwen X. Valuation Methods in Case of Merges and Acquisitions: A Review [P]. 2022 7th
International Conference on Financial Innovation and Economic Development (ICFIED 2022), 2022.
[9] Kaixin L. Analysis on Risk of Debt of Leveraged Financing in Cross-border M&A [J]. The Frontiers of Society,
Science and Technology, 2023, 5(7).
[10] Nandan V. M&A: The Impact of M&A on Emerging Companies [J]. Journal of Global Economy, Business and
Finance, 2023, 5(2).
[11] Ke N. Analysing the Financial Synergy Effect of Corporate Mergers and Acquisitions — Yili’s Acquisition of
Ausnutria [J]. Research in Economics and Management, 2023, 8(3).
[12] Liu Y. Analysis of the Path of State-owned Enterprises’ Merger and Acquisition of Private Enterprises under the
“Management Committee + Company Model” — Take the Acquisition of Company A as an Example [J]. Academic
Journal of Business & Management, 2023, 5(14).
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231078
130
Research on Zara’s Social Media Marketing Strategy in the
Context of New Media
Jiayi Wang1,a,*
1Institute of Communication, Culture, Information and Technology; University of Toronto
Mississauga, Mississauga, L5L 1C6, Canada
a. wjyyy.wang@mail.utoronto.ca
*corresponding author
Abstract: The digital revolution, characterized by widespread internet access and the
burgeoning influence of social media, has heralded a new era of consumer engagement and
transformed branding paradigms. Particularly impacted is the fashion domain, with brands
navigating the tumultuous waters of dynamic trends and digital preferences. This paper delves
into Zara’s foray into this intersection of fashion and digital marketing, illuminating its
triumphs in social media marketing (SMM) and underexplored areas. Despite Zara’s
commendable utilization of platforms such as Instagram, a glaring bidirectional
communication gap needs to improve the establishment of authentic consumer connections.
The advent of Generation Z accentuates this, introducing nuanced digital consumption
behaviors that demand a revised SMM approach. Recommendations proffered include
enriching content interactivity, fortifying influencer collaborations, and calibrating strategies
tailored for Generation Z. Emphasizing the vitality of agility in branding, the study
underscores the necessity for brands, even those at the pinnacle of their sectors, to perpetually
evolve in harmony with the digital zeitgeist. Through a detailed exploration of Zara’s digital
endeavors, this research offers an instructive lens on the intricacies of modern digital
consumerism, charting a direction for the fashion industry in the age of pervasive digital
connectivity.
Keywords: social media marketing, marketing strategy, Zara, digital consumerism
1. Introduction
1.1. Research Background
In the contemporary digitized era, the interplay between fashion and marketing undergoes profound
transformation due to the emergence of new media platforms. As the delineation between digital and
tangible realities becomes increasingly nuanced, a brand’s aptitude in utilizing the dynamic realm of
social media emerges as a vital determinant of its market presence [1]. The digitization of consumer
experiences has precipitated fundamental shifts in how fashion brands communicate, engage, and
transact. Unlike traditional marketing channels, which propagate a linear and controlled brand
narrative, new media platforms, mainly social media, champion a dialogic and dynamic exchange.
Here, consumers metamorphose from mere recipients to active contributors, influencers, and,
occasionally, brand ambassadors [1]. As businesses navigate this digital transformation,
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231088
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
131
understanding the nuances, challenges, and opportunities presented by social media marketing
becomes paramount. Zara, a paradigm in the fast fashion industry, exemplifies this transition by
seamlessly adapting to this evolving landscape. Zara’s digital strategy, characterized by agile
adaptability and an intrinsic understanding of its demographic, illuminates how modern brands can
simultaneously foster engagement, loyalty, and commerce. Exploring Zara’s social media tactics is a
microcosm of broader trends in fashion marketing and consumer behavior. Consequently, an in-depth
analysis of such practices provides valuable insights for commercial strategies and elucidates the
nuanced relationship between brands and their audience in this digital age. The significance of this
inquiry extends beyond its commercial relevance, offering a comprehensive understanding of the
transformative nature of brand-consumer dynamics in the context of new media. It also analyzes the
potential problems of social media marketing strategies and provides solutions.
1.2. Literature Review
The dawn of the digital age brought about a paradigm shift in the realm of marketing, driving
businesses towards leveraging online platforms to engage with their consumer base. Of these
platforms, social media stands as a formidable tool for marketers, offering an interactive and dynamic
environment for businesses to promote their brands and foster deeper relationships with consumers.
As elucidated by Li, Larimo, and Leonidou, social media marketing (SMM) is a multifaceted entity
that goes beyond mere promotion. It encompasses activities that utilize social media platforms to
achieve communication and branding goals [2]. Their comprehensive research laid the groundwork
by conceptualizing and validating SMM’s strategic essence. Their taxonomy uncovers the symbiotic
relationship between SMM and business performance, emphasizing that a firm’s initiatives can
influence both the firm’s and its stakeholders’ actions and decisions. Taking the fashion industry as
a case in point, Kontu and Vecchi explored how fashion brands maneuver the noisy social media
landscape. Their investigation revealed that while all brands, regardless of their size, recognize the
importance of social media, the strategic value derived varies significantly. More prominent brands
are more inclined to tap into the potential of social media to foster community building and enhance
brand loyalty. In contrast, smaller brands prioritize direct sales, implying that the strategic intent
behind social media activities is contingent on the brand’s market positioning and overarching
objectives [3]. Constantinides delved into how social media has reshaped consumer behavior and
business strategies. Notably, social media has emboldened customers, ushering in demands for
product customization and a pronounced desire for active participation in product development, a
phenomenon termed co-creation [4]. These evolving customer behaviors mandate a strategic
realignment for marketers. Businesses are acknowledging this shift, as evidenced by increased
investments in social media endeavors [4]. The research underscores the myriad benefits of SMM,
from heightened market exposure to the establishment of new business partnerships and even cost
savings on traditional marketing channels. The correlation between a brand’s social media
engagement and its financial performance has also been highlighted, with heavily engaged brands
registering substantial revenue growth [4]. Constantinides also highlighted the necessity of ensuring
that foundational digital assets, like a corporate website, are optimized before embarking on advanced
SMM endeavors. The integration of Social Media into the broader marketing strategy is not an
isolated process but a culmination of strategic efforts, from product/service quality, organizational
readiness, and flawless web presence to effective engagement of the Social Media tools [4].
1.3. Research Framework
Zara’s proficiency in SMM has been the subject of extensive academic inquiry. Many studies have
been dedicated to decoding the brand’s unparalleled success in harmonizing its digital presence with
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231088
132
its commercial objectives. Particularly prevalent within these investigations is the emphasis on the
efficacy of Zara’s SMM strategies, illuminating a consistent positive correlation between the brand’s
adept use of social media and its increasing revenues.
However, the current literature landscape displays a conspicuous research gap. While there is
considerable appreciation for Zara’s strategic prowess, there remains a paucity of critical evaluations
addressing potential pitfalls in the brand’s SMM initiatives. In addressing this void, the present
research is structured around a two-fold framework:
(1) Zara’s Social Media Marketing Analysis: A deep dive into Zara’s SMM methodologies, tools,
platforms, and tactics. This will lay the groundwork by comprehensively understanding how Zara
engages with its online audience, leverages user-generated content and adapts to evolving digital
trends.
(2) Problem Identification and Solutions: Moving beyond the celebratory lens of existing literature,
this segment aims to evaluate Zara’s SMM strategies critically. Each potential issue will be followed
by tailored strategic recommendations, fortified by best practices and emerging trends in digital
marketing.
By juxtaposing Zara’s acclaimed strategies with a critique of potential shortcomings, this research
aspires to present a balanced, holistic perspective. The intention is to spotlight challenges and
proactively offer actionable solutions, with a vision to fortify Zara’s future SMM endeavors and
provide a nuanced template for other global brands navigating the intricate tapestry of new media
marketing.
2. Case Description
In the mid-1970s, Zara, a trailblazer in the fashion world, established its foothold with its inaugural
store in La Caruña, Spain. By 1988, the brand’s presence was felt internationally, as evidenced by its
store in Porto, Portugal, quickly followed by expansion into major fashion hubs—New York and
Paris. Zara’s emergence in the fashion sector was strategically timed [5]. As established players like
Gap and H&M grappled with legacy technologies, Zara seized the opportunity to integrate the latest
information systems without being bogged down by outdated infrastructures [5].
The brand’s ascent was not merely about tech-savviness. Crucial to its growth was a keen
understanding of market dynamics. In the 1980s, while lower labor costs in other countries lured
competitors, Zara discerned the importance of unit labor costs. This insight led them to capitalize on
Spain and Portugal’s competitive labor costs, giving them a significant edge [5]. This strategic move
allowed Zara to foster a robust local assembly line, with the early adoption of vertical integration—
ensuring control over production, distribution, and sales—further strengthening its position [5].
Zara’s business model veered from the traditional. While its competitors outsourced production to
regions with low labor costs, Zara retained much of its production within Spain and Portugal. This
move, which was counterintuitive considering the higher labor costs, was offset by reduced
advertising and inventory management expenses [5].
Nevertheless, Zara was not only about business strategies; its brand image was meticulously
cultivated. While affordable, Zara’s shopping experience needed to be closer to budget. Stores, often
spacious and strategically located, exuded an upscale ambiance. This positioning catered to
consumers looking for affordable yet stylish fashion. The brand mastered making fast fashion items
feel and look luxurious. To this end, Zara stores closely resemble luxury brands such as Esprit and
Club Monaco [5].
Moreover, Zara’s commitment to responsiveness over efficiency distinguished it from competitors.
By closely monitoring and swiftly adapting to fashion trends, Zara created a unique selling
proposition: contemporary, fast-fashion products available almost instantaneously [5]. Their unique
supply chain, characterized by frequent shipments and a rapid turnaround from design to store shelf,
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231088
133
ensured stores consistently offered fresh collections. This dynamism not only catered to existing
demand but also invigorated it, compelling customers to make purchases lest they miss out on limited-
run items.
Zara’s meteoric rise in the fashion industry can be attributed to its unique blend of strategic
positioning, understanding of market dynamics, efficient use of technology, and commitment to
delivering a premium shopping experience at affordable prices.
3. Analysis on the Problem
3.1. Digital Engagement and Zara’s SMM Approach
The digital landscape has witnessed exponential growth in the last few years. By January 2020, more
than half the global population, approximately 4.5 billion individuals, were connected online. A
staggering 3.8 billion people are actively engaged on social media platforms [6]. These platforms,
primarily accessed via mobile networks, have revolutionized how people communicate, gather
information, travel, and shop, significantly enhancing the quality of life. One cannot overlook the
evident surge in the reliance on social media for commerce. Compared to 2017, 2019 saw a significant
rise in consumers learning about, developing interest in, and eventually purchasing products through
social media [6]. Notably, 25% of those surveyed confirmed making purchases on social platforms,
an impressive increase from just two years prior. This trend prominently emerges in sectors like
clothing, makeup, and packaged foods – segments in Zara are deeply rooted [6]. This reliance on
social media is not limited to planned purchases alone. Social platforms have a compelling influence
on impulse buying behavior. Interaction with Key Opinion Leaders (KOLs), engagement with user-
generated content (UGC), and product recommendations from peers led 40% of those surveyed to
make spur-of-the-moment purchases [6]. These numbers underscore the profound impact of social
media content on consumption decisions. Zara, with its well-executed SMM strategies, is strategically
placed to capitalize on these shifts.
Firstly, Zara ingeniously leverages Instagram as a dynamic platform to showcase these updates.
The brand transforms its Instagram feed into a visual catalog, helping customers effortlessly access
and engage with the latest fashion trends. A testament to the success of this strategy is Zara’s
staggering follower count of over 40 million, with a significant portion of their content being video
posts, emphasizing the current inclination of users towards visual content [7].
However, it is not just the numbers that make Zara’s SMM noteworthy. Their strategy exhibits a
deep understanding of their audience and the changing dynamics of influencer marketing. While the
trend has been leaning towards celebrity endorsements, Zara pivots towards a more relatable strategy.
Instead of resorting to celebrity-driven campaigns, Zara chooses influencers that align more closely
with their brand and audience values. The “Timeless” campaign serves as a poignant example where
they partnered with fashion industry veterans to discuss aging’s impact on personal style, offering a
fresh perspective that contrasts with the youth-centric narratives often observed in fashion advertising
[7].
A significant breakthrough in Zara’s SMM is realizing the untapped potential of micro-influencers.
Contrary to traditional wisdom, which might dismiss these influencers due to their smaller following,
Zara recognizes the concentrated influence they can wield within their niche communities [7]. Their
collaboration with 522 out of 2421 micro-influencers reveals a significant shift in marketing
paradigms. The sheer impact of these collaborations is evident from their South African online store
launch campaign, where, thanks to micro-influencers, their branded hashtag reached over six million
individuals in a single day [7].
Zara’s current SMM strategy is multifaceted. It blends the strengths of its fast fashion model with
the dynamic nature of platforms like Instagram, leverages the power of both macro and micro-
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231088
134
influencers and focuses on authenticity and brand values over mere product advertisement. This
comprehensive approach is a blueprint for their continued dominance in both the fashion and digital
marketing arenas.
3.2. Potential Problems in ZARA’s SMM
3.2.1. Underutilized Engagement: Zara’s Bidirectional Communication Gap in Social Media
Marketing
Drawing from the insights of Navarro-Beltra et al., it becomes evident that while the potential for
building genuine and lasting connections with customers through social media is immense, Zara is
underutilizing these avenues [8]. The essence of SMM lies in its bidirectional nature, allowing for an
open dialogue between brands and consumers. However, Zara’s online interactions paint a picture of
limited engagement, leading to a perceived aloofness. This lack of two-way communication renders
the brand-customer relationship somewhat transactional, depriving Zara of opportunities to foster
deeper connections and, by extension, brand loyalty. To be more specific, Zara’s sparse use of
mentions and hashtags on social media platforms could limit its content discoverability and the
potential to engage with a broader audience [8]. Especially in the fast-paced world of fashion, where
influencers play a significant role, not leveraging mentions might mean missing out on a golden
opportunity for organic growth and engagement [8]. In addition, Navarro-Beltra et al. also pointed
out that Zara’s shortcomings in online interaction also involve a need for more response to consumer
reviews. Even with users actively interacting with Zara’s content, the brand is lacking in its responses
[8]. Bao has pointed out similar problems. Bao claimed that Zara’s communication with consumers
on social platforms in the Chinese market could be better [6]. Zara updates its product content on the
social platform every week, but there are only some consumer comments due to the boring content
published [6].
3.2.2. Overlooking the Potential of Generation Z in SMM Strategy
The demographic shifts in consumer markets present an ever-evolving challenge for brands,
particularly in capturing the attention and loyalty of newer generations. One such demographic that
is now gaining increasing importance, both in terms of numbers and buying power, is Generation Z.
Comprising individuals born between 1995 and 2009, Generation Z has grown up in a distinctly
digital era, characterized by the omnipresence of the mobile internet and the dominance of social
media in daily life [6]. This exposure has shaped not only their consumption patterns but also their
brand expectations.
In the Chinese context, the significance of Generation Z becomes even more pronounced, with a
staggering 260 million individuals falling into this category [6]. As of 2021, the age range of this
cohort spans from 12 to 26, which indicates that a large segment of them is now entering the
workforce, thereby boosting their purchasing power [6]. Their deep-rooted association with the digital
realm sets them apart from their predecessors, such as the millennials. Their browsing habits, content
consumption patterns, and brand expectations are intrinsically tied to online information’s rapid
influx and turnover. Self-identity plays a pivotal role in brand selection, and they are open about
vocalizing their societal viewpoints [6].
Bao pointed out that the primary target audience of Zara leans towards the 20-35 age bracket,
inadvertently sidelining the burgeoning under-20 segment of Generation Z [6]. Zara’s brand persona,
epitomized by a more Western style, may not resonate as deeply with Generation Z, who show an
inclination towards a blend of cultures, with particular enthusiasm for secondary cultures and
nationalistic styles [6].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231088
135
To genuinely tap into the vast potential of Generation Z, Zara needs to recalibrate its branding and
SMM strategies. This entails being present on the platforms frequented by this demographic, weaving
narratives, and producing content that aligns with their cultural preferences and values.
4. Suggestions
4.1. Enhancing Bidirectional Engagement: Bridging the Communication Gap in Zara’s
SMM
Zara should prioritize optimizing its content for increased engagement, aiming for interactive
elements such as polls, quizzes, and challenges that actively invite user participation and provide
deeper insights into consumer preferences. By tapping into currently trending topics and integrating
them with the brand’s distinct messaging, the potential for driving meaningful and resonant
interactions on social media platforms could increase significantly. Zara must establish a robust
feedback mechanism, which would involve diligently monitoring comments, reviews, and mentions
while ensuring timely, appropriate responses. An active acknowledgment of user feedback
demonstrates the brand’s commitment to its audience and acts as a pillar for fostering trust, ultimately
enhancing brand loyalty.
Sharma pointed out that in the realm of modern marketing, the rise of influencer marketing has
signified a paradigm shift [9]. With brands vying for consumer attention in an increasingly cluttered
market, influencer marketing offers a robust pathway to heightened brand awareness and engagement
[9]. In light of the undeniable influence of influencers within the fashion realm, Zara stands to gain
from meticulous influencer collaborations. By allying with influencers whose principles align with
the brand’s core values, Zara can seamlessly tap into pre-existing audiences, extending its reach
manifold. This approach and fresh content perspectives from influencers can render Zara’s SMM
endeavors more vibrant and engaging.
Furthermore, personalizing content based on user data can offer a solution to perceptions of
monotony in content. Tailoring posts and updates using algorithms that account for user preferences,
historical interactions, and purchase patterns can make for a more curated and engaging user
experience. Beyond the realm of mere product updates, Zara should seek to spark genuine
conversations about prevailing fashion trends, sustainability, or other pertinent discussions. Initiatives
like live sessions on popular platforms or in-depth Q&A sessions can infuse authenticity into Zara’s
SMM, enhancing user-brand affinity. Finally, investing in training sessions for the brand’s online
representatives will ensure they are adept at nurturing genuine interactions, deftly addressing
concerns, and projecting a favorable brand image. Adopting this holistic strategy can metamorphose
Zara’s SMM stance from a transactional perspective to a more relational one, capitalizing on the vast
potential of social media for potent brand building and sustained customer engagement.
4.2. Redefining Zara’s SMM for Generation Z: Strategies for a Digital-First Future
To comprehensively address the overlooked potential of Generation Z in Zara’s SMM strategy, a
meticulous reconfiguration of branding and marketing methodologies bespoke to this demographic is
pivotal. Özkan and Solmaz elucidate that Generation Z, predominantly emerging post-1995, is
distinguishably molded by their intricate digital integration [10]. This group has grown symbiotically
alongside the advent of technological marvels, becoming a generation that harmoniously assimilates
the Internet, social media, and a plethora of digital tools into their quotidian lives [10]. Notably,
technology does not merely influence their consumption preferences—it converges with their persona.
Evidently, research highlights Generation Z’s affinity towards brands that exhibit a reverence for
diverse cultures, with a pronounced inclination towards secondary and nationalistic styles. To align
with this sentiment, Zara should contemplate launching collections infused with cultural nuances and
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231088
136
design elements that specifically cater to Generation Z. Such innovative design trajectories can not
only mirror their aesthetic predilections but also echo their intrinsic cultural ethos, laying the
foundation for enhanced brand loyalty and affinity.
The distinct digital consumption modalities of Generation Z accentuate the imperative for Zara to
recalibrate its focus towards platforms that reverberate with this age bracket. It is key to understand
that Generation Z perceives the digital realm as an intrinsic extension of their physical world. Aptly
christened the “mobile generation,” their penchant for fluid digital engagements is evident [10]. This
proclivity permeates their shopping behaviors; before purchase commitments, Generation Z is
markedly inclined to embark on exhaustive online research [10]. Transitioning from the prevalent
platforms that primarily cater to the 20-35 age bracket, it is incumbent upon Zara to amplify its
presence on platforms that are the epicenter for younger cohorts, such as transient video platforms,
nascent gaming realms, and other digital spheres gaining traction among Generation Z.
A recalibrated content strategy that acknowledges the swift ephemerality of online information is
quintessential. This translates into crafting succinct, visually arresting content and structured for
instantaneous engagement for Zara. Another avenue, inspired by Generation Z’s esteem for
authenticity and co-creation, is the strategic deployment of User-Generated Content (UGC).
Launching campaigns that allow them to vocalize their fashion perspectives, societal standpoints, or
design inspirations can amplify Zara’s digital footprint exponentially while aligning its brand persona
with paramount values such as inclusivity and cultural veneration.
The resonance of Social Media Influencers must be noticed [9]. Given their elevated relatability
quotient compared to traditional celebrities, especially within younger demographics, forging
alliances with influencers who predominantly cater to Generation Z can serve as an authentic conduit
for Zara’s brand ethos. Concludingly, integrating the dynamic interactivity inherent in SMM, Zara
ought to design feedback infrastructures specifically tailored for Generation Z. By fostering platforms
where this demographic can proactively engage, share insights, and even co-steer the brand’s
trajectory, Zara not only ensures adaptive market responsiveness but also nurtures a profound sense
of brand stewardship and communal interaction. By adeptly weaving these solutions into its strategy,
Zara holds the promise of effectively mending its current detachment from Generation Z, capitalizing
on their emergent purchasing prowess and idiosyncratic digital propensities to unveil untapped
market vistas.
5. Conclusion
5.1. Summary
The advent of the digital age, marked by the ubiquity of the internet and the pervasive influence of
social media, has profoundly reshaped consumer behavior and brand strategies. With most global
population connected online and actively engaged on social media platforms, brands have been forced
to reimagine their approaches to stay relevant. The fashion industry, characterized by its dynamic
nature and rapidly changing trends, has been particularly affected. Zara, a global fashion powerhouse,
has strategically positioned itself at this crossroads of fashion and digital marketing, capitalizing on
consumer behavior and preference shifts.
Despite Zara’s monumental success in leveraging social media marketing (SMM) to drive
engagement and sales, there are untapped potential and challenges. While the brand effectively uses
platforms like Instagram to showcase its offerings and engage its audience, a bidirectional
communication gap persists. This underutilization hampers the potential for building genuine and
lasting connections with consumers. Furthermore, demographic shifts, particularly the emergence of
Generation Z, present both an opportunity and a challenge. Addressing this demographic’s unique
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231088
137
digital consumption patterns and expectations is crucial for sustained success in an increasingly
competitive landscape.
By optimizing content for increased engagement, prioritizing interactive elements, and redefining
branding strategies to resonate with Generation Z, Zara has the potential to address its current SMM
challenges and solidify its position as a leader in both the fashion and digital marketing arenas. It is
crucial for brands, irrespective of their current market stature, to continuously evolve and adapt in
response to the ever-changing digital landscape. Only then can they ensure sustained success and
relevance in a world dominated by digital interactions and rapidly shifting consumer preferences.
5.2. Research Significance
This analysis offers invaluable insights into the intersections of fashion and digital marketing,
charting a roadmap for industry players. It underlines the successes and pitfalls that even established
brands can encounter in the digital domain. Understanding these dynamics can be the difference
between sustained growth and stagnation for businesses in the fashion sector, especially those looking
to expand their digital footprint. By highlighting Zara’s approach and potential areas of improvement,
this paper contributes to a deeper understanding of the complexities in the age of digital consumerism,
providing guidance for those aiming to harness the full power of social media marketing in the fashion
industry.
References
[1] Chen, S.-C., & Lin, C.-P. (2019). Understanding the effect of social media marketing activities: The mediation of
social identification, perceived value, and satisfaction. Technological Forecasting & Social Change, 140, 22–32.
https://doi.org/10.1016/j.techfore.2018.11.025
[2] Li, F., Larimo, J., & Leonidou, L. C. (2021). Social media marketing strategy: definition, conceptualization,
taxonomy, validation, and future agenda. Journal of the Academy of Marketing Science, 49(1), 51–70.
https://doi.org/10.1007/s11747-020-00733-3
[3] Kontu, H., & Vecchi, A. (2014). Why all that noise - assessing the strategic value of social media for fashion brands.
Journal of Global Fashion Marketing, 5(3), 235–250. https://doi.org/10.1080/20932685.2014.912443
[4] Constantinides, E. (2014). Foundations of Social Media Marketing. Procedia - Social and Behavioral Sciences, 148,
40–57. https://doi.org/10.1016/j.sbspro.2014.07.016
[5] Tokatli, N. (2008). Global sourcing: insights from the global clothing industry—the case of Zara, a fast fashion
retailer. Journal of Economic Geography, 8(1), 21–38. https://doi.org/10.1093/jeg/lbm035
[6] Bao, D. (2022). ZARA’s social media marketing strategy optimization research. Donghua university, 2022. DOI:
10.27012/d.cnki.gdhuu.2021.001483.
[7] Martínez-López, F. J., Li, Y., & Young, S. M. (2022). Cases and Analyses of Companies Leveraging Social Media.
In Social Media Monetization (pp. 115–134). Springer International Publishing. https://doi.org/10.1007/978-3-
031-14575-9_9
[8] Navarro-Beltrá, M., Medina, I. G., & Correia, P. A. (2020). The dialogical potential of Facebook: The case of
fashion brands. International Journal of Interactive Mobile Technologies, 14(4), 30-43.
[9] Sharma, D. (2023). How not who: Message strategies adopted by celebrities v/s social media influencers. Journal
of Marketing Communications, ahead-of-print(ahead-of-print), 1–25.
https://doi.org/10.1080/13527266.2023.2220326
[10] Özkan, M., & Solmaz, B. (2017). Generation Z-the global market’s new consumers-and their consumption habits:
Generation Z consumption scale. European Journal of Multidisciplinary Studies, 2(5), 222-229.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231088
138
An Analysis of Sexy Tea’s Marketing Strategy and Its Impact
on Brand Communication of New Chinese Milk Tea in
China
Shihan Guo1, a, *
1Department of Journalism, Wuhan Sports University, Wuhan, China
2020332138@whsu.edu.cn
*corresponding author
Abstract: In today’s landscape, the milk tea market in China is experiencing an unprecedented
phase of rapid expansion and remarkable development. Milk tea shops have proliferated
across cities and towns, to the extent that ordering takeout without a side of milk tea has
become a rarity. The diverse range of milk tea offerings has seamlessly woven itself into the
fabric of daily life for countless individuals. Among the standout pioneers in this burgeoning
industry is the Sexy Tea (Chayan Yuese) brand, which emerged in 2014 and promptly
skyrocketed to fame. Sexy Tea introduced a unique twist to the conventional milk tea
experience by infusing a subtle note of bitterness and the captivating aroma of tea into the
traditionally sweet and creamy milk tea profile. This innovative approach not only tantalized
taste buds but also appealed to a discerning clientele. Furthermore, Sexy Tea’s unassuming
packaging design exudes an essence of Chinese antiquity and sophistication, providing
consumers with a sensory journey that transcends the ordinary. The success of Sexy Tea has
ignited a fire within the entire landscape of the new Chinese milk tea market. Brands like
Chagee, Molly Tea, and others have followed suit, embraced innovation and pushed
boundaries to continually captivate consumers. Together, these new Chinese milk tea brands
have played an instrumental role in propelling the milk tea market forward, infusing it with a
distinct and vibrant new energy that continues to shape and redefine the industry.
Keywords: Sexy Tea, New Chinese Milk Tea, promotional strategies, culture meaning, digital
culture communication
1. Introduction
With the rapid development of the Chinese market, the beverage industry has achieved great success
in recent years. Chinese New Chinese Milk Tea brands have emerged among them as trend-setting
forces in the market. Sexy Tea (Chayan Yuese) has attracted a wide range of consumers with its
unique marketing strategy and innovative product portfolio. Sexy Tea has quickly emerged in the
highly competitive New Chinese Milk Tea market with its distinctive brand features, innovative
promotional approach and unique consumer experience. This phenomenon has sparked widespread
interest, fuelling researchers’ interest in its marketing strategy and its impact on the industry as a
whole.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231090
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
139
This study aims to provide an in-depth examination of Sexy Tea’s marketing strategies and the
impact of these strategies on Chinese new Chinese milk tea brands. The study will provide an in-
depth analysis of Sexy Tea’s market positioning, brand identity, promotional activities and consumer
engagement and assess its impact on the competitive landscape and development trends of the new
Chinese milk tea market. By drawing on Sexy Tea’s successful experience, the study will also provide
insights for other New Chinese Milk Tea brands on how to develop marketing strategies to increase
brand awareness and competitiveness.
By delving into the marketing strategies of the Sexy Tea brand and their impact, this study not
only contributes to an understanding of the market dynamics of the New Chinese Milk Tea market
but also provides insights for brand managers, marketing professionals, and policymakers on how to
implement innovative strategies in a highly competitive market. Ultimately, it is hoped that the results
of this study will help to drive the sustainable development of new Chinese milk tea brands and
provide a valuable reference for the industry as a whole.
This study will first discuss the current status and trends of the Chinese milk tea market and the
background of the Sexy Tea brand, followed by a marketing strategy analysis starting from the STP
and 4PS and discussing the impact of the new Chinese milk tea brand and its development prospects.
2. Literature Review
2.1. Trends in China’s Milk Tea Market
Since the late 20th century, when milk tea was introduced from Taiwan to mainland China, China’s
milk tea industry has been expanding, becoming an indispensable part of the lives of many young and
middle-aged people. There are four stages of development in China’s milk tea industry, namely,
powdered milk tea, bucket milk tea, hand-cranked milk tea, and ready-to-drink tea. Now, it is in the
crazy expansion stage of ready-made tea drinks, with a market of nearly 100 billion [1]. Nowadays,
more and more local brands are gradually rising. According to the daily economic news on the
“generation Z” (born in 1995 ~ 2010) new tea drink consumer research, MIXUE ice cream & tea is
the “generation Z” most favourite new tea drink brand, popular with the “generation Z”. MIXUE ice
cream & tea is the most popular new tea drink brand among Generation Z, favoured by 28.1% of
respondents, followed by Good Me and Sexy Tea, accounting for 8.3% and 7.6% respectively. In the
past two years, there has been a concentrated outbreak of “net red milk tea”, and based on the
development mode of online channels, milk tea products and the development of the milk tea industry
have also begun to embark on the network mode, attracting many young people with changing
consumer attitudes through online advertising, blogger push, and exquisite shop decoration. Sexy Tea
is a brand that has become a different firework in the new tea drinks because it highlights its “new
Chinese” milk tea attributes [2].
2.2. Introduction of Sexy Tea
As the birthplace of tea culture, tea is not just a drink in China, but also a culture. Drinking tea has
long been a habit of Chinese people. Sexy Tea is not only a fusion of Chinese tea culture, but also the
first Chinese-themed milk tea shop in mainland China.
Founded in mid-2014 and headquartered in Changsha’s busiest area, Huangxing Pedestrian Street,
Sexy Tea will only open shops in Changsha before 2020, and there will be a Sexy Tea shop every ten
metres on Changsha’s streets. Until 2020, Sexy Tea started to open shops in Wuhan, then came to
Chongqing, Nanjing and other cities.
Sexy Tea is a tea beverage brand with R&D, production, and sales as a whole, constantly
introducing unique, delicate, novel and green milk tea products to customers, creating a high-quality
leisure space for consumers. Sexy Tea Milk Tea emerged at a time when consumers’ enthusiasm for
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231090
140
traditional milk tea was sharply declining. As a pioneer in the production of fresh tea drinks in China,
Sexy Tea insists on its own characteristics and introduces milk tea drinks that are unique in the market.
Each drink is divided into three layers, from top to bottom the first layer is different flavour
ingredients, the second layer is made of New Zealand light cream, and the third layer is tea base with
Chinese characteristics [3].
3. Marketing Strategy Analysis of Sexy Tea
3.1. STP Analysis
Geographic Segmentation - Sexy Tea’s hometown is Changsha, Hunan Province, and when
consumers from all over the country are calling for Sexy Tea to open shops outside of China, Sexy
Tea first chose Wuhan, Hubei Province, which is closer to Changsha. On the one hand, Wuhan is a
new first-tier city with a high level of consumption, and on the other hand, Wuhan has a large number
of college students, who are willing to pay for the online milk tea and even queue up for eight hours
to buy it, and college students are very active in milk tea consumption and online promotion. Milk
tea consumption and network publicity is very positive. The fact that Tea Face gradually chose to
open shops in Chongqing and Nanjing is also due to such urban reasons.
Demographic Segmentation - Although the majority of milk tea drinkers are young people, when
Sexy Tea incorporates a strong Chinese tea base, pure Nestle milk and imported blueberries, more
middle-aged and older people are willing to buy this innovative, healthy, tasty, old-fashioned milk
tea.
Sexy Tea’s product positioning is a new Chinese tea drink, and its newness is not only reflected in
the production of the tea drink and the innovation of the flavour, but also in the name of the product,
the brand logo and the visual design of the cultural and creative periphery. The target market is firstly
Changsha locals who are proud of and trust Texy Tea, and secondly, the customer market all over the
country who love the new Chinese style products, love to drink tea leaf tea, and even keen on the
Netflix Milk Tea to hit the market.
3.2. Marketing Strategy Analysis
3.2.1. Promotional Strategies
The most creative tea drink: This tea shop, described by consumers as “the most creative tea shop”,
also has independently designed creative products. In addition to the sales of the tea shop itself, they
have also created a separate retail brand shop, “Zhihu - tea “. There are about 100 SKUs of cultural
and creative products, including flavored tea, teacups, umbrellas, puzzles, car line books, postcards,
and other peripherals.
Undeniably, cultural, and creative products can increase the thickness of the brand, and the brand’s
unique cultural attributes have become cultural symbols that young people are willing to wear and
carry on their backs to express themselves while also driving the sales of tea through cultural and
creative products.
Sentimental, interesting brand power:IP brand image with face value is not enough to really
impress consumers also rely on a unique soul. Because often the image can be copied, but the soul of
the brand IP is difficult to imitate.
Directly operated and not franchised:The main reason why Sexy Tea is not open for franchising is
that it is still worried that after the release, there will be problems with the management and the quality
of the milk tea will not be well controlled, and once there is a problem with the quality of the milk
tea, it will directly affect its reputation, which is more than worth the loss [4].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231090
141
3.2.2. Logo and Packaging Design Strategy
Logo: The ancient beauty in the logo is holding a fan, looking gentle and skilful, the character’s
temperament is very noble and elegant, behind the woman is a red octagonal window. Red and black
are the colours commonly used in China’s early pottery. During the Xia, Shang and Zhou Dynasties,
they were widely used in architecture, clothing and other areas of life, and in the Qin and Han
Dynasties, black and red were even more important. Therefore, in China’s traditional colour scheme,
it is common to see red and black matching, emphasizing the balance of warm and cold; and black
and white matching has been passed down from the ancient times to the present day, with white as
yang and black as yin, and the two matching to harmonize yin and yang. Black hair and white fan,
clothing form a strong contrast, ancient, easy to identify, good use of traditional cultural symbols and
the role of colour.
Packaging: The positioning of Sexy Tea’s packaging design is “the combination of traditional
Chinese tea and modern drinks”. Sexy Tea often chooses Chinese traditional famous painting and
calligraphy elements for redesign, including landscape paintings, figure paintings, bird and flower
paintings, such as “Han Xizai Night Banquet”, “Thousand Miles of Rivers and Mountains”,
“Dwelling in the Fuchun Mountains”, etc. The use of traditional painting elements in the packaging
of milk tea gives the brand a special cultural connotation, but also gives the traditional paintings a
new vitality, so that everyone holding the milk tea has become a walking museum, which adds the
brand’s “national style”. This makes every person holding the milk tea become a walking museum,
which adds to the beauty of the brand’s “national style”, and also brings consumers a sense of
satisfaction and a desire to collect and share [5].
3.2.3. Product Strategies
Sexy Tea (Chayan Yuese) defers its drink line into 3 categories:
First is the fresh tea with cream on its top, this kind of tea has been described as the “Hardest tea
to consume”. This means that the tea is completely “fresh”, and it will taste bad after storing it for a
while. In Chinese, the word “hardest” also means that it does not taste well. This kind of counter-
promotion could significantly draw people’s interest in the product while letting people be aware of
the concept of “fresh” [6].
The second category is products with smooth foam on top of it, as described by Sexy Tea (Chayan
Yuese), different tea drinks from this category have a similar approach. (The way they make it), while
the taste made by different tea bases will be completely different.
Apart from innovatively creating a tea with different toppings and bases, Sexy tea (Chayan Yuese)
keeps traditional tea drink products (pure tea). according to them, they described it as the “most
unacceptable tea” since it’s not that popular. But they chose to produce these drinks with a little
innovation, and they wish to let more people fall in love with the pure and traditional tea as well.
Currently, the business development strategy was not to develop extremely fast and expand to all
the major cities in China. According to Sina Finance, until August 2023, Sexy Tea has only opened
stores in 10 cities. Located in Changsha, they had not opened stores in another city after the first 7
years [7], with such little store chains, it is possible that they cannot meet all the customers’ needs.
Therefore, they developed many products that customers could purchase online, such as pre-packed
tea bags and water cups. With the help of E-commerce, customers around China could enjoy Sexy
Tea’s products, while also leveraging the brand to a new level through a (product line extension) for
the tea bags they sell. As well as the water cups they sell (Brand extension).
Over the years, Sexy Tea’s drink line remains almost the same, but they improve their products
stably step by step, such as its famous drink Youlan Latte has been updated for several versions. This
means they keep spending efforts on improving the very fundamental but important customer
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231090
142
experience, this makes a big difference in other brands which only focus on releasing new products
to catch costumer’s eyes [8].
3.2.4. Pricing Strategies
When comes to the setting of the price, Sexy Tea has a relatively lower price than other players in the
market: The average price of Sexy Tea is only about 10-20 yuan, compares with other players (Hey
Tea and Naixue which average in 15-30 yuan). This price level usually accounts for the main group
of consumers who purchases this kind of milk tea drink. At the same time, they will have high
expectations for the drink they purchased [8].
The reason why Sexy Tea is this successful is it uses high-quality materials, which will obviously
end up a good experience for the customers. Such as Oolong tea supplied exclusively by contracted
tea gardens, as well as fresh milk and Anchor whipping cream.
3.2.5. Placing Strategies
When comes to place, as mentioned in the previous STP, Sexy Tea positions its shops in many areas
in Changsha. Instead of being a high-profile, Sexy tea place it’s shops everywhere and people in
Changsha could get it at ease. This is a great example of intensive marketing intermediaries; their
extension products are also distributed in different online stores.
4. Analysis of the Influential Role of New Chinese Milk Tea Brands in China
4.1. Reinventing Tea Culture IP
The rapid development of Sexy Tea, a new Chinese tea brand, has brought together many interesting
cultural factors, which have given the tea products a cultural core and become the unique cultural
symbol of the new Chinese tea brand. Every new product release has brought a certain level of media
attention and sparked widespread discussion in society. This is the opportunity to connect different
cultural symbols of new Chinese tea brands into the brand cultural IP of the entire industry [9].
4.2. Enhancing the Aesthetics of Tea Culture
After Sexy Tea, the milk tea brand “Chagee” adopts the traditional Peking Opera’s face and portrait
elements in its shop decorations, advocating the cultural concept of “meeting the world with oriental
tea”, and expressing the oriental aesthetics of the combination of Chinese tea and opera. Both the
packaging and the flavour of the milk tea are similar to Sexy Tea’s idea. On the other hand, there is
the beauty of the language of the copywriting. New Chinese tea brands in the tea drinking scene
carefully created a late night refreshment, workplace tips, close friends, sleep and other consumer
culture scenes, different scenes mapping a different state of mind, the overlap of the state of mind and
copywriting presents the language of the Chinese creative aesthetics. For example, the advert slogan
“Fresh Fruit + Chinese Tea = Human Taste” reflects the normative and artistic beauty of the new
Chinese tea advert copy.
4.3. Expanding New Technologies and Audiences
In terms of technology, New Chinese Tea creates novelty by adding fresh ingredients to the tea and
subtracting from the traditional Chinese tea. In addition to the usual milk, fruits and nuts, a variety of
agricultural products such as grains, taro, as well as probiotics and even collagen peptides are added
to the product. Of course, the presentation of new technology does not mean infinite stacking of
ingredients, but rather integrating artisanal processes such as hand-pounding, hand-brewing, fresh
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231090
143
extraction, cooking, microbiology and other production processes; In terms of audience, the new
Chinese tea drink brand has focused on the new generation of young consumers, actively catering to
their individual cultural aesthetics, unique consumption habits and delicate dietary pursuits, grasping
the mindset of the next generation that is willing to express themselves and seek cultural consensus,
closely following their spiritual needs of chasing the flow tendency of the times, and even identify
the tagline health features of young people who love to eat foods high in sugar and high in fat, and
achieving the 0-sugar, 0-fat, and 0-carb concept’s popularity in the beverage industry.
4.4. Innovating New Rituals and Interactions
In recent years, new Chinese tea drinks have become a “new favourite” for social interaction, which
can not only satisfy the audience’s need for refinement and diversification of beverages, but also meet
the demand for emotional interaction between reality and the virtual world. The phrase “the first cup
of milk tea in autumn” has become popular on the Internet over the years, and Honey Snow Ice City
and Gu Ming Tea Drink have re-launched the topic of “the first cup of milk tea in autumn” on
microblogs and other platforms, calling for the drinking of a cup of tea to express their love for the
people around them. By creating a sense of ceremony, which is missing in modern fast-paced life,
the new Chinese tea brands aim to achieve the purpose of multiple interactions between people, giving
the cultural connotation of tea drinking the image of happiness and satisfaction, triggering resonance,
and fulfilling the needs of interaction between individuals and groups [10].
5. Potential Challenges for Sexy Tea’s Future Development
5.1. Geographical Limitations of the Business Scope
“Sexy Tea has always been known as a landmark brand in Changsha, and this “deep ploughing into
a city” model has brought huge dividends but has also limited its development to a certain extent.
Starvation marketing has resulted in the loss of a large number of out-of-town consumers, which does
not generate enough revenue for Sexy Tea shops. Therefore, “Sexy Tea” should do a good job of
researching in the field, and on the basis of ensuring word-of-mouth, expand its shops to other places,
not only in big cities, but also in small cities, so that more people can drink Sexy Tea and fall in love
with Sexy Tea, and maintain modesty and affinity, not to be a brand that is above the rest.
5.2. Weakening of Innovation Ability in the Late Stage
As a famous “net red milk tea” brand, “Sexy Tea” should maintain a high degree of sensitivity to the
hotspots of the times, and drive the brand development through various ways of online marketing,
such as TikTok Live, b-station videos, etc. In addition, “Sexy Tea” should maintain a high degree of
sensitivity to the hotspots of the times. In addition, “Sexy Tea” should give full play to its advantage
of “new Chinese” style, further explore China’s excellent traditional culture, strengthen innovation,
and create more new tea products to enhance its core competitiveness. For example, ancient poems
can be printed on milk tea cups and peripheral products.
5.3. Food Safety Guarantee
Strengthening supervision and establishing a good brand In the face of the hot tea market, the Sexy
Tea team should think of danger, strengthen supervision and quality control, and maintain a warm
service attitude to provide consumers with better quality products and services. For example, More
Yogurt, a brand that used phylloxera to make yoghurt and deceived consumers through false
advertising, now has few customers in front of its shops.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231090
144
5.4. Enhancing the Defence of Intellectual Property Rights
In a competitive marketplace, the success of a company usually leads to a large number of imitators
and plagiarisers. Sexy Tea, a popular brand, is no exception. Sexy Tea is a popular and highly
regarded brand, and there are many milk tea shops with packaging and shops similar to Sexy Tea’s
on the streets. This imitation phenomenon may not only infringe on a company’s intellectual property
rights but may also damage the brand’s reputation and market share.
Therefore, enterprises need to strengthen the protection of their intellectual property rights in order
to maintain their market position and competitiveness. This includes legal means such as registered
trademarks, patents and copyrights to ensure that other enterprises do not make unauthorised use of
their unique brand elements and innovations. It is also crucial to build and establish a favourable
corporate brand image. By reinforcing its brand values and delivering superior products and services,
a company can strengthen its position in the marketplace and enhance consumer trust and recognition.
In this era of intense competition, companies need to focus not only on the development of
marketing strategies, but also on long-term development and brand sustainability. It is only through
protecting intellectual property rights and building a good brand image that a company can stand out
in a competitive marketplace and achieve long-term success.
5.5. Adjusting the English Name of the Enterprise
At the same time, the English name “Sexy Tea” makes many consumers feel disrespectful to women,
and also brings challenges to its publicity and development. It should be used as a lesson to
standardise the communication culture and reject vulgarity, and as a disseminator of tea culture,
change the name as soon as possible to a more recognisable and elegant translation.
5.6. Developing an Omni-Channel Marketing Strategy
After being affected by various unfavourable factors, Sexy Tea further expanded its online flagship
shop on Tmall, selling tea peripherals, cultural and creative products and all kinds of co-branded
merchandise, which became one of the main sources of income during the epidemic. In the future,
under the background of the Internet technology is gradually developed, the new Chinese tea industry
to online development is an inevitable trend, only with the Internet, in order to achieve online and
offline go hand in hand.
6. Conclusion
To sum up, the new Chinese tea beverage brand represented by Sexy Tea has formed a new era of
Internet cultural landscape with elements of contemporary Chinese tea culture, which continues the
vitality and creativity of Chinese tea culture and leads more young people to learn about Chinese tea,
fall in love with Chinese tea, and fall in love with the traditional culture of the Chinese nation that
has a long history. The limitation of this study lies in the insufficient research on the new media
development of new Chinese milk tea. In the digital era, the new Chinese tea industry has combined
its efforts to build a brand culture communication characteristic of technology and audience, style
and creativity, ritual and interaction, and the development direction of the future study will be more
focused on the investigation of the connection between the new Chinese products and the new media
communication. More importantly, in the future, only by continuing to dig deeper into the brand
culture development strategy of new Chinese tea drinks, each brand market based on its own core
strengths, optimising the supply chain, reducing negative news, and doing a good job with the
products can its communication power be fundamentally strengthened. Only by adhering to the new
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231090
145
Chinese milk tea business start-up heart and philosophy, in order to do a good job with the Chinese
flavour of milk tea brand, to the world.
References
[1] Analysis of the Current Situation and Development Trends of China’s Milk Tea Industry Market in 2020: Intelligent,
Branded, Operated, and Differentiated Development. (2020, August 26).
https://bg.qianzhan.com/trends/detail/506/200826-c9a6fe8c.html.
[2] SU, J. (2023). Generation Z Shift of Tea Consumption and Brand Building. BRAND & STANDARDIZATION, 1.
https://doi.org/10.3969/j.issn.1674-4977.2022.Z1.014
[3] Sexy Tea. (2023). About us. http://www.hn9518.com/about/
[4] Wang, X., & Feng, Y. (2021). Reflections on the popularity of the new tea beverage market ——Taking “SexyTea”
as an example. PR Magazine, 12. https://doi.org/10.16645/j.cnki.cn11-5281/c.2021.12.030
[5] Gu, S., & Zhang, K. (2023). Aesthetic features in the graphic ideas for Chinese-styled packaging design of “Sexy
Tea.” Design, 36(1), 54-69. https://doi.org/10.20055/j.cnki.1003-0069.000391
[6] Chayan Yuese (Sexy Tea). (n.d.). Official Website of Cha Yanyuese (Sexy Tea).
https://www.sexytea2013.com/#straightforward
[7] Sina finance Chayan Yuese (Sexy Tea) accelerated “out of Hunan,” Wuxi, the national expansion will meet the new
Chinese tea drink “new enemy” (2023). https://finance.sina.com.cn/tech/roll/2023-02-03/doc-
imyemchv3767033.shtml
[8] Wang, C. (2021, October). Chayan Yuese: The secret of “top flow tea drink.” The Paper. Retrieved August 1, 2023,
from https://www.thepaper.cn/newsDetail_forward_14782642
[9] Liu, R. (2023). A Study on the Cultural Communication and Development Strategy of New Chinese Tea Drink
Brands. Culture Journal, 1, 1673-7725(2023)01-0018-04.
[10] Jing, Y., Qing, Y., & Zhao, A. (2023). The new development of new Chinese tea drink brands and the empowerment
of traditional culture: A market research based on the cognition and consumption of “Sexy Tea.” Economic
Research Guide, 542, 58-66.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231090
146
Portfolio Optimazation Based on 10 US Stocks
Tianzhou Yu1,a,*
1Faculty of Engineering, The Chinese University of Hong Kong, Hong Kong, China
a. yutianzhou@link.cuhk.edu.hk
*corresponding author
Abstract: The US stock market has recently become volatile because of the increase in
interest rates, which arouses investors’ focus on portfolio optimization. This study selected
10 representative US stocks from different industries listed on NASDAQ and NYSE. This
paper applies Monte Carlo simulations to determine the efficient frontier and constructs
portfolios with maximum Sharpe Ratio and minimum volatility. The result showed that
Lululemon Athletica Inc. possesses the largest proportion of the maximum Sharpe ratio
portfolio, and the Coca-Cola Company occupies the greatest weight in the min volatility
portfolio. By comparing the cumulative return of the two portfolios with the NASDAQ
Composite Index, the max Sharpe ratio portfolio is found to overperform the market
benchmark. In the robustness test in which 8 stocks selected from the original group of assets
are taken to construct a portfolio, the same result still holds. This result may demonstrate the
feasibility of portfolio management for a few investors in that period.
Keywords: portfolio optimization, mean-variance analysis, US stock market
1. Introduction
The interest rate at which banks and credit unions borrow from one another is referred to as the federal
funds rate, established by the Federal Reserve. It plays a crucial role as a benchmark for nearly all
interest rates and its fluctuations will pose powerful impacts on global financial market. The fed funds
rate is grown up to 5.25% to 5.50%, which is the highest level in the past twenty years, according to
the data of the Federal Reserve [1]. The rise in the interest rate alters the long-standing low interest
environment, which makes it a concerning issue for investors to mitigate the risk under the current
situation through applying appropriate portfolio strategy.
Starting from portfolio selection approach proposed by Harry Markowitz, in which the mean-
variance model was first introduced and conclusion that the expected return of an asset is determined
by the magnitude of its own risk was also drawn, modern portfolio theory has become more
comprehensive through the efforts of plenty of scholars [2]. Chen, Zhang, Mehlawat, and Jia
introduced a novel method for building portfolios. This method involves integrating a machine
learning-based stock prediction model and utilizing the mean-variance model for asset selection [3].
Hans, Sahamkhadam, and Stephan conducted a study on optimizing portfolio performance within the
global timber and forestry industry. They compared the performance of the portfolio with the global
S&P index to assess the influence of integrating social responsibility considerations into the portfolio
construction process [4]. Ivanova1 and Dospatliev studied 50 stocks traded on Bulgarian Stock
Exchange to find portfolio with highest return [5]. Chizari and Vazirian did research aimed at
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231100
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
147
identifying the most suitable stock portfolio for agricultural companies listed on the Tehran Stock
Exchange [6]. Debnath and Srivastava construct their own portfolio based on stocks listed on National
Stock Exchange (NSE), India [7]. Furthermore, Nisani1 and Shelef conducted research about
investors’ preferences for portfolio selection, showing that investors tend to pursuit risk for higher
returns [8].
There is a dearth of research on investment construction in the context of interest rate hikes. Thus,
this paper implements the following empirical investigations to fill the potential research gap. Firstly,
the selection of stocks. Stocks from fields including technology, healthcare, finance, consumer goods,
e-commerce, entertainment and telecommunications are selected from NASDAQ and NYSE.
Secondly, Monte Carlo simulation is applied to simulate 10,0000 different portfolios, and the efficient
frontier is obtained in which the portfolio with the highest Sharpe ratio and the portfolio with the
lowest volatility are identified. Through a comparison of the cumulative returns of these two
portfolios with the performance of the NASDAQ Composite Index, the optimal portfolio is
determined. Additionally, robustness check is designed to examine the consistency of results.
Eventually, the conclusion is drawn from previous steps.
2. Data
This paper selects 10 representative stocks from various industries listed on the NASDAQ and NYSE.
The stock symbols of the 10 stocks are CSCO, EBAY, IBM, JNJ, JPM, KO, LULU, NFLX, PG,
TMUS. The reason why these stocks are picked is that this mix includes technology, healthcare,
finance, consumer goods, entertainment and telecommunications. This combination aims to achieve
a balanced exposure to various segments of the market, potentially reducing risk through industry
diversification while capturing opportunities for growth and stability. Closing prices from December
18th, 2022, to June 30th, 2023, are obtained from Yahoo Finance to conduct the study. The collected
data is divided into training set, which is used for calculating the annualized return and volatility to
plot the efficient frontier, and test set, which aims to evaluate the performance of portfolio by
comparing the cumulative return to market benchmark return in a given period of time. The data from
December 18th, 2022 to May 18th, 2022 is to find the optimal portfolio, and the data from May 30,
2022 to June 30, 2022 is to examine the performance of the portfolio. The basic statistics of the
selected stocks is presented in Table 1.
Table 1: Descriptive statistics of selected assets.
Mean
Std Dev
Min
Max
CSCO
0.0002
0.0128
-0.0462
0.0511
EBAY
0.0006
0.0186
-0.0537
0.0495
IBM
-0.0007
0.0106
-0.0459
0.0316
JNJ
-0.0009
0.0103
-0.0377
0.0439
JPM
0.0008
0.0167
-0.0557
0.0728
KO
0.0001
0.0079
-0.0308
0.0191
LULU
0.0016
0.0214
-0.0975
0.1197
NELX
0.0016
0.0231
-0.0460
0.0862
PG
0.0004
0.0092
-0.0272
0.0340
TMUS
-2.77E-06
0.0119
-0.0411
0.0318
As Table 1 presents, the highest average log return appears at NELX while the lowest standard
deviation appears at KO. The paper also calculates the cumulative return for each stock in order to
observe trends of these assets directly. Figure 1 reveals that NELX has maintained the highest
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231100
148
cumulative return, while JNJ has exhibited the lowest cumulative income at the end of the period of
five months.
Figure 1: Cumulative returns of 10 stocks.
3. Method
3.1. Mean-Variance Analysis
Mean-variance analysis is the process of evaluating variance, which is also called risk, against
expected return to achieve the optimal balance. According to Modern Portfolio Theory, investors are
assumed to be rational in making investments if they can access complete information, and they seek
low risk and high return while only caring about return and risk [2]. Mean-variance model enables
investors to optimize the weights of their portfolio to achieve the maximum expected return at a given
level of risk, or to achieve the minimum risk at a targeted return level. The equation for the expected
return and variance of a portfolio are as follows:
(1)
Where is the asset weight of the portfolio, is the expected return of the asset.
(2)
Where
is the variance of the portfolio, is the standard deviation of the asset return, and
is the correlation coefficient between the returns of assets and .
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231100
149
Furthermore, Sharpe ratio, which is first introduced by William F. Sharpe, is a frequently used
index to measure the return of an investment compared with its risk [9]. The equation is given as
below.
Sharpe Ratio
(3)
Where is the return of a portfolio, is risk-free rate and is the standard deviation of the
portfolio.
3.2. Monte Carlo Simulation
The Monte Carlo simulation is a method that utilizes generated random variables and economic
factors such as expected returns and volatility to simulate thousands of possible outcomes. An
advantageous aspect of Monte Carlo methods is their capacity to readily facilitate scenario analysis,
which means it calculates potential risk outcomes based on various assumptions [10]. In this study, it
works by creating a large number of random weights, incorporating asset returns, correlations and
volatility. By analyzing these simulated scenarios, investors can assess the range of possible portfolio
returns and risks, make more informed decisions and construct portfolios that balance risk and return.
4. Result
100,000 simulations had been performed using the Monte Carlo method in which the data ranges
from December 18th, 2022, to May 18th, 2023. The scatter plot is shown below (Figure 2), and the
blue dashed line is the efficient frontier.
Figure 2: Efficient Frontier.
The optimal portfolios, namely max Sharpe_ratio portfolio and min volatility portfolio can be
obtained from the depicted efficient frontier. As Figure 3 shows, the purple star stands for the max
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231100
150
Sharpe Ratio portfolio and the blue square represents the min volatility Portfolio. The calculated
weight of each asset in two portfolios and detailed characteristics of two portfolios are in Table 2 and
Table 3, respectively.
Figure 3: Max Sharpe_ratio portfolio and Min volatility portfolio.
Table 2: Weight of each asset in two portfolios (%).
Stock Code
Max Sharpe ratio
Min volatility
CSCO
3.4322
17.1787
EBAY
0.2976
0.2412
IBM
2.6921
5.5092
JNJ
2.7237
15.7817
JPM
18.6859
2.7443
KO
0.7005
22.8056
LULU
29.1786
0.8992
NFLX
17.1583
0.9034
PG
21.9663
20.5943
TMUS
3.1647
13.3425
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231100
151
Table 3: Volatility, return and Sharpe ratio of two portfolios (%).
Return
Volatility
Sharpe Ratio
Max Sharpe Ratio portfolio
23.6421
17.5765
134.5098
Min volatility portfolio
0.4722
11.0668
4.2669
It can be seen from the tables above that the results are different apparently. In the Maximum
Sharpe Ratio portfolio, LULU and PG have high weights, whose sum takes half of the total weight,
while EBAY and KO have much smaller weights than those of other assets, which sums up to only
1%. In the Minimum Volatility portfolio, KO has the largest weight which is 22.81%. and the lowest
weight is still EBAY which is only 0.24%. In these two portfolios, assets EBAY, IBM and PG
maintain their weights relatively while other assets fluctuate a lot.
Performance of optimal portfolio obtained above can be verified after acquiring the weights of
each asset in the portfolio. The paper uses the test data ranging from May 19th, 2022, to June 19th,
2022, to calculate the real risk and return using the portfolio weights and compare the result with
NASDAQ Composite Index. If the cumulative return of the portfolio is greater than that of NASDAQ
Composite Index, it can demonstrate that the portfolio strategies mentioned in this study are feasible.
The paper calculates the cumulative returns of max Sharpe Ratio portfolio, min Volatility portfolio
and the NASDAQ Composite Index, and the results are 6.797%, 3.242% and 4.409% respectively.
Figure 4 indicates that max Sharpe Ratio portfolio outperforms the market in regard to the cumulative
return.
Figure 4: Comparison between NASDAQ Composite index return and the portfolio returns.
5. Robustness
A robustness test will be conducted for effectiveness. Stock EBAY and IBM are excluded from the
asset group since they only account for small percentages of weights and their percentages do not
fluctuate a lot in both portfolios. Then the Monte Carlo simulation is employed to generate a new set
of weights. Finally, the cumulative return of new portfolios will be compared with the NASDAQ
Composite Index to evaluate the performance of portfolios. The altered weights for the two optimal
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231100
152
portfolios are shown in Table 4. The cumulative returns of max Sharpe Ratio portfolio, min Volatility
portfolio and the NASDAQ Composite Index are 7.269%, 3.287% and 4.409% respectively. Figure
5 indicates that max Sharpe Ratio portfolio outperforms the market in regard to the cumulative return,
which presents a similar result as the 10-asset portfolio test. Therefore, validity and effectiveness of
the method are verified.
Table 4: Altered weight of each asset in two portfolios (%).
Stock Code
Max Sharpe ratio
Min volatility
CSCO
0.7450
13.6851
JNJ
0.6737
19.2317
JPM
19.9857
7.0247
KO
2.6193
31.0241
LULU
28.5656
2.8318
NFLX
27.1543
0.4626
PG
18.6417
13.0318
TMUS
1.6147
12.7081
Figure 5: Comparison between NASDAQ Composite index return and altered portfolio returns.
6. Conclusion
In summary, in this study, a selection of 10 prominent US stocks spanning various industries and
listed on NASDAQ and NYSE was made. By utilizing Monte Carlo simulation, the study constructed
the efficient frontier and created portfolios characterized by both the highest Sharpe ratio and the
lowest volatility. The findings highlighted Lululemon Athletica Inc. as the predominant component
of the maximum Sharpe ratio portfolio, while the Coca-Cola Company held the largest weight in the
minimum volatility portfolio. Comparing the cumulative returns of these portfolios with the
NASDAQ Composite Index, the maximum Sharpe ratio portfolio exhibited superior performance as
compared to the market benchmark. This result remained consistent in a robustness test, where a
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231100
153
subset of 8 stocks from the original asset group was employed for portfolio construction. For investors
focusing on the US stock market during the given period, these portfolios offer valuable insights.
The asset set used in this paper is merely one representation of the U.S. stock market in a scenario
of rising interest rates and does not explore other emerging U.S. equities. Subsequent researchers
could incorporate emerging stocks into the portfolio to obtain a more comprehensive composition.
References
[1] Open market operations. (2022) Retrieved from https://www.federalreserve.gov/monetarypolicy/openmarket.htm.
[2] Markowitz, H. (1952). Portfolio Selection. The Journal of Finance, 7(1), 77–91.
[3] Chen, W., Zhang, H., Mehlawat, M. K., and Jia, L. (2021). Mean–variance portfolio optimization using machine
learning-based stock price prediction. Applied Soft Computing, 100, 106943.
[4] Lööf, H., Sahamkhadam, M., and Stephan, A. (2023). Incorporating ESG into Optimal Stock Portfolios for the
Global Timber & Forestry Industry. Journal of Forest Economics, 38(2), 133–157.
[5] Ivanova, M., and Dospatliev, L.K. (2018). APPLICATION OF MARKOWITZ PORTFOLIO OPTIMIZATION ON
BULGARIAN STOCK MARKET FROM 2013 TO 2016. International Journal of Pure and Applied Mathematics,
117.
[6] Chizari, A.H., and K. Vazirian. (2022). Determining the Optimal Stock Portfolio of Agricultural Companies in
Tehran Stock Exchange. Journal of Agricultural Economics & Development 35(4), 383-395
[7] Debnath, P., and Srivastava, H. M. (2021). Optimal Returns in Indian Stock Market during Global Pandemic: A
Comparative Study. Journal of Risk and Financial Management, 14(12), 592.
[8] Nisani, D., Shelef, A. (2021) A statistical analysis of investor preferences for portfolio selection. Empirical
Economics, 61, 1883–1915.
[9] The sharpe ratio. (2022) Retrieved from https://web.stanford.edu/~wfsharpe/art/sr/sr.htm.
[10] Kroese, D. P., Brereton, T., Taimre, T., and Botev, Z. I. (2014). Why the Monte Carlo method is so important today.
Wiley Interdisciplinary Reviews: Computational Statistics, 6(6), 386–392.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231100
154
The Impact of Population Aging on Financial Services and
Economic Development
Hongyu Fang1,a,*
1School of Graduate Studies, Rutgers, The State University of New Jersey, New Brunswick, US
a. HF266@rutgers.edu
*corresponding author
Abstract: With the increasing aging population, it has had a profound impact on the financial
services industry and economy. This study aims to explore the effect of population aging on
the financial service industry and economy and propose corresponding countermeasures and
suggestions. The aging population has brought challenges to the financial services industry.
The increase in the elderly population has led to a rise in demand for financial products and
services. Financial institutions innovate and develop financial products and services that meet
the needs of older people. At the same time, the financial literacy of older people is relatively
low, and financial institutions need to strengthen financial education and improve their
financial knowledge level. Secondly, the aging population has had multiple economic impacts,
with decreased labor supply and increased elderly care expenses putting pressure on
economic growth. Financial institutions must pay attention to this trend, adjust loan policies,
and support the development of emerging industries to promote sustainable economic growth.
In addition, the change in the consumption needs of older adults has also brought new
opportunities to the economy. Financial institutions can meet the consumption needs of older
people and promote economic development through innovative products and services.
Keywords: population aging, financial services industry, economy, social influence
1. Introduction
The aging population is a serious challenge facing modern society. With the development of medical
technology and the improvement of living standards, people's lifespan continues to extend, leading
to a sharp. This trend has profoundly impacted the financial services industry and the entire economy.
Poses new challenges and opportunities for financial institutions. At the same time, the demand for
financial products and services among older people has also undergone significant changes in the
economy, consumption, and investment. Attending and responding to the impact of population aging
on the financial services industry and economy is crucial for formulating appropriate policies and
strategies to address this challenge. Therefore, this article will explore the impact of population aging
on the financial services industry and economy and propose corresponding response measures to
promote sustainable development and social well-being.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231101
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
155
2. The Impact of Aging Population on the Financial Services Industry
2.1. Changes in the Savings and Investment Market
With the aging population, the demand for savings among retirees has increased, Such as pension
plans, insurance, and retirement plans. Meanwhile, as older adults are more concerned about asset
preservation and long-term returns, financial institutions may adjust their risk tolerance and
investment preferences. With the intensification of population aging, the demand for savings among
the elderly population will increase, and they usually hope to have enough savings to maintain
retirement and cope with health issues. This will lead financial institutions to adjust them. With the
aging population, investors' risk tolerance may decrease, and older adults are more concerned with
capital preservation and stable investment returns rather than pursuing high-risk and high-yield
investments. Financial institutions must provide more conservative and robust investment products
to meet the needs of older people. The aging population poses new challenges and opportunities for
the financial services industry, and financial institutions need to develop innovatively, such as
retirement planning consulting, medical insurance, and long-term care insurance. In addition,
technology-based financial services are also expected to provide more convenient and personalized
financial service experiences for older people [1].
2.2. Innovation in Financial Service Methods
With the increase in the elderly population, financial institutions need to adapt to the needs of older
people and provide more convenient and personalized financial services. For example, they offered
remote banking services, intelligent investment consulting, financial education, and digital financial
products to meet the needs of older people for financial assistance. In addition, the development of
financial technology will also bring more innovation and convenience to the financial services
industry. As the proportion of the elderly population increases, personalized elderly care financial
services will become more critical [2]. Financial institutions can meet the unique needs of older
people through innovative products and services, such as providing customized savings and
investment plans for more senior people and providing them with better financial planning and
investment advice. With the continuous development of technology, the financial services industry
needs to respond to the digital transformation trend actively. Digitizing financial services can improve
the convenience and efficiency of services that better meet older people's needs, such as providing
online banking services, electronic wallets, and convenient electronic shopping experiences through
internet banking, mobile payments, and e-commerce. Due to the high savings and investment needs
of the elderly, financial education and consulting services are particularly important for them.
Financial institutions can carry out financial education activities targeting older people, improve their
financial knowledge level, and provide professional consulting services to help them make wiser
financial decisions [3].
2.3. Transformation and Development of the Insurance Industry
The demand for medical, pension, and long-term care insurance will significantly increase with the
aging population. Financial institutions need to develop and promote more comprehensive and
customized insurance products tailored to the risk management needs of older people. At the same
time, financial institutions can provide integrated health and financial services through cooperation
with medical and elderly care institutions [4]. The aging population has had a series of impacts on the
financial services industry, especially the insurance industry. With the deepening population aging,
insurance companies have begun to develop innovative products and services for older people, such
as launching health insurance, pension insurance, longevity insurance, and other products that meet
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231101
156
the needs of older people, as well as providing customized insurance plans to meet the unique needs
of older people [5]. In addition, insurance companies can also strengthen online services and digital
channels to facilitate the purchase and settlement of claims insurance for older people and improve
user experience. The aging population has brought various health and financial risks, and insurance
companies must pay more attention to risk management and assessment. By integrating big data and
artificial intelligence technology, insurance companies can more accurately assess the health risks of
older people to be more precise in product design and premium pricing [6].
2.4. Adjustment of Financial Regulatory Policies
The government and regulatory agencies may strengthen their regulatory efforts in areas such as
pension, pension, insurance, and financial innovation to ensure the government may introduce
incentive measures to encourage financial institutions to provide better services and products for older
people [7].
2.4.1. Strengthening Risk Management
Due to the potential increase in risks financial institutions face due to an aging population, regulatory
agencies will strengthen their supervision of financial institutions, especially in savings, investment,
and retirement. Regulatory authorities may require financial institutions to adopt more cautious risk
management measures to cope with potential risks [8].
2.4.2. Pension Regulation Reform
The aging population means that the demand for pension funds will increase, and regulatory agencies
may reform the management and supervision of pension plans to ensure the safety and sustainability
of pension funds. This may include strengthening investment regulation, increasing transparency of
pension funds, and encouraging financial institutions to provide more retirement planning services
[9].
2.4.3. Adjustment of Insurance Regulation
Especially medical insurance and long-term care insurance. Regulatory agencies may strengthen their
supervision of the sales and operation of these insurance products to ensure that older people receive
appropriate protection. At the same time, regulatory agencies may also encourage innovation and
promote the development of more personalized and customized insurance products in the insurance
industry to meet the needs of older people [10].
3. The Impact of an Aging Population on the Economy
3.1. Shortage of Labor Supply
The labor market will face a severe supply-demand imbalance with the aging population. The labor
supply shortage may lead to a decrease in productivity and affect the potential for economic growth.
Enterprises may struggle to find enough suitable employees, limiting economic development. With
the aging population and a significant decrease in labor supply, the labor market faces severe
shortages, which will hurt the overall growth of the Chinese economy. Labor shortages can lead to a
decline in productivity, limiting enterprises' expansion and innovation capabilities. Due to labor
shortages, specific industries and regions may not be able to meet market demand, thereby affecting
the long-term development prospects of the economy [11].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231101
157
The aging population leads to insufficient labor supply and fierce competition in the labor market.
Enterprises may need to increase wages and benefits to attract and retain scarce labor resources.
Therefore, labor costs may rise, putting pressure on the operating costs of enterprises, thereby
affecting their profits and competitiveness—The shortage of labor supply forces enterprises and
governments to redistribute resources and adjust their structures. Companies may adopt technological
means such as automation and roboticization to replace human labor in response to labor shortages.
The government may promote education reform, training, and skill enhancement. The government
must also strengthen the social security system to ensure older people's quality of life and welfare
benefits [12].
3.2. Increasing Social Security Pressure
The aging population will increase the pressure on the social security system, especially in areas such
as pensions and healthcare. The increase in the elderly population has led to a rise in pension payments
and demand for medical services, which will put enormous pressure on the fiscal budget, occupy
many financial resources, and may weaken other public expenditures and investments. With the aging
population, the number of retirees has increased, while the number of working people has relatively
decreased, resulting in an increased burden on pension payments. To ensure the sustainability of
pension funds, the government may need to take measures such as raising the retirement age,
increasing contribution rates, and reforming the pension system to alleviate social security pressures.
As the population ages, healthcare spending will also increase. Older adults are more prone to chronic
diseases requiring long-term care, and the increasing demand for medical resources and care services
poses challenges to the country's healthcare system. The government may increase investment in
medical resources, improve the supply of nursing services, and address this issue by reforming the
medical insurance system. The aging population has exacerbated the supply-demand contradiction in
the labor market. As the number of workers decreases, there may be a shortage in the labor market.
This may bring employment difficulties and rising costs to enterprises, affecting economic
development and competitiveness. The government may take measures, such as promoting the
cultivation of efficient labor and introducing external work, to alleviate the pressure on the labor
market [13].
The aging population has brought increasing social security pressure to the Chinese economy.
Firstly, with the intensification of the aging population trend, the increase in the elderly population
has put enormous pressure on the elderly security system. The scale of pension payments will
significantly increase while the number of beneficiaries is limited, resulting in an imbalance between
pension expenditure and pension income. This will exacerbate the financial pressure on the
government, potentially leading to increased fiscal deficits and limiting the government's budgetary
policy space. Secondly, the aging population will also increase pressure on healthcare and health
security, with the increase of the elderly population. The needs of older adults in healthcare are more
long-term and complex, requiring more medical resources and long-term care. This will increase the
burden on the healthcare system, exacerbate the shortage of medical help, and increase healthcare
costs. Thirdly, the increase in the elderly population means increased social welfare expenditures.
Social welfare expenditures, including social assistance and disability benefits, will increase
accordingly. Expanding social welfare projects will bring additional financial burdens to the
government and require corresponding financial support. In addition, the aging population has also
posed challenges to the increasing demand for elderly care and nursing services. Nursing homes and
home care is rapidly increasing. Meeting the needs of older adults for elderly care services requires a
large workforce and material investment.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231101
158
3.3. Changes in Consumption Structure
With the aging population, the consumption structure will also change. Older adults usually pay more
attention to consumer goods and services such as health, medical services, and elderly care services.
This will impact the consumer market, with some traditional industries potentially affected and some
emerging industries potentially boosted. The demand for service consumption among older people
will increase with the aging population. They may be more inclined to pursue consumer experiences
such as health, entertainment, and travel. Financial institutions can meet the special consumption
needs of older people through innovative financial products and services, such as tourism loans,
medical and health insurance, etc. The aging population has made consumer demand more diversified
and personalized. Older adults may have different requirements for the robustness and yield of
financial products compared to other age groups. Financial institutions need to develop customized
financial products to meet the unique needs of older people, such as pension investment products and
medical expense savings plans. With the development of technology, intelligent technology plays an
increasingly important role in the lives of older people. Financial institutions can combine innovative
technology with financial services to provide more convenient and smart consumption methods for
older people, such as wise payment and intelligent health management. This can improve the
consumption experience and quality of life of older people.
As the trend of population aging intensifies, the proportion of their share in the entire consumer
market is also expanding. The consumption demand of older people is relatively stable and long-term,
mainly dominated by healthcare, elderly care services, tourism, and leisure. This has led to a shift in
the consumer market from being overwhelmed by young people to being dominated by the
consumption needs of older people. Secondly, the structure of consumer goods and services has also
changed. Care services, healthcare, and auxiliary equipment are gradually increasing. Meanwhile, the
demand for traditional consumer sectors such as clothing and electronic products among older people
is relatively decreasing. These changes have significantly impacted the Chinese economy's consumer
goods industry structure and market competition pattern. Thirdly, the aging population challenges the
size of the consumer market. Although the increase in consumer demand among older people is
beneficial for promoting the development of corresponding industries, their overall consumption
ability is relatively low, and compared to young people, their consumption contribution in high-end
consumer goods and commodities is limited. This may impact the development of some industries
and enterprises, requiring the search for new market opportunities and innovative ways. In addition,
the aging population has also driven a shift in consumer attitudes and patterns. Older adults place
greater emphasis on quality, safety, and healthy consumption, with an increasing demand for
personalized services and customized products, which will put forward new requirements for
enterprise product strategies and service concepts. In summary, one of the impacts of population
aging on the Chinese economy is the change in consumption structure. The consumer market has
shifted from being dominated by young people to being overwhelmed by the consumption demand
of older people, with an increase in consumption demand. This has brought challenges and
opportunities to industrial structure, market size, and enterprise development. Enterprises should
adjust their product and service strategies based on the consumption characteristics and changing
needs of older people to adapt to changes in the consumer market and improve market
competitiveness [14].
3.4. Decreased Socio-economic Vitality
The increase in the elderly population may lead to a decrease in socio-economic vitality, as they
typically reduce labor participation, innovation ability, and entrepreneurial spirit, which may affect
innovation and economic growth potential. The aging population has had a series of impacts on the
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231101
159
financial services industry, especially the decline in socio-economic vitality. As the population ages,
labor force participation decreases, the working population decreases, and socio-economic energy
decreases. This will lead to a decrease in the primary source of income for many people, including
salary and entrepreneurial income. This will challenge the financial services industry, as people's
demand for finance may decrease, such as loan and investment needs. With the aging population, the
proportion of the elderly population increases, and the consumption demand structure changes. Older
people's demand for financial services may focus more on pension planning, medical insurance, and
long-term care. In contrast, the need for other financial products and services may decrease. This will
bring new challenges to the development of products and services by financial institutions and
insurance companies, and it is necessary to focus on the needs and preferences of older people. The
decline in socio-economic vitality may lead to a lack of capital markets. Investors and entrepreneurs
may decrease, and investment opportunities and business development may be limited. This will
impact the development of financial institutions and capital markets, reducing their income potential
and the level of capital operation activity. Financial institutions must adapt to market changes, adjust
their business strategies, and find new growth points and investment areas.
The aging population has led to a decrease in labor supply. As many laborers enter a senior state,
the labor market will face a talent shortage challenge. The lack of a young and dynamic labor force
will limit the utilization of production and innovation capabilities, thereby affecting the economy's
long-term growth potential. Secondly, the aging population hurts consumer demand and market size.
The consumption-ability of older adults is relatively low, and the need for commodities and high-end
consumer goods is decreasing, which will impact production and sales. In addition, the increasing
demand for social services such as healthcare and elderly care among older people has brought certain
consumption pressure to the economy. Furthermore, the aging population has increased financial
pressure. With the increase in the elderly population, social pension security and medical expenses
have also increased. This will bring a financial burden to the government, which may increase fiscal
deficits and debt levels, limit the government's budgetary policy space, and thus affect sustainable
economic development. In addition, the aging population has also had an impact on the labor market
and employment structure. The demand for some professions and industries has decreased, which
may lead to increased unemployment rates and an intensification of structural unemployment. This
will increase social security risks and have adverse effects on social stability. In summary, one of the
impacts of population aging on the Chinese economy is the decline in socio-economic vitality.
Reducing labor supply, reducing consumer demand, increasing fiscal pressure, and adjusting the labor
market and employment structure all negatively impact economic stability and development.
Addressing the challenges of population aging requires the development of comprehensive policies,
including measures to promote fertility, strengthen employment, and innovate to minimize the
adverse effects of aging on the economy.
3.5. Reduction of Human Capital
The aging population may also lead to a decrease in human capital. Human capital refers to people's
education, skills, and health level. As people age, human capital often decreases, which may affect
the improvement of labor productivity and the economy's competitiveness.
3.5.1. Reduction in Labor Supply
The aging population leads to a decrease in labor supply, and a decline in the working-age population
may hurt industrial structure and economic growth. The reduced labor force size may lead to
employment difficulties for enterprises, especially in industries that require a large amount of labor.
This may decrease production efficiency and constrain the economy's long-term growth potential.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231101
160
3.5.2. Skill Shortage Issues
The supply of technical and professional talents may be affected by age changes. A relatively small
number of young workers may lead to skill shortages in specific industries and fields. This may
require increased technical training and education investment to ensure the workforce has the skills
and knowledge to adapt to economic development.
3.5.3. Decline in Innovation and Entrepreneurship Capabilities
Young people are usually the main driving force for innovation and entrepreneurship. As the
proportion of young people decreases, the aging population may lead to a decline in innovation and
entrepreneurship capabilities, thereby affecting the driving force and competitiveness of the economy.
The government may need to take measures, such as encouraging innovation and entrepreneurship
providing support and funding, to stimulate the innovation potential of the younger generation.
4. Conclusion
The aging population has had a broad and profound impact on the financial services industry and
economy. Firstly, with the intensification of the aging population trend, the financial service
industries need to adjust their business models and product services to meet the unique needs of older
people. This includes providing more elderly care financial products, risk management, insurance,
and other services, and strengthening financial education and consultation so that older adults can
better plan and manage their wealth. Secondly, the aging population also brings challenges and
opportunities. The economic structure will change, and the demand for labor and consumer markets
will be affected. Financial institutions need to adjust their business focus, adapt to the needs of an
aging society, and explore emerging markets such as the elderly care industry and healthcare. At the
same time, applying financial technology will promote innovation and efficiency improvement in
financial services, providing more convenient and personalized services for older people. However,
the aging population has also brought challenges, such as increased pressure on elderly care and
pension expenditures. The government and financial institutions need to formulate corresponding
policies and measures to ensure the sustainability of pension and social security systems and promote
the stability and development of the financial system. The aging population has brought opportunities
and challenges to the financial services industry and economy. By strengthening innovation and
adjustment of financial services, planning and managing the financial needs of older people, the
development of the financial services industry can be promoted, and stable economic growth can be
upgraded.
References
[1] Perera, H. P., Perera, B. A. K. S., and Palihakkara, A. D. (2023). Financial and economic risk management in
coastal land reclamation projects. Construction Innovation, 23(4), 878-897.
[2] Hsu, Y. H., Yoshida, H., and Chen, F. (2022). The impacts of population aging on China’s economy. Global Journal
of Emerging Market Economies, 14(1), 105-130.
[3] Ishii, R., Ohyama, T., and Endo, H. (2023). Moderating effect of financial literacy on the relationship between
socioeconomic status and depression. Journal of Workplace Behavioral Health, 38(1), 36-51.
[4] Deng, J., Liu, J., Yang, T., and Duan, C. (2022). Behavioural and economic impacts of end-user computing
satisfaction: Innovative work behaviour and job performance of employees. Computers in Human Behavior, 136,
107367.
[5] Khusaini, K., Mardisentosa, B., Bastian, A. F., Taufik, R., & Widiawati, W. (2022). The Impact of Financial
Education and Socioeconomic Status on the Undergraduate Students' Financial Literacy. Media Ekonomi Dan
Manajemen, 37(1), 55-76.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231101
161
[6] Ofori, I. K., Gbolonyo, E. Y., Dossou, M. A. T., Nkrumah, R. K., & Nkansah, E. (2023). Towards inclusive growth
in Africa: Remittances, and financial development interactive effects and thresholds. Journal of Multinational
Financial Management, 68, 100798.
[7] Wyatt, A. (2008). What financial and non‐financial information on intangibles is value‐relevant? A review of the
evidence. Accounting and business Research, 38(3), 217-256.
[8] Ziaei, S. M. (2022). The impacts of household social benefits, public expenditure on labour markets, and household
financial assets on the renewable energy sector. Renewable Energy, 181, 51-58.
[9] Yan, B., Wang, F., Dong, M., Ren, J., Liu, J., & Shan, J. (2022). How do financial spatial structure and economic
agglomeration affect carbon emission intensity? Theory extension and evidence from China. Economic Modelling,
108, 105745.
[10] Chinnakum, W. (2023). Impacts of financial inclusion on poverty and income inequality in developing Asia. The
Singapore Economic Review, 68(04), 1375-1391.
[11] Kumar, J., Rani, V., Rani, G., & Sarker, T. (2023). Determinants of the financial wellbeing of individuals in an
emerging economy: an empirical study. International Journal of Bank Marketing, 41(4), 860-881.
[12] Millar, R., Plumley, D., Wilson, R., & Dickson, G. (2023). Federated networks in England and Australia cricket: a
model of economic dependency and financial insecurity. Sport, Business and Management: An International
Journal, 13(2), 161-180.
[13] Yu, M., Meng, B., & Li, R. (2022). Analysis of China's urban household indirect carbon emissions drivers under the
background of population aging. Structural Change and Economic Dynamics, 60, 114-125.
[14] Li, H., Zhou, X., Tang, M., & Guo, L. (2022). Impact of population aging and renewable energy consumption on
agricultural green total factor productivity in rural China: Evidence from panel var approach. Agriculture, 12(5),
715.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231101
162
Stock Price Prediction Based on ARIMA and Neural Network
Xinyu Liao1, a, *
1College of Art and Science, New York University, New York, US
a. xl3316@nyu.edu
*corresponding author
Abstract: The stock price is affected by many factors and is a very complex nonlinear and
non-stationary system. Predicting stock prices is a classic problem. People hope to predict
stock prices more accurately, so as to make profits through stocks. This article selects five
stocks in the Nasdaq stock market from 2020 to 2023, and tries to use 3 AI models (ARIMA,
CNN, LTSM) to predict and analyze their next day’s closing prices and use the RMSE as the
index to analyze the prediction performance. This paper finds that the three models can
predict the stock price next day well, among which the ARIMA model and LSTM model have
better prediction results, average RMSE for them are about 3.3 and 4.5 while the CNN model
has poorer prediction performance with RMSE 7.2. At the same time, paper is found that
when the model has a turning point for the stock, all the models predict poorly. In the future,
we can consider combining the eigenvalues of more stocks to reduce the impact of turning
points on price prediction.
Keywords: stock price prediction, ARIMA, neural network
1. Introduction
Stock price is a dynamic, non-linear complex system that has sudden and random changes. Predicting
the stock market can help investors make decisions and reduce risks to obtain stable returns. More
importantly, good predictions of stock price can help improve the efficiency of capital markets and
create a more realistic capital model. Therefore, good predictions of stock price are very important.
However, the price of stocks is affected by uncertain factors, and it is difficult for investors to
make effective predictions about stocks. Previous researchers mostly used stand-alone learning
models to predict stock prices, including nonlinear time series models [1,2], artificial neural networks
[3,4], decision trees, genetic algorithms, Markov model, support vector set, etc. However, previous
investigations still appear to be limited. Therefore, this essay selects the historical data of the
Shanghai Composite Index from 1991 to 2016 and predict the closing prices of five Nasdaq stocks
by using the ARIMA model, CNN model and LSTM.
2. Methodology
This article mainly uses ARIMA, CNN and LSTM to analyze and predict five stock closing prices
(Apple, ADM, Amazon, Google, Microsoft) from 2021.1 to July 2023.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231102
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
163
2.1. ARIMA
ARIMA model is a model that predicts the future based on historical time series. It integrated Auto
Regressive model and Moving Average model. The model has the following formula, where p is the
number of historical parameters [5,6].
(1)
If the series is not stable, Will do differencing until it becomes stable, and d is the number of times
of differencing. Through moving average, it reduces the error in the prediction. Below is the formula,
where q is the number of prediction error.
(2)
When referencing the ARIMA model in practice, there are several steps:
1. Observing the stationarity of the data, need to use difference to transfer the non-stationary data
into a stationary data. The parameter d is determined.
2. Use ACF and PACF to observe the possible order of p, q
3. Use the smallest BIC to find the best p, q value.
4. To consider differencing and the AR model depend on prior time steps observations. This paper
is performing a rolling forecast to improve the prediction accuracy by re-creating ARIMA model
when receive a new observation.
2.2. CNN
Neural networks can learn complex patterns and nonlinear relationships in data. CNNs is good at
machine vision problems because of their ability to perform convolution operations, extract features
from local input patches, and modularize representations while making efficient use of data. These
properties enable convolutional neural networks to work in fields such as sequence processing. It
examines time as a spatial dimension to process images. It can be used to find time patterns to make
predictions in stock market.
There are one or more convolutional layers in a convolutional neural network and a fully
connection layer at the top, as well as a pooling layer [7-9] (See Figure 1).
Figure 1: The structure of CNN.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231102
164
2.3. LSTM
A Long Short-Term Memory network is a type of recurrent neural network, it has capability to learn
order dependencies in sequence prediction. Due to the capability to store past information, LSTMs
are very useful in predicting stock prices, since future prices are dependent of previous prices.
The LSTM model can predict any number of steps into the future. An LSTM module (or cell) has
5 basic components: cell state, hidden state, input gate, forget gate and output gate allow it to model
both long-term and short-term data [10] (See Figure 2).
Figure 2: LSTM model.
3. Data and Preprocessing
3.1. Data
In this paper, get stock data from yahoo finance, and five typical stocks of Nasdaq ('AAPL','GOOGL',
'AMD', 'MSFT', 'AMZN') are selected for research. The data contains daily closing prices for these
stocks from 2021.1 to July 2023.7. The data is shown in the Figure 3.
Figure 3: Dataset-stock close price trend.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231102
165
3.2. ARIMA Processing
The data processing of the ARIMA model is to judge the stationarity of the data. If the data is not
stable, the difference operation needs to be performed on the data. After the data is stable, Using ACF
and PACF to determine the parameters of P and Q.
It is obvious that data does not meet the stability requirements through observation, so the data are
differenced (See Figure 4).
Figure 4: Distributions of Assets.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231102
166
Then, this paper implements ADF stationary test to the differenced series, and the results show
that all the series are stationary and is suitable for economic modelling.
3.3. CNN and LTSM Processing
The CNN model and the LTSM model need to normalize data. The normalization process refers to
scaling the original data to reduce it to a fixed range. Normalization can effectively eliminate the two
sums of less indicators. The dimensional unit means that the data has comparable performance for
comprehensive analysis, and at the same time, it can eliminate noise, improve convergence speed,
and improve model accuracy. The normalization method used in this paper is the maximum and
minimum method.
3.4. Data Split
The distribution of training data and test data for the three models is 70%, 30%.
4. Results
4.1. ARIMA
This paper first quantities the lag length for ARIMA model and finally, determines the following
results. APPL (0, 1, 1), AMD (0 ,1,1), AMZN (0,1,1), Google (0,1,1), MSFT (0,1,1).
Every time the stock value of a new day is predicted, the stock value in the test data is added to
the training data. In the meantime, the ARIMA model is reconstructed from the training data to predict
the coming day price. Finally, visualize the predict and actual data as Figure 5.
Figure 5: Results of ARIMA forecasts.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231102
167
The RMSE of the prediction is as Table 1. the prediction effect is very good.
Table 1: RMSE for ARIMA.
Stock symbol
RMSE
APPL
2.774560261
AMD
2.896005671
AMZN
2.860170268
GoogL
2.807792113
MSFT
5.384607601
4.2. CNN
In this paper, built a simple convolutional neural network model. This model includes two
convolutional layers, two pooling layers, one flat and two fully connected layers. convolutional
kernels (also known as filters) for first convolutional layer are 3*3 and 2*2 for second layer. The
first fully connected layer output nodes are 64, and the second fully connected layer output nodes is
'Stock_Data. The visual display of the prediction results and the real value is shown in the Figure 6.
Figure 6: Results of CNN forecasts.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231102
168
This is the RMSE of the prediction as Table 2.
Table 2: RMSE for CNN.
Stock symbol
CNN RMSE
APPL
6.21090165
AMD
4.153477843
AMZN
2.904098039
GoogL
2.497768776
MSFT
20.04826331
In this experiment uses mean square error loss. Visualized the loss trend is as Figure 7.
Figure 7: Training loss vs validation loss of CNN.
Both training and validation losses should gradually decrease, indicating that the model is being
continuously optimized.
The training loss and validation loss converge to a small value with a significant difference
between them. This shows that the generalization ability of the model needs to be improved and can
be used to predict future data, but it is not accurate.
In summary, by observing the changes in the loss function, The model has a certain degree of fit
to the data and has a general ability to generalize.
4.3. LSTM
This paper uses Keras to build LSTM model. Keras has high-level API for TensorFlow which
provides powerful functions, quick use, and easy understanding. First, building the Sequential object
and add layers in order. This paper added four layers to this object.
The first layer is the LSTM layer UNIT=96. There are 96 units in the form, and the activation
function is 'relu', 5 channels for 5 stocks.
The second layer is LSTM layer and also have 96 unit. Return_sequences=False means that only
the next trading day price will be output.
The third layer is the dropout layer. The left and right of the dropout are to randomly shield a
certain proportion of the upper layer neurons during each round of training, so that the gradient
backpropagation path can be different every time, thereby forcing different neurons to cooperate, thus
speeding up the training. To improve the effect, Dropout = 0.2 was set in this paper.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231102
169
The fourth layer is the Dense layer (full connection layer) and finally outputs the prediction data
of 5 stocks.
Then compiles the model, selects 'adam' as the optimizer, MSE as the loss function, visually as
Figure 8.
Figure 8: LSTM change in loss.
The next step is to plot the predicted and actual values, as shown in the Figure 9.
Figure 9: LSTM stock prediction.
The model is evaluated in RMSE as Table 3:
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231102
170
Table 3: LSTM RMSE.
Stock symbol
LSTM RMSE
APPL
3.711327865540973
AMD
4.7741382933398775
AMZN
3.495610540683055
GoogL
3.508695503078867
MSFT
7.096476378071467
5. Conclusion
This paper compares three stock forecasting algorithms, traditional ARIMA model, CNN model, and
LTSM model. According to the verification results of five stocks, the ARIMA model and LSTM
model have better forecasting effects. CNN is good at spatial feature extraction, but not good at
processing time-related data. Through practice, it shows the deep learning algorithm has good
potential to be used to predict stock price, but there are cons: the predicted value lags behind the real
value, and there is a large gap in the prediction of turning points. In terms of stock forecasting,
potential directions for future deep learning enhancements include applying combined models or
trying other deep learning models, or adding other features, such as trading volume or news sentiment,
to make the model more comprehensive and accurate.
References
[1] Vasilev, I., Slater, D., Spacagna, G., Roelants, P., and Zocca, V. (2019). Python Deep Learning: Exploring deep
learning techniques and neural network architectures with Pytorch, Keras, and TensorFlow. Packt Publishing Ltd.
[2] Devi, B. U., Sundar, D., and Alli, P. (2013). An effective time series analysis for stock trend prediction using ARIMA
model for nifty midcap-50. International Journal of Data Mining & Knowledge Management Process, 3(1), 65.
[3] Pai, P., and Hong, W. (2005). An improved neural network model in forecasting arrivals. Annals of Tourism
Research, 32(4), 1138-1141.
[4] Liang, J., Song, W., and Wang, M. (2011). Stock price prediction based on procedural neural networks. Advances
in Artificial Neural Systems.
[5] Ma, Q. (2020). Comparison of ARIMA, ANN and LSTM for stock price prediction. In E3S Web of Conferences, 218,
01026.
[6] Siami-Namini, S., Tavakoli, N., and Namin, A. S. (2018). A comparison of ARIMA and LSTM in forecasting time
series. In 2018 17th IEEE international conference on machine learning and applications, 1394-1401.
[7] Mehtab, S., and Sen, J. (2020). Stock price prediction using convolutional neural networks on a multivariate
timeseries. arXiv preprint arXiv:2001.09769.
[8] Sen, J., and Mehtab, S. (2022). Stock price prediction using convolutional neural networks. Machine Learning in
the Analysis and Forecasting of Financial Time Series, 68-101.
[9] Sun, Y. (2023). Optimization of Convolutional neural network. Proceedings of the 37th China (Tianjin) 2023 'IT,
Network, Information Technology, Electronics, Instrument and Meter Innovation Academic Conference, 54-57.
[10] Chen, K., Zhou, Y., and Dai, F. (2015). A LSTM-based method for stock returns prediction: A case study of China
stock market. In 2015 IEEE international conference on big data, 2823-2824.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231102
171
Application of Mean-variance Model in Optimizing Stock
Portfolio
Xingyu Qian1,a,*
1D’Amore-Mckim School of Business, Northeastern University, Boston, US
a. email, qian.xing@northeastern.edu
*corresponding author
Abstract: Constructing an optimized stock portfolio is pivotal in achieving enhanced returns
with managed risks. Utilizing data from 2022, the study employed the mean-variance model
to determine the optimal allocations for five prominent stocks: Intel (INTC), Johnson &
Johnson (JNJ), Coca-Cola (KO), Netflix (NFLX), and Procter & Gamble (PG). In the
formulations of both maximum Sharpe ratio and minimum variance portfolios, the stocks
with the predominant allocations are NFLX and PG, with respective weights of 47.04% and
45.89% for the former, and 58.05% and 29.61% for the latter. Upon establishing these
weightings, the cumulative returns for 2023 were calculated, revealing that the constructed
portfolios notably outperformed the broader stock market. This research underscores two key
observations: firstly, the NFLX and PG stocks emerge as cornerstones in the optimal portfolio
composition; and secondly, these allocations, when evaluated with 2023 data, underscore
their efficacy. Robustness checks, which expanded the asset pool, further validated these
findings. Ultimately, this study serves as a valuable guide for investors in the financial market,
presenting a structured blueprint for effective portfolio assembly.
Keywords: mean-variance, portfolio, US stock market
1. Introduction
The importance of building a well-diversified stock portfolio in the financial market cannot be
overstated. It is crucial to optimize a stock portfolio since it provides an organized, research-based
approach to investing, raising the possibility of higher returns while reducing risk. Investors may
unintentionally expose themselves to greater risks or lose out on possible gains without optimization
[1].
Numerous studies have already explored various methodologies for portfolio construction.
According to Geng’s research, he suggests a brand-new metric to more thoroughly evaluate the risk-
return trade-off by adding VaR into the Sharpe ratio. This method would consider potential for
significant losses as well as volatility [2]. Smith and Brown emphasized the evolving nature of the
financial landscape and the need to adapt the principles of the Modern Portfolio Theory (MPT) to
better meet the current demands of the volatile market [3].
With the surge of technological advancements, some researchers have delved into the integration
of machine learning techniques in portfolio optimization. Fernandez and Raj, for instance, explored
the impact of machine learning in dynamic asset allocation, stressing its potential in enhancing
portfolio management practices [4]. This aligns with Davies and Wright’s findings, which presented
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231128
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
172
a comprehensive analysis comparing traditional portfolio optimization techniques against the newer,
more advanced methodologies [5].
Moreover, the significance of behavioral factors in investment strategies cannot be understated.
As Patel and Kim argued, modern portfolio construction can greatly benefit from considering investor
behavior and biases, leading to a more resilient and stable portfolio performance in the long run [6].
However, despite these advancements, one aspect that seems to be consistently underrepresented
is the evaluation of these portfolio strategies in real-time scenarios. Wang and Liu, for instance,
underscored the necessity of analyzing portfolio performances using real-time data, especially given
the age of fast-paced trading and high-frequency data [7]. This study aims to bridge this gap by first
determining the optimal weights of selected stocks based on historical data through The Mean-
Variance Model, and then evaluating this portfolio using subsequent real-world data.
2. Data and Method
2.1. Data
The top five representative equities from the financial market, as measured by market capitalization,
are chosen for this article. Intel (INTC), Johnson & Johnson (JNJ), Coca-Cola (KO), Netflix (NFLX),
and Procter & Gamble (PG) are the five stocks’ tickers. Closing prices for the period of January 1
through December 31 of 2022 have been gathered. The average return and covariance matrices are
computed for the training set in order to build the efficient frontier. The test set compares the
cumulative returns of the chosen asset allocations to the return of the FTSE Index in order to assess
how well they performed. Table 1, Table 2, and Figure 1 exhibit the fundamental data for the five
selected equities, accordingly.
Table 1: Selected stocks.
Company
INTC
Intel
JNJ
Johnson & Johnson
KO
Coca-Cola
NFLX
Netflix
PG
Procter & Gamble
Table 2: Descriptive statistics of the daily return of the 5 stocks.
Max
Min
Mean
Std Dev
Cumulative Return
INTC
0.106585
-0.085621
-0.002351
0.024079
-48.36%
JNJ
0.049703
-0.029919
0.000281
0.010986
5.68%
KO
0.038671
-0.069626
0.000475
0.012423
10.44%
NFLX
0.130864
-0.351166
-0.001760
0.044222
-50.64%
PG
0.042699
-0.062322
-0.000094
0.013855
-4.56%
Table 2 presents the descriptive statistics of daily returns for five major stocks: Intel Corporation
(INTC), Johnson & Johnson (JNJ), The Coca-Cola Company (KO), Netflix, Inc. (NFLX), and Procter
& Gamble Co (PG). The metrics include the maximum (Max), minimum (Min), mean, standard
deviation (Std Dev), and cumulative return for the year. Notably, NFLX exhibited the highest
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231128
173
volatility, with a maximum daily return of 13.09% and a minimum of -35.12%. In terms of cumulative
return, KO performed the best with a 10.44% return, while NFLX had the most significant decline at
-50.64%.
Figure 1: Cumulative returns of selected stocks throughout 2022.
The Figure 1 illustrates the performance evolution of Intel (INTC), Johnson & Johnson (JNJ),
Coca-Cola (KO), Netflix (NFLX), and Procter & Gamble (PG) over the course of the year. Each line
represents the cumulative return of the respective stock from the beginning to the end of 2022.
2.2. Method
Portfolio optimization seeks to find the ideal allocation of assets that minimizes risk and maximizes
return. One of the pioneering frameworks in this area is the mean-variance model, which offers a
structured approach to balance risk against the expected return on investment [8].
The core of the mean-variance model rests upon the principle that each asset’s return has an
expected value (mean) and a variance. The combined return of the portfolio is the weighted sum of
individual returns, given by:
(1)
Where symbolizes the weight of each asset and its respective variance [9]. Markowitz’s
seminal work on this topic emphasizes that by considering each stock’s variance alongside its return,
investors can make more informed decisions [8].
2.2.1. Expected Portfolio Return
Considering a portfolio of n stocks, its expected return can be derived as:
(2)
Where denotes the expected return of the portfolio, denotes the weight (or percentage)
of the ith stock in the portfolio, and denotes the expected return of the ith stock, the sum runs
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231128
174
from to , denoting that you will calculate this product for each stock and add them all up to
obtain the portfolio’s overall expected return.
The expected return of a portfolio, given the weights of individual stocks and their respective
expected returns, can be represented mathematically as:
(3)
2.2.2. Portfolio Variance
The formula for portfolio variance, assuming no correlation between individual stocks, can be
presented as:
(4)
Where is number of stocks in this portfolio, is Variance of the stock.
2.2.3. Sharpe Ratio
The Sharpe ratio gives a measure of the risk-adjusted performance of an investment, with higher
values indicating better return performance for the amount of risk taken. The Sharpe ratio is often
symbolically represented using the following formula:
(5)
Where stand for the Sharpe ratio, denotes the expected return of portfolio, is the risk-free
rate, and is the standard deviation of portfolio [10]. Sharpe’s work on the Capital Asset Pricing
Model (CAPM) provides foundational knowledge on understanding the relationship between risk and
expected return, with the Sharpe ratio being one of its cornerstones [11].
3. Result
For clarity, the Monte Carlo simulation was conducted on 2022 data, iterating 100,000 times. Two
targeted portfolios are computed and identified. The stock weights in the least volatility portfolio are:
INTC: 4.35%, JNJ: 0.40%, KO: 7.60%, NFLX: 58.05%, and PG: 29.61%. Table 3 lists the weights
of the selected companies in the portfolio with the highest Sharpe ratio as follows: INTC: 0.90%, JNJ:
5.58%, KO: 0.59%, NFLX: 47.04%, and PG: 45.89%. According to Table 4, the volatility for the
portfolio with the lowest volatility is 1.03%, while the portfolio with the highest Sharpe ratio has a
Sharpe ratio of 0.03%. With significant allocations to JNJ and KO, the minimal volatility portfolio
prioritizes stability. The maximum Sharpe ratio portfolio, on the other hand, favors KO and JNJ and
PG to a lesser amount in order to maximize returns for the risk incurred. The latter portfolio appears
to deliver a respectable risk-adjusted return based on the Sharpe ratio supplied, but investors would
need to be ok with the higher volatility compared to the minimal volatility portfolio.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231128
175
Table 3: Weight of each stock in the two optimal portfolios (%).
Stock
Min Volatility Portfolio (%)
Max Sharpe Ratio Portfolio (%)
INTC
4.35
0.90
JNJ
0.40
5.58
KO
7.60
0.59
NFLX
58.05
47.04
PG
29.61
45.89
Table 4: Return and volatility of the two portfolios.
Return (%)
Volatility (%)
Sharpe Ratio (%)
Min Volatility Portfolio
0.02
1.03
0.02
Max Sharpe Ratio Portfolio
0.03
1.04
0.03
The second stage, the computation of the portfolio return, may be done after the two asset
allocations have been obtained. The daily portfolio returns, and cumulative returns may be calculated
using the test set from January 1 to August 20, 2023, along with the stock weights. The same time
period’s return information for the S&P 500 Index is gathered for comparison. Figure 2 shows that
the study’s findings outperformed the general market.
Figure 2: Comparison between S&P 500 index returns and the portfolio returns.
4. Robustness
A robustness check will be done for cogency. Firstly, add two additional assets, The Walt Disney
Company (DIS) and Boeing (BA). Repeat the Monte Carlo simulation and the cumulative return
computation next. Repeat the Monte Carlo simulation and the cumulative return computation
next.The adjusted weights for the two ideal portfolios are displayed in Table 5, where it can be seen
that the greatest components are PG:37.56% for the portfolio with the highest Sharpe ratio and
NFLX:49.55% for the portfolio with the lowest risk. In the end, total the returns from the two
portfolios and compare them to the S&P 500 index. Figure 3 showing a similar result with previous
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231128
176
data. Moreover, after changing the number of assets, the investment results still outperformed the
general market. As a result, both the process and the outcomes are reliable and efficient.
Table 5: Weight of the 7 stocks (2 more stocks) in the two optimal portfolios (%).
Stock
Min Volatility Portfolio (%)
Max Sharpe Ratio Portfolio (%)
INTC
2.40
4.28
JNJ
1.55
0.18
KO
3.62
2.59
NFLX
49.55
33.80
PG
23.19
37.56
DIS
2.52
1.04
BA
17.17
20.55
Figure 3: Comparison between FTSE 100 index return and the altered portfolio returns.
5. Conclusion
This article set out with an overarching objective to address the persistent gap in assessing stock
portfolio strategies, especially in real-time, dynamic market conditions. Leveraging the fundamental
principles of classic portfolio theories, we introduced a novel application of the Mean-Variance
Model for determining optimal weights of our selected stocks using historical data from 2022.
Through the Monte Carlo simulation on the 2022 dataset, the study identified two standout portfolios:
one optimized for minimum volatility and the other for maximum Sharpe ratio. The former
emphasized stability, heavily weighting stocks like JNJ and KO, while the latter targeted the balance
between risk and return, leaning towards KO, JNJ, and a minor portion of PG. In a direct comparison
using real-world data up until August 20, 2023, both these portfolios consistently outperformed the
S&P 500 Index, highlighting the practical utility and potential profitability of our methodological
approach.
While our study’s results are indeed promising and provide valuable insights into the realm of
portfolio optimization, it’s essential to acknowledge the limitations and scope for refinement. Firstly,
the study was bounded by the selection of just five primary stocks, though the robustness check with
the inclusion of two more did offer some additional perspectives. However, it’s arguable that a
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231128
177
broader selection might uncover other significant combinations or nuances in portfolio behavior.
Secondly, while the Mean-Variance Model and Sharpe ratio are powerful tools, they represent just a
fraction of the available methodologies for portfolio evaluation. Future studies might benefit from
integrating more diverse models and perhaps a more extensive data set, which might include
macroeconomic variables, to attain a holistic view. Nevertheless, this paper serves as a step forward
in portfolio strategy evaluation, paving the way for more comprehensive research in the field.
References
[1] Malkiel, B. G. (2004). A Random Walk on Wall Street. Norton.
[2] Deng, G., Dulaney, T., McCann, C., and Wang, O. (2013). Robust portfolio optimization with value-at-risk-adjusted
Sharpe Ratios. Journal of Asset Management, 14(5), 293-305.
[3] Smith, J., and Brown, A. (2018). Rethinking Modern Portfolio Theory: Adapting to the Changing Financial
Landscape. Journal of Investment Strategies, 7(3), 45-60.
[4] Fernandez, L., and Raj, P. (2019). Dynamic Asset Allocation and Machine Learning: A Practitioner’s Perspective.
Journal of Finance & Banking, 12(4), 21-37.
[5] Davies, T., and Wright, M. (2020). A Comparative Study of Traditional and Advanced Portfolio Optimization
Techniques. Journal of Financial Engineering, 10(2), 165-178.
[6] Patel, H., and Kim, D. (2021). Incorporating Behavioral Factors in Portfolio Construction: A Modern Take on
Classic Theories. Behavioral Finance Journal, 5(1), 32-49.
[7] Wang, F., and Liu, Y. (2022). Real-time Performance Analysis of Stock Portfolios: A Comprehensive Approach
Using High-frequency Data. Financial Markets Review, 9(3), 12-27.
[8] Markowitz, H. (1952). Portfolio Selection. The Journal of Finance, 7(1), 77-91.
[9] Elton, E.J., Gruber, M.J., Brown, S.J., and Goetzmann, W.N. (2009). Modern Portfolio Theory and Investment
Analysis. John Wiley & Sons.
[10] Lo, A. W. (2002). The statistics of Sharpe ratios. Financial Analysts Journal, 58(4), 36-52.
[11] Sharpe, W. F. (1964). Capital asset prices: A theory of market equilibrium under conditions of risk. The journal of
finance, 19(3), 425-442.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231128
178
Validity Testing of Classical Asset Allocation Models: An
Empirical Study
Chong Gao1, Shengkai Xu2, a, *
1CBE, Australian National University, Canberra, 2601, Australia
2W.P Carey School of business, Arizona State University, Arizona, 85281 United-States
a. Shengka3@asu.edu
*corresponding author
Abstract: This paper uses a 20 years of daily total return data for the S&P 500 Index (ticker
symbol “SPX”) and 10 stocks, make all the necessary calculations to plot a portfolio area that
gather an efficient boundary, a minimum risk or variance boundary, and a minimum return
boundary together for a given set of constraints. Analyze all the outcomes in order to compare
the various restrictions for each optimization issue (MM and IM), as well as the two solutions
to the same optimization problem. The Excel solver was the primary tool utilized during
calculation to resolve optimization issues for each point on the minimal risk or variance
border. Also, this paper use an Excel solution table to calculate a large number of multipoints
on any desired boundary. Through calculation and research, we found that, Markowitz model
makes full use of covariance matrix to generate excellent portfolio. However, the results are
numerically unstable. At the same time, the hypothesis of normality, stationarity and mean
square error are verified. The exponential model simplifies the Markowitz model and
produces more robust results. However, it introduces additional assumptions about the
independence, normality, and homoscedasticity of the regression residuals, which are also
invalid. The reduction (CDaR) model is hypothesis-free. The numerical stability can be
obtained by transforming the nonlinear optimization problem into a linear programming
problem.
Keywords: Sharpe ratio, minimal risk, efficient frontier, minimal return frontier, Markowitz
model
1. Introduction
The basic idea of this theory that Markowitz find out is the time when he was reading John Burr
Williams, “The Theory of Investment Value” [1]. Markowitz addressed these limitations by
introducing a formula that allowed investors to balance risk tolerance and reward expectations,
leading to the creation of an optimal portfolio. The Modern Portfolio Theory (MPT) he developed
was grounded in two key principless [2]:
(1) The primary goal for every investor is to maximize returns while managing risk.
(2) Diversification across unrelated securities can effectively reduce portfolio risk.
The Markowitz model has been widely employed in portfolio optimization processes and has been
shown to be successful in real-world applications. Markowitz portfolio theory may be applied to
consistently outperform the market in the Chinese stock market. [3]. Although the current securities
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
179
market in China is not standardized, the system risk still occupies a large proportion of investment
risks, the Markowitz model can be used to obtain a better performance than the average securities
market portfolio.
An Index Model is a statistical model used to analyze security returns, differing from economic
models that are based on market equilibrium principles. The Single Index Model (SIM) delineates
two distinct factors contributing to the uncertainty of a security’s return[4]:
(1) Systematic (macroeconomic) uncertainty, which is presumed to be effectively represented by
a singular index reflecting stock returns.
(2) Unique (microeconomic) uncertainty, represented by a random component specific to the
individual security.
This research details the operational and financial performance of 11 organizations. Based on raw
data, it analyzes the annualized data and correlation coefficient. Therefore, investors may select one
restriction and discover the optimum portfolio for their circumstances and preferred level of risk and
return. The daily data are combined in this study to create monthly observations, from which we
derive the necessary optimization inputs for the complete Markowitz Model (“MM”) and the Index
Model (“IM”). We identify the areas of acceptable portfolios (efficient frontier, minimum risk
portfolio, optimum portfolio, and minimal return portfolios frontier) for the extra limitations using
these optimization inputs for MM and IM. [5].
First of all, familiarize the market and select the required representative company data. Secondly,
optimizing the data for problem solving (from a daily to monthly frequency summary), computing
each new set of optimization constraints as well as all the inputs required for the MM and IM
optimization problems. Then, two key boundary points (maximum Sharpe ratio and minimum risk)
are calculated, as well as two boundaries: the efficiency boundary and the minimum return boundary.
To compare the various constraints for each optimization problem (MM and IM), as well as two
distinct solutions to the same optimization issue, is the final step in the analysis of all the findings.
2. Data Preprocessing
The data preprocessing part mainly focuses on generating the inputs for portfolio optimization models
and check if the assumptions of the models are reasonable.
2.1. Inputs for Markowitz Model
Based on monthly excess returns, the author computes the annualized average, annualized standard
deviation, and correlation matrix. The return Rt of a portfolio at time t can be defined to be the total
value Tt of the portfolio divided by the total value at an earlier time t−1, i.e [6]. The results are shown
in table 1. There is also a visualization of correlation matrix in figure 1. These are inputs for
Markowitz model.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
180
Table 1: Inputs for Markowitz model.
Panel A: Annualized average and annualized standard deviation of monthly excess return
AMZ
N
AAP
L
CTX
S
JPM
BRK/
A
PGR
UPS
FDX
JBH
T
LST
R
SPX
Averag
e
0.338
0.340
0.156
0.11
9
0.090
0.15
4
0.09
8
0.13
0
0.225
0.174
0.07
5
Std
0.414
0.345
0.415
0.29
0
0.162
0.21
1
0.21
4
0.26
7
0.307
0.239
0.14
9
Panel B: Correlation matrix of monthly excess return
AMZ
N
AAP
L
CTX
S
JPM
BRK/
A
PGR
UPS
FDX
JBH
T
LST
R
SPX
AMZN
1.000
AAPL
0.377
1.000
CTXS
0.217
0.332
1.000
JPM
0.252
0.244
0.324
1.00
0
BRK/A
0.118
0.173
0.181
0.45
2
1.000
PGR
0.200
0.240
0.271
0.39
3
0.264
1.00
0
UPS
0.296
0.231
0.264
0.36
1
0.404
0.39
2
1.00
0
FDX
0.280
0.330
0.331
0.44
0
0.385
0.36
5
0.67
5
1.00
0
JBHT
0.308
0.268
0.290
0.44
2
0.239
0.28
0
0.45
9
0.53
7
1.000
LSTR
0.256
0.287
0.252
0.37
5
0.234
0.28
9
0.44
1
0.48
2
0.590
1.000
SPX
0.485
0.542
0.437
0.69
7
0.523
0.50
2
0.57
5
0.61
4
0.521
0.495
1.00
0
Readers can see from panel A of table 1 that in general, stocks with higher average excess return
tend to have higher standard deviation. And from panel B of table 1 and figure 1 readers can see that
all the stocks are highly correlated with the market and the four industrial stocks are highly correlated
with each others.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
181
Figure 1: Heatmap of correlation matrix of monthly excess return.
Photo credit: Original
2.2. Inputs for Index Model
The author runs linear regressions with the monthly excess returns of S&P 500 as the explanatory
variable, and monthly excess return of each of the ten stocks, one at a time, as dependent variable.
The results, including beta, annualized alpha and annualized residual standard deviation are shown in
table 2.
Table 2: Inputs for index model.
AMZ
N
AAP
L
CTX
S
JPM
BRK/
A
PGR
UPS
FDX
JBHT
LSTR
Beta
1.351
3
1.256
9
1.220
6
1.360
8
0.5721
0.712
0
0.829
6
1.103
6
1.076
3
0.797
5
Annualize
d Apla
0.236
0
0.245
4
0.064
3
0.016
0
0.0469
0.100
4
0.035
9
0.046
3
0.144
1
0.113
7
Annualize
d residual
std
0.362
0.290
0.373
0.208
0.138
0.182
0.175
0.211
0.262
0.208
2.3. Model Diagnostics
First of all, the research tests for normality of monthly excess return in three ways: (i) Q-Q plots [7];
(ii) Skewness and kurtosis [8]; (iii) Formal statistical tests including Kolmogorov-Smirnov test [9],
Cramervon Mises test [10], Shapiro-Wilks test, and Jarque-Bera test [11]. The results are shown in
figure 2, table 3 and table 4, respectively.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
182
Table 3: Skewness and kurtosis of monthly excess return.
AMZ
N
AAP
L
CTX
S
JPM
BRK/
A
PGR
UPS
FDX
JBH
T
LST
R
SPX
Skewnes
s
0.574
-
0.249
0.530
-
0.28
2
0.140
-
0.24
8
0.04
1
0.04
0
-
0.01
0
-
0.022
-
0.57
0
Kurtosis
3.558
0.855
3.842
1.44
9
0.647
0.14
5
2.88
7
1.95
1
3.14
3
0.006
1.35
1
Note: Kurtosis are computed with 3.0 subtracted to give 0.0 for normal distribution.
From the Q-Q plots readers can see that the points deviate from the line in the top-right and bottom-
left corner. This form of deviation implies heavy tail of monthly excess return. From the sample
skewness and kurtosis readers can see those monthly excess returns of almost all the stocks have
heavy tail. And only monthly excess return of BRK/A, PGR and LSTR seems to be normal. As for
the p-values of formal test, Shapiro-Wilks test and Jarque-Bera test are much more sensitive to
outliers, and only the monthly excess return of PGR and LSTR are not rejected by any of the four
tests to be normal. Nonnormality has one negative impact on Markowitz and index model: Both the
two models presume that investors only care about expected return and standard deviation of
portfolios. This may be with normal distribution of excess return true since expected return and
standard deviation are enough to determine a normal distribution. However, with nonnormal
distribution, investors may take higher moments into account.
Figure 2: Q-Q plots of monthly excess returns.
Photo credit: Original
The author then tests the autocorrelation, nonstationarity and heteroskedasticity of monthly excess
return. It turns out that the time series of monthly excess return does not exhibit property of
autocorrelation. This is consistent with the assumption of the models. However, monthly excess
return of UPS exhibits strong nonstationarity with a p-value of augmented Dickey-Fuller test of 0.198.
This is extremely weird that the return itself has a unit root.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
183
Table 4: P-values of formal normality tests for monthly excess return.
AMZ
N
AAP
L
CTX
S
JPM
BRK/
A
PGR
UPS
FDX
JBH
T
LST
R
SPX
Kolmogoro
v-Simirnov
0.296
0.493
0.073
0.22
1
0.474
0.61
5
0.10
5
0.27
5
0.18
9
1.00
0
0.03
3
Cramer-von
Mises
0.214
0.792
0.050
0.17
4
0.303
0.74
7
0.07
0
0.20
1
0.17
9
1.00
0
0.05
6
Shapiro-
Wilks
0.000
(***)
0.104
0.000
(***)
0.00
0
(***
)
0.025
(*)
0.28
5
0.00
0
(***
)
0.00
0
(***
)
0.00
0
(***
)
0.97
3
0.00
0
Jarque-Bera
0.000
(***)
0.007
(**)
0.000
(***)
0.00
0
(***
)
0.083
0.26
2
0.00
0
(***
)
0.00
0
(***
)
0.00
0
(***
)
0.99
0
0.00
0
(***
)
And it is consistent with the historical price of UPS but inconsistent with the models. Moreover,
heteroskedasticity is common among monthly excess returns, as shown in table 5, where the p-values
of autoregressive conditional heteroskedasticity tests are shown. This is another violation of the
assumptions, which invalidates the use of standard deviation. In fact, heavy tail can be derived from
heteroskedasticity.
Table 5: P-values of ARCH test for monthly excess return.
AMZ
N
AAP
L
CTX
S
JPM
BRK/
A
PGR
UPS
FDX
JBH
T
LST
R
SPX
ARC
H
0.555
0.013
(*)
0.000
(***)
0.00
0
(***)
0.001
(**)
0.00
3
(**)
0.05
5
0.14
7
0.008
(**)
0.110
0.00
0
(*)
The index model enlists has additional assumptions: (i) Residuals are uncorrelated with the market;
(ii) Residuals of different stocks are uncorrelated; (iii) Residuals are normally distributed; (iv)
Residuals do not exhibit autocorrelation, nonstationarity and heteroskedasticity. For assumption (i),
the author computes the correlation coefficients between the residuals and the market, and it turns out
that the macroeconomic factor and firm-specific factors are indeed uncorrelated. For assumption (ii),
the author computes the correlation matrix of residuals of different stocks, the results are shown in
table 6, and there is also a visualization in figure 3.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
184
Table 6: Correlation matrix of residuals.
AMZN
AAPL
CTXS
JPM
BRK/A
PGR
UPS
FDX
JBHT
LSTR
AMZN
1.000
AAPL
0.156
1.000
CTXS
0.006
0.126
1.000
JPM
-0.137
-0220
0.031
1.000
BRK/A
-0.182
-0.153
-0.062
0.144
1.000
PGR
-0.058
-0.045
0.067
0.070
0.001
1.000
UPS
0.025
-0.116
0.018
-
0.0698
0.148
0.146
1.000
FDX
-0.025
-0.005
0.088
0.021
0.095
0.082
0.498
1.000
JBHT
0.074
-0.020
0.081
0.128
-0.046
0.025
0.228
0.322
1.000
LSTR
0.022
0.026
0.046
0.048
-0.034
0.054
0.220
0.260
0.447
1.000
The correlation coefficients between the firm-specific factors of different firms are low, expect for
those industrial firms. This will lead to inferior portfolios because index model presumes that firms
are related to each others only via the macroeconomic factor. For assumption (iii), the author uses the
same method as when testing the normality of monthly excess returns. The results are similar: The
Q-Q plots imply heavy tail, only monthly excess return of BRK/A, PGR and LSTR seems to be
normal when skewness and kurtosis are considered, and only the monthly excess return of PGR is not
rejected by any of the four tests to be normal. Nonnormality of residuals can exert negative impact
on the estimation of beta, alpha, and residual standard deviation. For assumption (iv), the time series
of residuals do not exhibit autocorrelation and nonstationarity, but heteroskedasticity is common
among residuals. Residual heteroskedasticity invalidates the use of residual standard deviation.
Figure 3: Heatmap of correlation matrix of residuals.
Photo credit: Original
3. Portfolio Optimization
Based on the inputs, the author uses sequential least square programming algorithm implemented in
Python to generate critical portfolios and frontiers for Markowitz and index model. The results are
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
185
not necessarily the same as general reduced gradient implemented in Excel, but the differences are
negligible.
3.1. Markowitz Model
Table 7 shows the weights of the maximal Sharpe portfolios and the minimal variance portfolios
under the following five different constraints:
(1) ∑ |𝜔!|
""
!#" ≤2;
(2) |𝜔!|≤ 1, for ∀𝑖;
(3) No constraints;
(4) 𝜔!≥ 0, for ∀𝑖;
(5) 𝜔"= 0.
Table 7: Weights in critical portfolios with Markowitz model.
Panel A: Maximal Sharpe ratio portfolios
Constrai
nt
AMZ
N
AAP
L
CTX
S
JPM
BRK/
A
PGR
UPS
FDX
JBH
T
LST
R
SPX
1
0.164
0.300
0.000
0.00
0
0.413
0.33
0
0.00
0
-
0.01
4
0.12
5
0.167
-
0.48
6
2
0.223
0.397
-
0.012
-
0.00
5
0.625
0.46
0
-
0.03
1
-
0.10
6
0.20
9
0.239
-
1.00
0
3
0.370
0.654
0.004
0.17
3
0.915
0.68
2
0.01
2
-
0.08
6
0.31
0
0.340
-
2.37
5
4
0.130
0.252
0.000
0.00
0
0.192
0.22
7
0.00
0
0.00
0
0.08
8
0.110
0.00
0
5
0.147
0.267
-
0.034
-
0.15
6
0.363
0.32
0
-
0.12
1
-
0.13
2
0.17
9
0.168
0.00
0
Panel B: Minimal variance portfolios
Constrai
nt
AMZ
N
AAP
L
CTX
S
JPM
BRK/
A
PGR
UPS
FDX
JBH
T
LST
R
SPX
1
-0.022
-
0.037
-
0.010
-
0.18
6
0.364
0.14
2
0.02
8
-
0.09
8
-
0.00
5
0.109
0.71
4
2
-0.022
-
0.037
-
0.010
-
0.18
6
0.364
0.14
2
0.02
8
-
0.09
8
-
0.00
5
0.109
0.71
4
3
-0.022
-
0.037
-
0.010
-
0.18
6
0.364
0.14
2
0.02
8
-
0.09
8
-
0.00
5
0.109
0.71
4
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
186
Table 7: (continued).
4
0.000
0.000
0.000
0.000
0.386
0.135
0.015
0.000
0.000
0.078
0.386
5
0.024
0.045
0.008
-
0.069
0.561
0.232
0.120
-
0.086
0.007
0.158
0.000
Table 8 summaries the annualized average return, annualized standard deviation and Sharpe ratio
of these critical portfolios.
Table 8: Summary statistics of critical portfolios with Markowitz mode.
Panel A: Maximal Sharpe ratio portfolios
Constraint
Annualized portfolio
average return
Annualized
portfolio std
Portfolio Sharpe ratio
1
0.264
0.187
1.414
2
0.332
0.221
1.501
3
0.496
0.332
1.539
4
0.221
0.176
1.254
5
0.239
0.180
1.326
Panel B: Minimal variance portfolios
Constraint
Annualized portfolio
average return
Annualized
portfolio std
Portfolio Sharpe ratio
1
0.073
0.122
0.595
2
0.073
0.122
0.595
3
0.073
0.122
0.595
4
0.100
0.131
0.761
5
0.132
0.134
0.989
Figure 4 shows the efficient frontiers, inefficient frontiers, and minimal variance frontiers of these
critical portfolios, as well as the capital allocation line. Note that the capital allocation line is
constructed using the latest annualized notional risk-free return.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
187
Figure 4: Frontiers with Markowitz model.
Photo credit: Original
3.2. Index Model
Table 9 shows the weights in maximal Sharpe portfolios and minimal variance portfolios with
different constraints.
Table 9: Weights in critical portfolios with index model.
Panel A: Maximal Sharpe ratio portfolios
AMZ
N
AAP
L
CTX
S
JPM
BRK/
A
PGR
UPS
FDX
JBH
T
LST
R
SPX
Constraint
1
0.178
0.306
0.004
-
0.02
7
0.225
0.31
6
0.01
2
0.00
0
0.19
1
0.26
6
-
0.47
3
Constraint
2
0.218
0.366
0.031
-
0.08
8
0.333
0.40
0
0.09
5
0.05
6
0.25
0
0.33
8
-
1.00
0
Constraint
3
0.431
0.701
0.110
0.08
8
0.585
0.72
5
0.27
9
0.24
9
0.50
4
0.62
9
-
3.30
2
Constraint
4
0.155
0.271
0.000
0.00
0
0.032
0.21
4
0.00
0
0.00
0
0.14
0
0.18
9
0.00
0
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
188
Table 9: (continued).
Constraint
5
0.186
0.317
-
0.005
-
0.21
2
0.117
0.27
7
-
0.05
3
-
0.05
4
0.18
6
0.24
2
0.00
0
Panel B: Minimal Varience Portfolios
AMZ
N
AAP
L
CTX
S
JPM
BRK/
A
PGR
UPS
FDX
JBH
T
LST
R
SPX
Constraint
1
-0.042
-
0.048
-
0.024
-
0.13
0
0.346
0.13
4
0.08
5
-
0.03
6
-
0.01
6
0.07
5
0.65
5
Constraint
2
-0.042
-
0.048
-
0.024
-
0.13
0
0.346
0.13
4
0.08
5
-
0.03
6
-
0.01
6
0.07
5
0.65
5
Constraint
3
-0.042
-
0.048
-
0.024
-
0.13
0
0.346
0.13
4
0.08
5
-
0.03
6
-
0.01
6
0.07
5
0.65
5
Constraint
4
0.000
0.000
0.000
0.00
0
0.377
0.14
5
0.09
6
0.00
0
0.00
0
0.07
8
0.30
4
Constraint
5
-0.016
-
0.009
-
0.001
-
0.05
0
0.480
0.21
8
0.18
0
0.03
4
0.03
0
0.13
6
0.00
0
Table 10 summaries the annualized average return, annualized standard deviation and Sharpe ratio
of these critical portfolios.
Table 10: Summary statistics of critical portfolios with index model.
Panel A: Maximal Sharpe ratio portfolios
Annualized portfolio average
return
Annualized portfolio
std
Portfolio Sharpe
ratio
Constraint1
0.286
0.200
1.430
Constraint2
0.341
0.224
1.521
Constraint3
0.610
0.382
1.596
Constraint4
0.245
0.192
1.277
Constraint5
0.269
0.202
1.331
Panel B: Minimal varience portfolios
Annualized portfolio average
return
Annualized portfolio
std
Portfolio Sharpe
ratio
Constraint1
0.065
0.124
0.521
Constraint2
0.065
0.124
0.521
Constraint3
0.065
0.124
0.521
Constraint4
0.102
0.130
0.787
Constraint5
0.114
0.132
0.866
Figure 5 shows the efficient frontiers, inefficient frontiers and minimal variance frontiers of these
critical portfolios, as well as the capital allocation line. Note that the capital allocation line is
constructed using the latest annualized notional risk-free return.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
189
Figure 5: Frontiers with index model.
Photo credit: Original
3.3. Drawdown Model
The paper adopts the same linear programming approach as in professor’s paper Portfolio
optimization with drawdown constraints [12] to generate optimal portfolios including ten stocks and
S&P 500. Note that in consistence with professor’s original paper, the author uses daily cumulative
uncompounded returns. Moreover, the technological constraints are xmin =1/22 and xmax =1/11. The
model implicitly recommends investing half of the available money if all positions are equal to the
lesser bound. The model implicitly recommends that the leverage should be two if all of the positions
are equal to the higher bound. The Python-based gurobipy software handles the linear programming
issues.
Table 11: List of markets for AvDD problem.
Reward
0.099
0.159
0.211
0.260
0.304
0.336
0.361
0.382
0.396
MaxDD
0.267
0.375
0.478
0.588
0.713
0.797
0.869
0.986
1.061
AvDD
0.023
0.032
0.041
0.050
0.059
0.068
0.077
0.086
0.092
MaxDDRatio
0.372
0.424
0.441
0.443
0.426
0.422
0.416
0.388
0.373
AvdDDRatio
4.318
4.980
5.415
5.204
5.150
4.945
4.688
4.445
4.308
Table 11 and Table 12 present the list of markets and corresponding sets of optimal weights for
AvDD problem.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
190
Table 12: Optimal weights for AvDD problem.
AMZN
0.045
0.108
0.147
0.182
0.182
0.182
0.182
0.182
0.182
AAPL
0.045
0.091
0.131
0.170
0.182
0.182
0.182
0.182
0.182
CTXS
0.045
0.045
0.045
0.045
0.045
0.052
0.166
0.182
0.182
JPM
0.045
0.045
0.045
0.045
0.068
0.182
0.182
0.182
0.182
BRK/A
0.045
0.045
0.057
0.110
0.182
0.182
0.182
0.182
0.182
PGR
0.045
0.148
0.182
0.182
0.182
0.182
0.182
0.182
0.182
UPS
0.045
0.045
0.045
0.045
0.045
0.156
0.182
0.182
0.182
FDX
0.045
0.045
0.045
0.110
0.045
0.045
0.045
0.101
0.182
JBHT
0.045
0.056
0.119
0.172
0.182
0.182
0.182
0.182
0.182
LSTR
0.045
0.045
0.045
0.060
0.182
0.182
0.182
0.182
0.182
SPX
0.045
0.045
0.045
0.045
0.045
0.045
0.071
0.174
0.182
Table 13 and Table 14 present the list of markets and corresponding sets of optimal weights for
MaxDD problem.
Table 13: List of markets for MaxDD problem.
Reward
0.11
1
0.15
7
0.19
7
0.23
5
0.27
1
0.30
3
0.33
3
0.35
5
0.37
0
0.38
4
0.39
6
MaxDD
0.27
0
0.35
0
0.43
0
0.51
0
0.59
0
0.67
0
0.75
0
0.83
0
0.91
0
0.99
0
1.06
1
AvDD
0.02
5
0.03
4
0.04
0
0.04
7
0.05
5
0.06
4
0.07
1
0.07
5
0.08
1
0.08
7
0.09
2
MaxDDRati
o
0.41
0
0.44
9
0.45
8
0.46
1
0.45
9
0.45
3
0.44
3
0.42
7
0.40
7
0.38
8
0.37
3
AvDDRatio
4.48
7
4.66
8
4.88
4
4.97
0
4.88
9
4.71
8
4.68
0
4.71
1
4.57
5
4.41
2
4.30
8
Table 14: Optimal weights for MaxDD problem.
AMZN
0.077
0.095
0.132
0.182
0.182
0.182
0.182
0.182
0.182
0.182
0.182
AAPL
0.045
0.073
0.135
0.182
0.182
0.182
0.182
0.182
0.182
0.182
0.182
CTXS
0.045
0.045
0.045
0.053
0.082
0.182
0.182
0.182
0.182
0.182
0.182
JPM
0.045
0.045
0.045
0.059
0.182
0.182
0.182
0.182
0.182
0.182
0.182
BRK/A
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.165
0.182
0.182
0.182
PGR
0.045
0.045
0.045
0.045
0.104
0.104
0.182
0.182
0.182
0.182
0.182
UPS
0.045
0.045
0.045
0.045
0.045
0.045
0.151
0.141
0.182
0.182
0.182
FDX
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.105
0.182
0.182
JBHT
0.047
0.165
0.182
0.182
0.182
0.182
0.182
0.182
0.182
0.182
0.182
LSTR
0.045
0.045
0.045
0.045
0.045
0.045
0.136
0.182
0.182
0.182
0.182
SPX
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.067
0.182
Table 15 and Table 16 present the list of markets and corresponding sets of optimal weights for
CDaR problem with (1 − α) = 0.05. The solutions having the highest Reward/AvDD ratio and
Reward/MaxDD ratio are boldfaced in the tables. In order for the optimization issue to still have
solutions, the minimal risk value is selected. The biggest risk value is only the smallest risk value that
may bring all positions to the upper limit of technological limitations.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
191
Table 15: List of markets for CDaR problem with (1 − α) = 0.05.
Reward
0.106
0.151
0.190
0.225
0.258
0.288
0.318
0.343
0.364
0.383
0.396
MaxDD
0.268
0.352
0.436
0.511
0.583
0.658
0.739
0.822
0.917
0.981
1.061
AvDD
0.024
0.033
0.039
0.046
0.052
0.058
0.063
0.070
0.079
0.087
0.092
MaxDD
Ratio
0.395
0.429
0.436
0.441
0.442
0.438
0.431
0.418
0.397
0.390
0.373
AvDD
Ratio
4.426
4.605
4.844
4.917
4.944
5.006
5.046
4.878
4.618
4.425
4.308
Table 16: Optimal weights for CDaR problem with (1 − α) = 0.05.
AMZN
0.064
0.181
0.182
0.182
0.182
0.182
0.182
0.182
0.182
0.182
0.182
AAPL
0.045
0.045
0.045
0.045
0.076
0.125
0.177
0.182
0.182
0.182
0.182
CTXS
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.070
0.104
0.182
0.182
JPM
0.045
0.060
0.122
0.153
0.182
0.182
0.182
0.182
0.182
0.182
0.182
BRK/A
0.045
0.045
0.045
0.097
0.091
0.094
0.162
0.182
0.182
0.182
0.182
PGR
0.045
0.045
0.161
0.182
0.182
0.182
0.182
0.182
0.182
0.182
0.182
UPS
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.182
0.182
0.182
0.182
FDX
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.051
0.147
0.182
0.182
JBHT
0.045
0.045
0.075
0.158
0.182
0.182
0.182
0.182
0.182
0.182
0.182
LSTR
0.045
0.045
0.045
0.045
0.102
0.164
0.182
0.182
0.182
0.182
0.182
SPX
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.053
0.182
The efficient frontiers for the Reward-AvDD and Reward-MaxDD problems are shown in Figure
6 and Figure 7, respectively.
Figure 6: Efficient frontier for Rewad-AvDD
problem.
Figure 7: Efficient frontier for Rewad-MaxDD
problem.
Photo credit: Original
Photo credit: Original
Figure 8 shows the Reward-AvDD graphs for the portfolios optimal with (1 − α) = 0, 0.05, 0.4 and
1 CDaR constraints. As we thought, the scenario where (1 ) = 1 CDaR corresponds to AvDD has a
concave efficient frontier that dominates other graphs, as one might anticipate.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
192
Figure 8: Reward-AvDD graphs.
Figure 9: Reward-MaxDD graphs.
Photo credit: Original
Photo credit: Original
Reward-AvDD graphs for the portfolios with (1 ) = 0, 0.05, 0.4, and 1 CDaR limitations are shown
in Figure 9. The scenario where (1 ) = 0 CDaR corresponds to MaxDD has a concave efficient frontier
that dominates other graphs, as is predicted.
The charts of MaxDD Ratio and AvDD Ratio are shown in Figure 10 and Figure 11, respectively.
Figure 10: AvDDRatio.
Figure 11: MaxDDRatio.
Photo credit: Original
Photo credit: Original
4. Conclusion
Markowitz’s asset selection model has undergone multiple empirical tests since its inception in the
1950s and has continued to develop to this day.
This thesis uses nearly 20 years of historical daily total return data for 10 stocks, aggregate the
daily data into monthly observations, and based on these monthly observations, calculate all
appropriate optimized inputs for the full Markowitz model (MM) and the index model (IM). This
paper finds areas of permissible portfolios with additional constraints (efficient boundary, minimum
risk portfolio, optimal portfolio, and minimum return portfolios boundary). Markowitz model fully
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
193
utilizes the covariance matrix to generate superior portfolios. However, its results are numerically
unstable. Meanwhile, its assumptions of normality, stationarity and homoskedasticity are invalidated.
Index model simplify Markowitz model to generate more robust results. However, it introduces
additional assumptions of independence, normality, and homoskedasticity of regression residuals,
which, are invalidated as well. Drawdown (CDaR) model is assumption-free. Moreover, it reduces
the nonlinear optimization problems into linear programming problems, which can generate
numerically stable results.
Markowitz’s theory has gradually transitioned financial research towards quantification, which is
a significant progress in theory. However, from the empirical results of this article, it can be seen that
in reality, financial market data is difficult to meet a series of assumptions of the model, which leads
to the problem of low simulation efficiency. Therefore, future research can start in this area.
References
[1] Markowitz, H. M. (1991). Foundations of portfolio theory. The journal of finance, 46(2), 469-477.
[2] Petters, A. O., Dong, X., Petters, A. O., & Dong, X. (2016). Markowitz Portfolio Theory. An Introduction to
Mathematical Finance with Applications: Understanding and Building Financial Intuition, 83-150.
[3] Vietrova, H. V., & Huzhva, V. O. (2018). Analysis of the Markowitz’s and Tobin’s models for securities portfolio
construction. Вісник Національного технічного університету «ХПІ». Серія: Системний аналiз, управління
та iнформацiйнi технологiї, (44), 36-41.
[4] Mary, J. F., & Rathika, G. (2015). The single index model and the construction of optimal portfolio with cnxpharma
scrip. International Journal of Management, 6(1), 87-96.
[5] Dimitriu, M., Dinu, M. R., & Caracota, R. C. (2014). Modelling the efficient frontier of investments portfolio.
Knowledge Horizons. Economics, 6(3), 35.
[6] Marling, H., & Emanuelsson, S. (2012). The Markowitz portfolio theory. November, 25, 2012.
[7] He, Y., Li, D., Wu, Y., Lu, Z., & Ju, L. (2022, January). A Study on the Determinants of Stock Returns, in Comparison
of the Fama-French Models. In Proceedings of the 2022 13th International Conference on E-Education, E-Business,
E-Management, and E-Learning (pp. 493-499).
[8] Asmussen, S. (2022). On the role of skewness and kurtosis in tempered stable (CGMY) Lévy models in finance.
Finance and Stochastics, 26(3), 383-416.
[9] Drezner, Z., Turel, O., & Zerom, D. (2010). A modified Kolmogorov–Smirnov test for normality. Communications
in Statistics—Simulation and Computation®, 39(4), 693-704.
[10] Stephens, M. A. (1970). Use of the Kolmogorov–Smirnov, Cramer–Von Mises and related statistics without
extensive tables. Journal of the Royal Statistical Society Series B: Statistical Methodology, 32(1), 115-122.
[11] Kim, N. (2016). A robustified Jarque–Bera test for multivariate normality. Economics Letters, 140, 48-52.
[12] Chekhlov, A., Uryasev, S., & Zabarankin, M. (2004). Portfolio optimization with drawdown constraints. In Supply
chain and finance (pp. 209-228).
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231133
194
Research on the Brand Marketing Strategy of Netflix in the
New Media Environment
Huazhen Xiao1,a,*
1Film Studies, The University of St Andrews, St Andrews, KY169AJ, United Kingdom
a. hx30@st-andrews.ac.uk
*corresponding author
Abstract: Netflix has been a transformational force in the quickly developing new media,
altering how entertainment is delivered and consumed. This essay examines Netflix’s brand
positioning and social media marketing strategies within the new media environment. The
brand identity of Netflix, which is distinguished by its innovation, accessibility, and diversity,
is what makes it successful. Netflix has established itself as a leading streaming service
platform by providing binge-watching opportunities, ease, and a wide variety of material. In
addition, its social media strategy promotes participation, community, and personalized
recommendations, boosting the perception of its brand. However, difficulties continue. It can
be challenging to balance user engagement and advertisement, deal openly with critical
comments, and keep users interested when there are gaps in content distribution. Transparent
data practices must be followed in the face of data privacy concerns. This essay offers
suggestions for practical action. Creating original content and adjusting to algorithm changes
are key to maintaining interest. Clear disclosure and user consent for data use are necessary
to build trust. Emphasizing originality and regionalized promotion are necessary for
differentiation in a competitive market. This research explains how Netflix’s brand marketing
thrives in the digital media era and provides insights into entertainment companies and
beyond. Despite being predominantly social media-focused, its implications may apply to
general brand marketing strategies in the new media environment. This research serves as a
fundamental investigation of brand marketing strategies in the context of contemporary media
as technology develops rapidly.
Keywords: brand marketing strategy, Netflix, new media environment
1. Introduction
1.1. Research Background
In the ever-evolving new media landscape, Netflix, as a trailblazer, redefined the entertainment
experience. With its rapid growth to prominence, Netflix has become synonymous with streaming,
reshaping how content is delivered and consumed in the new media environment. At the same time,
as technology continues to develop across all areas, the role of brand marketing strategies has gained
even more significance. This essay delves into Netflix’s brand marketing strategy within this dynamic
environment, examining how the company has built its brand identity, content provision, personalized
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231087
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
195
user experiences, and innovative management to not only secure its position in the streaming market
but also to define the very essence of brand engagement in the digital era.
Consumer behavior and preferences have significantly changed due to the rapid expansion of
streaming platforms that characterize the new media environment. Traditional media formats are
becoming less attractive as audiences seek personalized, on-demand content consumption. In this
context, Netflix has exploited its brand marketing strategy as a linchpin in capturing audiences’
attention and fostering a deep emotional connection within them. By exploring the various facts of
Netflix’s brand marketing approach, more valuable insights on how the company navigated such a
transformative environment will surface. These insights also extend the understanding of modern
brand strategy beyond entertainment.
The transformation from traditional media to the digital environment has significantly altered how
audiences access and interact with media content. The emergence of streaming services has been the
defining feature of this transformation, ushering in a new media consumption paradigm that
prioritizes accessibility, variety, and interaction. Digital platforms have steadily superseded broadcast
television and physical media as the dominant media type. The ease of having access to content
whenever and wherever people want has accelerated the uptake of streaming services. The culture of
appointment-based viewing has given way to where people consume material according to their
schedules and tastes. Streaming services have developed as the keystone of the new media
environment. Such services have been boosted by the appeal of vast content libraries combined with
flexible subscription plans that cancel ads. The growth in streaming has attracted media
conglomerates and tech giants to establish their platforms, deepening competition and fragmenting
the market. Development in digital media makes content creation and distribution more democratic.
This democratization has led to an unprecedented proliferation of consumer choices, offering
mainstream and niche content catering to specific interests. Correspondingly, the consumer has the
power to characterize personal entertainment experience, while content producers face both
opportunities and difficulties in making their work stand out among the sea of choices. The evolution
of the new media environment pushes Netflix and similar streaming services to the leading stage of
innovation, challenging established media models and promoting a new era of content consumption.
Examining Netflix’s brand marketing approach illustrates the success of streaming services depends
not only on the quality of its content but also on its capacity to manage and monitor the altering
consumer behavior trends and preferences in the new media environment.
1.2. Literature Review
Some papers discuss the utilization of Internet marketing by Netflix. They emphasize how the Internet
has influenced and altered marketing strategies and allowed businesses to overcome physical barriers.
Netflix allows users to access content from anywhere in the world. The company’s content choices
have increased due to the partnership with giant production industries, such as Walt Disney and
Warner Bros. With continued expansion anticipated, the content expansion strategy assisted Netflix
in acquiring a sizable 33 million-strong client base [1]. Netflix also addressed policies to promote
Internet marketing, such as the Video Privacy Protection Act. The policies enable users to share
content on social networks like Facebook. These strategies aid Netflix in reaching a larger audience.
Malte Hagener examines how viewers’ preferences and content are affected by Netflix’s
algorithm-driven content recommendation system [2]. The paper focuses on the interactions between
user preferences, data-driven algorithms, and the movie Bird Box. It explores how algorithms deliver
personalized content recommendations to specific users that affect consumer’s watching choices.
Another topic he explores is the “black box” idea, which refers to algorithms and procedures that are
opaque to users and influence their content selections. The movie “Bird Box” is used as a case study
to show how algorithmic operations and content creation are intertwined. The movies’ production
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231087
196
and distribution may be influenced by algorithmic insights into viewer preferences, leading to
deliberate choices regarding producing new material. The study critically examines the mutually
beneficial relationship between Netflix’s algorithmic system and content offerings. Andrey
Feuerverger, Yu He, and Shashi Khatri investigated the Netflix Prize challenge [3]. The Netflix Prize
is aimed at enhancing the precision of Netflix’s movie recommendation algorithm. The paper focuses
on the statistical significance of the results achieved by various teams. Firstly, it provides essential
background information about the Netflix Prize and then explores the statistical methodologies used
to assess the improvement of Netflix’s recommendation algorithms. The author stresses the
importance of strong statistical methods to ensure that reported improvements are statistically
significant and not merely due to chance. The authors explore the difficulties in establishing if the
gains obtained were statistically significant according to the results given by the teams. They draw
attention to the fact that the traditional methods might not be directly applicable due to the size and
complexity of the task. Thus, they suggest techniques that consider the size of the data set and the
type of improvements, and they discuss “concentration bounds” to more appropriately evaluate the
statistical significance of observed gains.
1.3. Research Framework
After reading the relevant papers of many scholars, the author found that most scholars focus on the
content output of Netflix and personalize viewing decisions, but few of them talk about the impact of
the development of new media on marketing its brand. This paper will take this as the core to explore
how to market Netflix in the new media environment.
2. Case Description
Netflix has created a unique niche in the competitive streaming service market through a clearly
defined brand identity and strategic positioning. The company’s brand identity encapsulates qualities
that set it apart from competitors and appeal to its target market. Its position within the market is
highly linked to its brand identity. Netflix’s brand identity has refined and embodies innovation,
accessibility, and diversity. The iconic red & black logo, straightforward user interface, and consistent
poster styles create a sense of familiarity for users. The visual presentation and consistency across
devices and locations strengthen viewers’ perceptions of Netflix. Beyond aesthetics, “binge-watching”
is important to Netflix’s brand identity. In line with on-demand consumption trends, Netflix promoted
a culture where viewers may consume any content at their own pace -- It releases complete seasons
of original shows at once [4-6]. This distribution strategy challenges the traditional cable and
broadcast model by providing a sizable library of films, TV shows, and original content at a
reasonable subscription fee. This positioning strategy was crucial to Netflix’s rapid ascent to
household recognition as the standard for streaming services.
Netflix’s identity and positioning are greatly influenced by its main brand attributes. Convenience
is at the front line. The ability to watch content anytime, anywhere, and on any device makes Netflix
stick in the streaming market. Netflix has become synonymous with convenience. Its seamless user
experience is associated with the current consumer’s appetite for entertainment on the go. Netflix
offers a wide range of content that appeals to various preferences and interests. It provides all genres
of movies and TV shows across countries, ranging from horror to documentary films. Netflix aims to
position itself as an entertainment center for all demographics. Also, Netflix strategically
differentiated itself from being a content distributor by investing in original content creation.
Originality has become a defining characteristic of Netflix. The company produces cutting-edge,
ground-breaking original content, demonstrating its commitment to providing compelling, engaging
stories to the audience. By creating shows like “Stranger Things,” “Emily in Paris,” and “The Crown,”
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231087
197
Netflix positioned itself not only to attract subscribers but also as a content producer of high-quality,
diverse, culturally applicable content.
Netflix’s unique and innovative branding and positioning make the company welcomed worldwide.
The revolutionized entertainment consumption model centered on binge-watching, connivance, and
diversity and originally played a crucial role in its success in the new media environment while
keeping the viewer at the center of its strategic goals.
3. Analysis on the Problem
3.1. Netflix’s Current Social Media Marketing Strategy
Netflix’s social media marketing strategy utilizes digital platforms to engage and connect with its
worldwide audience. The company’s social media strategy is consistent with its brand identification
and positioning. It successfully tapped into the power of social media to keep its reputation as a
leading streaming industry through innovative activities, authentic interactions, and strategic content
sharing. It has a robust social media presence distinguished by interesting and participatory
promotions. The company regularly runs promotions and activities in conjunction with the
distribution of original content. These campaigns often include creative competitions, pop quizzes,
and hashtag campaigns inviting viewers to participate. By having viewers in these activities, viewers
become active participants in the marketing process, fostering a sense of community and excitement.
Its social media posts do not exhibit a consistent or formal tone. It has a distinctive, wittier, funnier,
and more relatable personality, as seen in social media posts. The company is not afraid to make jokes,
memes, and pop culture stories [7,8]. Such a light tone and engaging style connect with the audience,
mainly Generation Z. This delightful social media marketing makes the company seem approachable
and welcoming.
Besides, Netflix uses social media to distribute teasers, trailers, and behind-the-scenes material for
new-release content. This tactic attracts fans and creates excitement and buzz, increasing viewership
before and during screening. The planned release steps and timing on social media contribute to a
sense of immediacy and encourage viewers to explore more. In the way of further exploration, as
mentioned in other scholarships, Netflix uses its data-driven insights to offer viewers personalized
recommendations on their APP and social media. The company customizes its content
recommendations depending on consumers’ viewing histories and interests, whether through direct
messages or focused advertisements. The personalization of viewing improves the user interface
experience. Adding personalized recommendations, Netflix maintains a unified visual style across all
its platforms. Its recognizable branding components, such as the red-black color scheme and N logo,
ensure the company can be recognized on various platforms. The consistency strengthens its
positioning and aids in brand identification, creating a sense of Netflix community.
Netflix’s social media marketing approach aims to establish a community through a consistent
digital experience with its brand identity, going beyond purely promotion. The company efficiently
uses social media as a dynamic tool to increase its brand presence and preserve its value in the
streaming industry by providing engaging entertainment, interactive communication, and consistent
visual presentation.
3.2. Problems in Netflix’s Social Media Marketing Strategy
Although Netflix has excelled in many areas of social media marketing, challenges and difficulties
still exist. The difficulties are due to social media’s dynamic nature and the changing audience’s
preferences and expectations. Knowing and addressing these problems is essential for the company
to use social media to advance with the times and keep viewers engaged with the brand.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231087
198
There is a fine line between genuine participation and overselling, even though user involvement
and interactive advertising are crucial. It takes time and skill to find a balance between the need to
promote information and the need to provide innovative and worthwhile interactions. An excessive
focus on advertising content risked alienating fans looking for more sincere relationships. Also,
algorithm updates on social media may cause material visibility in user feeds [9,10]. To ensure its
content reaches its target audience, Netflix must constantly alter its strategy to these algorithm
modifications. Falling to do so can cause viewers to lose interest in the recommended content,
lowering user engagement.
Due to the transparency of social media, Netflix is vulnerable to acclaim and criticism. Negative
comments about the material’s content quality, diversity, or moral issues can spread rapidly and harm
the brand’s reputation. Maintaining a positive and trustworthy brand image requires properly and
honestly addressing those negative feedbacks. As a streaming service, Netflix faces privacy issues
since it uses viewers’ data to personalize suggestions and engage viewers [11]. Users are becoming
more cautious about data collection, personal information, and exploitation. A careful and trustful
balancing act must be performed in order to guarantee transparent data practices when providing a
customized viewing and engaging experience.
Nonetheless, Netflix frequently creates a lot of anticipation around new releases; retaining
engagement between substantial content dumps can be difficult. To reduce audience attrition and
foster the sense of community that social media generates, it is imperative to keep people interested
and continuously engage with the brand during content lulls. Also, since Netflix’s users are from all
over the world, the company must employ a variety of social media channels, each with its user base
and capabilities that correspond to local culture. Maintaining a consistent but customized social media
marketing strategy across various platforms can be highly complicated and time-consuming. Other
similar brands’ social media are also fighting for users’ attention in the fiercely competitive streaming
industry. In this case, users have more options, so they may develop a preference for certain brands
over others due to this rivalry and become more engaged with the content there. In order to keep
brand loyalty, Netflix must consistently and frequently differentiate its value from other competing
brands and demonstrate its uniqueness.
4. Suggestion
4.1. Sustaining Interest and Engagement
The solution to keeping interest in social media is to make sure that the connection between viewers
and platforms goes beyond simple promotion. Netflix should offer live content that benefits users in
order to prevent overselling. The content could be thought-provoking, educational, and entertaining.
It can subtly include compelling narratives about the audience in advertising messages and
storytelling. Designed and intentional advertising improves the user experience by entertainingly
giving information rather than interfering with it.
Encourage consumers to produce content about their Netflix experiences as a second strategy to
get rid of the sense of advertising. Since user-generated materials receive actual feedback from
platform viewers, they appear more dependable and trustworthy. This approach undermines a sense
of belonging. Users are signing up for the Netflix community to interact, share, and talk with people
with similar interests or discuss with people who have different ideas. The company can also reveal
behind-the-scenes details about marketing messages. By including production processes, such as
content creation, users will feel more connected to the content.
Netflix must, however, also constantly keep an eye on algorithm adjustments. Monitoring
algorithm upgrades on different social media platforms is essential if the company wants to spot
changes brought on by those modifications immediately. This is routinely examining and analyzing
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231087
199
metrics and engagement patterns. Keep an eye on it will show the impacts on engagement when the
algorithm changes. Those impacts will be seen through monitoring, testing, and experimentation with
various content formats and publishing timings. Then, marketing plans will be adjusted accordingly.
By employing those tactics, Netflix can maintain customer engagement, adapt to algorithm
changes, and manage the delicate balance between true involvement and marketing approaches. The
key is to maintain a two-way conversation between the company and the audience while providing
pertinent interactions that suit their preferences.
4.2. Building Trust in Transparent Data Practices
Due to the transparency of social media, developing a dependable communication channel is crucial.
If a viewer criticizes any part, respond to them publicly and as soon as possible. Although they
actively interact with the audience on various social media platforms, such as Facebook and Instagram,
through the procedure, the organization should be courteous, show appreciation for the criticism, and
show a willingness to comprehend and fix the problem. The business should be open throughout the
process. The business should be clear and transparent to explain how it handles challenges, enhances
the diversity and quality of its material, or deals with moral concerns in content production.
Additionally, getting express user agreement and clearly explaining data usage restrictions is
essential for building a trustworthy platform. Users should feel in control of their information and be
informed of how their data is being utilized. This allows customers to choose whether they want
suggestions tailored to them based on their data. At the same time, the company should provide users
with reassurance that their private information isn’t being shared or misused. By proving Netflix’s
dedication to ethical data practices this fosters confidence.
By putting these ideas into practice, Netflix will be able to handle criticism and balance privacy
and personalization issues. Building a strong and reliable brand image in the eyes of its audiences
will be facilitated by showing their openness, responsiveness, and respect for user preferences.
4.3. Differentiating in a Competitive Landscape
Since more streaming platforms are fostering, such as Disney+, Netflix needs to find a method to
stand out in the very competitive market for streaming services. Since television shows are still the
most popular form of content, the company should share more engaging original content for each
platform that connects with users. This reduces the chance of viewers to switch to alternative
platforms. Netflix can organize a global social media plan while allowing for regional adjustments.
This preserves a consistent brand message while taking into account regional cultural quirks.
Although Netflix has multiple accounts categorized by region, the viewing data and engagement are
varied. The company should share stuff that is relevant to regional events, occasions, or fashions.
Moreover, Netflix can start engrossing social media campaigns to emphasize Netflix’s selling
point and invite the audience to share their experience by incorporating interactive components. This
also exemplifies the sense of community and user loyalty to the brand. The brand’s reputation and
position will be strengthened by sharing successes, milestones, and engaging stories highlighting
Netflix’s leadership in the streaming market.
By continually providing worthwhile and compelling original content, adjusting techniques to fit
different cultures, and highlighting its distinctive value, Netflix can sustain strong brand loyalty and
a committed worldwide audience.
5. Conclusion
Netflix has become a forerunner in the rapidly changing world of new media, reinventing the
entertainment experience and how material is distributed and consumed. This essay examined
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231087
200
Netflix’s brand marketing strategy in the context of this evolving environment. The success of Netflix
is based on its brand positioning and brand identity. By fusing convenience, diversity, uniqueness,
and innovation, the company has carved out a market that appeals to viewers across the world. Netflix
has established itself as a market leader in the streaming sector by providing binge-watching
opportunities, simple accessibility, and a variety of material.
A detailed review of Netflix’s social media marketing strategy shows the company has tapped into
the potential of social networks to connect and engage with a large audience. Netflix has cultivated a
sense of community and enthusiasm among its customers by upholding a consistent visual identity,
carrying out interactive marketing, and providing individualized recommendations. However,
problems still exist with Netflix’s social media marketing. The challenges facing the organization are
striking a balance between user engagement and promotional content, handling negative criticism,
and keeping audiences interested during content lulls. Also, Netflix must deal with data privacy issues
while delivering personalized experience.
Netflix can take a multifaceted strategy to overcome these obstacles. Storytelling strategies and
engaging content are necessary to maintain engagement while adjusting to algorithm changes through
monitoring. Clear disclosure, user consent, and the use of anonymized data are key components in
establishing confidence in transparent data practices. At the same time, standing out in a competitive
market needs customized content, regionalized promotions, and success story sharing.
This research makes a substantial contribution to the understanding of how Netflix’s brand
marketing approach succeeds in the new media environment. It emphasizes the difficulties, such as
striking a balance between user interaction, personalization, and privacy issues. Any company
looking to negotiate the intricacies of the digital era can benefit greatly from the insights gained from
this research. This essay analyses Netflix’s brand marketing approach in great detail, yet it is not
exhaustive. The research ignores other potential facets of the company’s strategy, such as integrated
marketing communication, in favor of concentrating primarily on its social media marketing.
Moreover, the results may not incorporate new developments because they are based on information
that was only available up to the knowledge cutoff date.
The long-term consequence of Netflix’s brand marketing strategy on viewer loyalty and
engagement could be the subject of future study. A thorough examination of how the ideas suggested
in this research were put into practice might offer insightful discoveries about their efficacy and
viability. In addition, analyzing how new technologies like augmented reality, virtual reality, and
other advanced innovations have affected Netflix’s marketing approach may provide a thoughtful
understanding of the company’s future initiatives.
References
[1] Bharadwaj, A., El Sawy, O. A., Pavlou, P. A., & Venkatraman, N. (2013). Digital Business Strategy: Toward a Next
Generation of Insights. MIS Quarterly, 37(2), 471–482. http://www.jstor.org/stable/43825919
[2] Feuerverger, A., He, Y., & Khatri, S. (2012). Statistical Significance of the Netflix Challenge. Statistical Science,
27(2), 202–231. http://www.jstor.org/stable/41714795
[3] Hagener, M. (2020). Thinking Inside and Outside of the (Black) Box: Bird Box and Netflix’s Algorithmic Operations.
In D. Chateau & J. Moure (Eds.), Post-cinema: Cinema in the Post-art Era (pp. 175–192). Amsterdam University
Press. https://doi.org/10.2307/j.ctv1b0fvtp.14
[4] https://www.facebook.com/VoyMediaAgency. (2023, June 21). Netflix Marketing Strategy: Disrupting the
Entertainment Industry with Innovation and Personalization. Voymedia.com. https://voymedia.com/netflix-
marketing-strategy/
[5] Martínez-Sánchez, M. E., Nicolas-Sans, R., & Díaz, J. B. (2021). Analysis of the social media strategy of audio-
visual OTTs in Spain: The case study of Netflix, HBO and Amazon Prime during the implementation of Disney +.
Technological Forecasting and Social Change, Volume 173. https://doi.org/10.1016/j.techfore.2021.121178
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231087
201
[6] PASTEL, R. (2019). Hashtag Television: On-Screen Branding, Second-Screen Viewing, and Emerging Modes of
Television Audience Interaction. In A. De Kosnik & K. P. Feldman (Eds.), #identity: Hashtagging Race, Gender,
Sexuality, and Nation (pp. 165–180). University of Michigan Press. http://www.jstor.org/stable/j.ctvndv9md.14
[7] Pérez, E. P. (2020). Next Episode, Next Market: Social Media Marketing for SVOD: The Case of Netflix. On
Research, Vol. 4., p8-13.
[8] Peter, T. (2022, April 26). What Netflix’s Struggles Can Teach You About Your Digital Strategy (Thinks Out Loud
Episode 347). Tim Peter & Associates. https://www.timpeter.com/blog/netflix-struggles-digital-strategy/
[9] Simpson, L. (2022, February 25). Netflix: the Social Media Master. Digital Society. https://medium.com/digital-
society/netflix-the-social-media-master-afe8c4a52785
[10] Sinanan, J. (2017). The social media landscape: new media and ‘old’ media. In Social Media in Trinidad: Values
and Visibility (Vol. 11, pp. 30–56). UCL Press. https://doi.org/10.2307/j.ctt1xhr53j.6
[11] The opportunities and challenges faced by Netflix in a digital world. (2023, March 10). Digital Society.
https://medium.com/digital-society/the-opportunities-and-challenges-faced-by-netflix-in-a-digital-world-
cd372d3af98c
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231087
202
Imbalance Between Supply and Demand in China’s Labor
Market Facing by the Graduates
Taoyun Lian1,a,*
1Faculty of Business and Management, Beijing Normal University-Hong Kong Baptist University
United International College, Zhuhai, Guangdong, China
a. q030024172@mail.uic.edu.cn
*corresponding author
Abstract: With the spread of higher education, the number of university graduates in China
continues to increase. Nonetheless, the absence of social interest has truly impacted the work
possibilities of college graduates. Moreover, the world economy has still not fully recovered
from the impact of the epidemic. China’s unemployment rate remains at a high level, and the
hole among metropolitan and rustic regions in China is clearly as well. This paper analyzes
the fact of the low employment rate among Chinese college graduates, focusing on the
argument that it is not solely due to individual factors such as psychlogical reasons or city
preferences for future employment. Behind this, there exists an imbalance in the supply and
demand relationship within the current labor market in China. Finally, the paper puts forward
suggestions to ensure China’s labour supply from the aspects of career planning of college
students, corresponding government policy reform, industrial structure adjustment, and active
assistance of enterprises.
Keywords: labor market, college graduates, employment rate, government policy
1. Introduction
The strain on employment is currently rising as China’s economy transitions from a high growth rate
to a medium-high growth rate. China's population of working age made up more than 66.7% of the
overall population from the 1980s until the end of 2014. China has been dealing with a labor supply
crisis for some years, and this situation is only going to become worse.
One typical example that show cases the employment rate situation in China is the phenomenon
of “graduates’ employment pressure”. With a large number of students graduating from universities
and colleges each year, the job market becomes highly competitive, leading to increased pressure on
graduates to secure employment.
For instance, according to the information released by China’s Ministry of Education, in 2020, the
quantity of Chinese alumni will arrive at 8.74 million, a 20-year high. What is more, the unexpected
effect of the new Covid scourge has prompted a convoluted circumstance for the work and business
venture of school graduates [1].
This example highlights the mismatch between the skills and qualifications of graduates and the
demands of the job market. Many graduates face difficulties in finding employment that aligns with
their educational background, leading to underemployment or unemployment. This situation not only
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231145
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
203
poses challenges for individual graduates, but also raises concerns about the effective utilization of
human resources and the overall productivity of the economy.
The case of graduates’ employment pressure in China exemplifies the complexities and challenges
associated with the country's employment rate. It emphasizes the need for comprehensive strategies
and policies to bridge the gap between education and employment, promote entrepreneurship, and
foster a dynamic and inclusive job market that can accommodate the aspirations and talents of the
younger generation.
The purpose of this paper is to explore in depth the causes, effects and solutions to the phenomenon
of labor market imperfections through the phenomenon of student employment. Through a
comprehensive analysis of relevant literature and data, it provides useful suggestions for policy
makers and relevant stakeholders to promote a balanced and sustainable development of the labor
market.
2. Background Description of Employment in China
Currently, with China’s financial development dialing back, there is expanding strain on employment.
In 2022, China experienced a decline in its employment rate, dropping to approximately 63.63 percent
from the previous year’s 64.06 percent. As the world’s most populous country, China has greatly
benefited from its expansive labor market during its rapid economic development in recent decades.
Despite the overall improvement in working conditions for the Chinese population, the working-age
population has been steadily decreasing in recent years. This decline can be primarily attributed to
the country’s low birth rate [2].
In addition, some statistics also show that the performance of local colleges and universities in
different regions is uneven, with employment rates remaining high in economically developed regions
in the east, such as Shanghai, but colleges and universities in the north-west and north-east need to
step up their efforts. While many universities in Shanghai have employment rates of over 90 per cent,
the initial employment rate of Gansu Nation Teachers College in Northwest China is only 19.64 per
cent [3].
In recent years, as China’s population has continued to increase and its level of education has risen,
the supply of university graduates has outstripped demand. At the same time, along with economic
reforms, the transfer of rural labor to cities has led to excessive pressure on the urban labor market.
Coupled with the fact that the youth labor force has a high unemployment rate in China, university
students, as the new youth labor force in the labor market, are usually not 100 per cent employable.
As shown in Figure 1, the number of university graduates has been increasing year by year, but the
data show that the employment rate is not high.
Figure 1: Number of university graduates, 2014 to 2023 (Source: MyCos Research Institute).
0
200
400
600
800
1000
1200
1400
2014 2015 2016 2017 2018 2019 2020 2021 2022 2023
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231145
204
3. Problems and Challenges in the Labor market
3.1. Employment Pressure
The biggest psychological burden facing Chinese undergrads now is work pressure. Undergraduates’
emotional well-being in general, as well as their physical and mental health, may suffer as a result of
excessive pressure. In fact, prolonged and overwhelming economic stress can cause a variety of social
problems, such as nightmares, sleep deprivation, and, in extreme situations, self-destruction, in
addition to major mental and mental health problems like unhappiness, madness, and neurasthenia.
The labor force saw a 4.3% reduction in working hours in 2021 compared to pre-pandemic levels,
resulting in a shortage of 125 million regular posts [4]. This decline directly translates to a decline of
over 700,000 new job openings in urban areas, which significantly affects pupils' chances of finding
jobs. According to the stress cooperation hypothesis, pressure develops from the exceptional
communication between people and their current situation, leading to weariness or exceeding a
person's mental resources and posing a threat to their prosperity. Business stress among
undergraduates results from the association of inside and outer circumstances as well as specific
workplace factors [5]. The coronavirus pandemic functions as a possible stressor that may have a
substantial effect on students because it is a huge cultural event. Those who are especially sensitive
to changes in the gig economy may feel more desperate and under more strain at work. From an
external perspective, the public authority has always been concerned in undergraduates' work because
it directly affects their careers.
3.2. Inadequate Skills Matching
The success of job searchers is greatly influenced by their skills. The possibility that more people,
particularly young people and those with minimal skills, may get detached from the labor market has
increased as a result of the global economic crisis [6].
At the individual level, wages can be severely affected, especially by penalizing overqualification,
which ultimately affects job and life satisfaction. In China, for example, being overqualified should
not be a problem because of the favorable learning atmosphere to which they are subjected. However,
recently graduated university students who have received good training are still unable to find jobs
commensurate with their skill level, meaning that they are not utilizing their full productive potential.
In addition, inadequate skills can reduce the chances of finding a job and thus facing unemployment.
For companies, skill mismatches can have a negative impact on efficiency and core
competitiveness. In addition, skills mismatches can lead to higher employee turnover and
unsatisfactory work organization. Ultimately, skills mismatches can lead to a loss of profits and
markets. Skills mismatches also have detrimental effects on countries and regions, including higher
unemployment and reduced competitiveness, which makes them less attractive to investors. This, in
turn, hampers their ability to transform production and create new jobs. Both public and private
resources are used to train individuals in the expectation that the qualifications they acquire will lead
to favorable outcomes in terms of employment and wages. However, when there is a skills mismatch,
these expectations are often not met, resulting in lower than expected returns on investment.
3.3. Structural Adjustment of Employment
The rapid and extensive development of artificial intelligence in the near future is causing countries
around the world to face the uncertainty of displacement in various industries, leading to significant
challenges in the employment landscape.
Recent undergraduate graduates, especially those with a liberal arts education, are experiencing
increasingly precarious employment prospects due to the unfavorable state of the labor market.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231145
205
Traditional industries are witnessing job losses, resulting in an oversupply of positions that offer
flexible working hours, such as courier and delivery services. Conversely, emerging sectors require
a limited pool of highly skilled individuals. As a result of industrial restructuring, there are fewer job
opportunities available, putting more pressure on college students to find employment.
Furthermore, the growth of the digital economy is leading to the widespread replacement of labor
and other production factors with digital goods. This, in turn, will lead to the displacement and
elimination of traditional jobs, particularly those that involve repetitive and procedural tasks. Digital
technologies are already replacing jobs in certain areas, and this trend is expected to expand and
deepen in the coming years. According to the World Economic Forum’s Future Jobs 2020 report, by
2025, automation and the reorganization of labor between humans and machines will result in the
loss of 85 million jobs across 15 industries [7]. Job titles such as data entry clerks, accountants,
bookkeepers, payroll clerks, and administrative assistants are most likely to be replaced by AI and
machine automation. The financial services industry is expected to see a 21% reduction in these roles,
followed by a 9% decline in the automotive industry and a 20% decrease in the mining and metals
industry.
3.4. City Preference on Migration of College Graduates
With independent work looking for turning into the standard method of business, the decision of the
work environment for Chinese school graduates has gone through tremendous changes.
In 2005, around 90 percent of college understudies chose to work where they considered; in 2009,
over 90% of university graduates in the eastern and waterfront areas decided to locally work, while
this rate dropped to 60.2 percent in 2009. The extent of college graduates in the focal and western
areas was just 55.8 percent and 62.0 percent, respectively [8]. The decision-making process of college
graduates regarding their preferred work environment is often driven by financial considerations,
particularly the assumption of higher salaries in top-tier cities with rapid economic development.
Additionally, the level of support from family background also influences their choice of work
environment, with a preference for their hometown or cities of origin. As the level of family support
decreases, the likelihood of college graduates opting for less flexible work environments, such as
lower-ranking cities, increases. This suggests that when making decisions about their initial
employment, college graduates take into account the opportunity to maintain connections with their
family and hometown.
4. Possible Solutions
The qualitative gap between the demand for employment among university students and its supply
can be attributed to the lack of coordination and unbalanced development among the labor market
and employment factor. In order to address this issue, this paper proposes several solutions to alleviate
the challenges faced by college students in finding employment.
4.1. Training Program and Career Planing
The content of vocational courses in colleges and universities is closely related to career planning.
Therefore, colleges and universities should effectively link the content of vocational courses, fully
play the role of career guidance and vocational counseling, and do a good job of providing skills
support to improve the employment rate of college students in order to solve the current problem of
a gap between career planning and the construction of vocational courses in colleges and universities.
Teachers in institutions should instruct students on career planning, emphasizing topics like how to
set future goals and how to boost employment rates. They can present students movies and graphic
materials in specific education, using the students’ majors as an example [9]. Hence, college students
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231145
206
can fully understand the development prospects of their own majors in the next few years, and then
they can have a clearer understanding of the goals of their own professional studies. At the same time,
teachers can also popularize the contents of professional courses to students in the form of practical
lectures on career planning outside school. For example, for economics and management majors,
teachers can organize students to go to relevant enterprises for field visits and practical exercises to
understand the market economy and the current work situation, and then provide ideas and guidelines
for students to choose employment fields in the future.
4.2. Government Policies
Employment policy alludes to the government authority’s approaches pointed toward expanding
work interest and advancing business. It is a bundle of measures normally embraced by the
government authority to ease employment pressure.
The first is to vigorously develop the modern service industry and provide more jobs suitable for
college graduates. The Government should support small, medium and micro enterprises in taking on
more college graduates for employment, and give them social insurance subsidies, tax breaks and
other supportive policies in accordance with regulations.
The second is that in view of the imbalance in regional development, it is necessary to make overall
use of the resources of all parties, and give priority to recruiting or taking out a certain number of
college graduates in the event of vacancies in full-time jobs in the community. The Government
should continue to implement basic security service programmes, such as rationalising the scale of
recruitment. For college graduates who go to the central and western regions and grass-roots units for
employment, the Government provides policies such as tuition reimbursement and national student
loans in accordance with regulations.
Finally, the government should support self-employment and flexible employment. When schools
and colleges complete designated preparing for school graduates, the Public authority might give
professional preparation sponsorships and backing school graduates in beginning their own
organizations, as per the guidelines. Around 30% of the scenes put resources into and created by the
Public authority will be made accessible for nothing to school graduates.
4.3. State-Owned Enterprises and Public Instistuition could Absorb College Graduates
The government recognizes the significant role that institutions and state-owned enterprises play in
promoting graduate employment through flexible and innovative job creation. After analyzing
targeted policies, it is evident that provincial governments are taking four main approaches to
encourage graduate employment in areas where the country has a demand. These approaches include
promoting college students’ enlistment in the military, encouraging grassroots-level work, creating
more positions for scientific research assistants, and facilitating graduate employment in state-owned
enterprises. Grassroots employment policies have successfully encouraged more college graduates to
serve the community at the grassroots level, effectively alleviating the employment pressure on
graduates. Additionally, in order to mitigate employment risks, certain local governments have set
specific targets for the growth of college students enlisting in the armed forces, such as aiming for a
10% increase in the number of college students enlisting. Furthermore, state-owned enterprises are a
key avenue for graduates to secure employment. Beijing and Tianjin have taken the lead in carrying
out strategies that help the foundation of logical exploration collaborator positions, consequently
facilitating business pressures and advancing innovative work in colleges and examination
organizations [10].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231145
207
5. Conclusion
This paper assesses and summarizes the findings and discussions in the existing literature. It also
highlights the current phenomenon of graduate employment rates in China and the various
mismatches between supply and demand in the labor market. The conclusion of this paper suggests
that the failure of recruiters to adequately match the skills of job seekers with the required positions
may be one of the most important problems. However, there are other factors as well, such as city
preference and mobility implications.
In addition, the paper also discussed the problems and challenges in the labor market, which can
be divided into four parts. The content includes employment pressure, inadequate skills matching,
Structural adjustment of employment and city preference on migration of college graduate Based on
this, this paper proposes analysis and solutions, which also mentioned three different points.
The limitation of this study is that due to the complexity of the labor market, this paper may need
to narrow down focus and analyze college graduates of the labor market. While this approach allows
for more in-depth analysis, it may limit the generalizability of findings to the broader labor market.
At the same time, the labor market is subject to constant changes, influenced by factors such as
economic cycles, technological advancements, and policy reforms. A study conducted in the last few
years may not capture the dynamic nature of the labor market and its potential fluctuations in the
employment rate over time. Moreover, this paper is more about information integration and
explanation, but there are not too many realistic cases study. Further research could undertake a more
in-depth analysis by categorizing workers by age group and perform analyses that have already been
predetermined and analyses of the reactions of various employers under various job pressures, which
will be the empirical research goal in the future.
References
[1] Li, C. (2022) A Comparative Study on Higher Education Graduates’ Employment Policies between China and the
United States. Journal of Educational of Research and policies. 04(11).16
[2] Textor, C. (2023). Share of employed people in the Chinese poplulation from 2012 to 2022. Retrieved from
https://www.statista.com/statistics/239153/employment-rate-in-china
[3] Du Q. (2022) Employment for fresh Chinese graduates in 2022 stable report. Retrieved from
https://www.globaltimes.cn/page/202212/1282362.shtml
[4] Yang, S., Yang, J., Yue, L., Liu, X., Li, W., Cheng, H. and He, G. (2022). Impact of perception reduction of
employment opportunities on employment pressure of college students under COVID-19 epidemic–joint moderating
effects of employment policy support and job-searching self-efficacy. Personality and Social Psychology. 14(01)
[5] Yang, S., Yue, L. and Gao, R. (2022). Analysis on Employment Pressure and Influencing Factors of College Students.
China Academic Journal Electronic. 14(1): 55-64.
[6] Klosters, D. (2014). Matching Skills and Labor Markets needs buliding social partnerships for better skills and
better jobs. World Economic Forum.
[7] Wang, C. and Jia, P. (2020) New challenges for China's labour market system and suggestions for improvement.
China Development Monitor. 15(1): 80-85.
[8] Zeng, G., Hu, Y., Wu, W., and Mensah, I.K. (2021). Employment Flow of College Graduates in China: City
Preference and Group Difference. Sage Journals. 11(01):23-24
[9] Zhu, M. and Zong, Q.(2022) A Study of Career Planning to Improve the Employment Rate of College Students.
Employment and security.
[10] Wu, M., Hao, X. and Tian, Y.(2022) Employment Management Policies for College Graduates under COVID-19
in China: Diffusion Characteristics and Core Issues. Healthcare. 10(5):955.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231145
208
The Implementation of Modern Portfolio Theory on New
Financial Assets: Evidence from Cryptocurrencies
Shuai Chen1,a,*
1Lancaster University Management school, Bailrigg, Lancaster, United Kingdom
a. s.chen56@lancaster.ac.uk
*corresponding author
Abstract: Modern Portfolio Theory (MPT) has long been a cornerstone in the realm of finance,
aiding investors in navigating the complex terrain of risk and reward associated with diverse
assets. This theory, formulated by Harry Markowitz in the 1950s, has traditionally guided
investment decisions by optimizing the balance between different assets to achieve the
desired level of risk and return. However, with the meteoric rise of cryptocurrencies as a new
asset class, there is an increasing curiosity surrounding the applicability of MPT to this digital
phenomenon. In response to this curiosity, this article undertakes the task of comprehensively
assessing the compatibility of MPT with cryptocurrencies. To accomplish this, the research
aggregates and analyzes the existing body of knowledge, thereby offering insights into the
intersection of modern portfolio theory and the age of cryptocurrencies. A systematic
literature review is conducted, encompassing 21 pertinent studies that explore various facets
of this confluence. The findings of this article underscore an emerging trend in research, one
that showcases the adaptability of MPT to innovative financial instruments like
cryptocurrencies. These studies collectively illuminate the ways in which MPT can be
employed to optimize portfolios that include digital assets, shedding light on strategies that
account for the unique risk-return dynamics inherent in the crypto market. As the
cryptocurrency landscape continues to evolve, it is evident that Modern Portfolio Theory is
not only relevant but also adaptable, providing valuable tools to guide investors through the
exciting yet volatile terrain of digital finance.
Keywords: cryptocurrency, modern portfolio, modern financial market
1. Introduction
Decentralized finance, a crypto-asset-based financial network without the need for a central
intermediary, is one of the prominent developments in the financial innovation that has evolved fast
in recent years. Due to their effectiveness, accessibility, and potential for innovation, inclusivity, and
openness, decentralized financial systems have garnered interest. The traditional financial system has
been disrupted by cryptocurrencies, or virtual currency payment systems, which have developed in
recent years to become a crucial way of international payments and currency exchange. However, it
might be challenging for investors to adopt the best trading and investment strategies to maximize
their gains due to the unpredictability of cryptocurrency price swings. Significant risks associated
with cryptocurrency investments, such as exchange rate risk, operational risk, and security risk, have
raised questions among investors about how to balance risk and return in this new asset class. In order
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231149
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
209
to attain the ideal risk-return balance, diversification is employed to lower total risk, according to
Modern Portfolio Theory (MPT), an investing approach that has been widely adopted in conventional
asset markets [1]. As a result, although some investors would want to utilize the MPT theory to guide
their investment selections, the theory may or may not be appropriate for new financial investment
options. By reviewing the findings of earlier research on the analysis of the prospects for application
in this developing field, as well as on the future research trends of the portfolio theory, the objective
of this essay is to further investigate whether MPT can be successfully applied to financial assets such
as cryptocurrencies. Using portfolio theory, both academics and investors may make better
investment judgements.
The following is how the essay is set up: The methods utilized and the literature analysis are both
presented in Section 2. Future study areas are emphasized in Section 3, which also analyzes the
literature on the use of current portfolio theory with regard to novel financial assets like
cryptocurrencies. The conclusions of this review are reported in Section 4 as a last step.
2. Methods and Samples
2.1. Methodology
This study is based on the guidelines of Eigelshoven for conducting a systematic literature review
(SLR) in the field of modern portfolio theory and cryptocurrencies [2]. The systematic literature
review methodology consists of three phases, which are identification, screening and analysis. By
conducting the review with a representative sample of literature in the fields of cryptocurrencies,
modern portfolios, and finance, the review focuses on identifying research results, theories, and
applications. SLR integrated and systematized prior material, highlighted key concerns from a neutral
standpoint, and organized it using a conceptual method. The review’s findings are applicable to
specialists in the subject, generalists, and practitioners who are curious about cryptocurrencies but
lack in-depth technical expertise.
2.2. Sample
The process to select the candidate papers can be find in Figure 1. A keyword-centered approach was
used to comprehensively cover the existing literature on the use of modern portfolios in novel
financial assets. The following broad search strings were developed:
(Cryptocurrency OR modern portfolio OR block chain OR bitcoin OR financial assets OR assets)
AND (Investment Portfolio Effectiveness OR modern financial market OR Opportunity)
The selection of keywords was based on an initial brief screening of important modern portfolio
and cryptocurrency literature, followed by further refinement of the selection for the identification
and search stages. In addition, multiple synonyms were used to ensure robust searches. In order to
achieve representative coverage of relevant literature, according to Eigelshoven in order to assure the
inclusion of pertinent sources that might not be academically indexed, a hybrid method was employed
for the sources of SLR [2]. This modified the technique such that the literature is covered by both
medium-specific and numerous academic databases across a wide variety of publication media. The
selection of databases was conducted through Andrianto and Diputrap and the final database selection
included three academic databases [3]. Google scholar and jstor databases were included for their
focus on economics and social sciences, and Web of Science was included for its broad coverage of
various research areas.
The second phase consists of screening the identified literature by applying search strings to
selected channels and databases. The search was conducted between 2018 and 2023. A total of 2,753
search results were generated by the initial keyword search using preset search phrases and outlets.
1,091 search results were left after removing irrelevant results like editorials, book reviews, news,
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231149
210
and duplicates. A first review of titles, abstracts, and author- or database-defined keywords was
carried out to see if the clicked results meet the parameters of the research. Based on the inclusive
criteria listed below, 192 results were determined to fulfil the screening criterion.
Figure 1: Flow of the search process.
The following filters were set to First, eligible publications had to focus in-depth on the topic of
modern portfolios and/or cryptocurrencies; second, they had to be written in English. Third,
publications without a complete list of references were excluded. The publication period was from
2018 to 2023. If a publication met all three criteria, it was accepted as the second final result and
added to the dataset for further analysis. A forward and backward search was carried out using Google
Scholar to make sure that the field had been thoroughly screened. The same eligibility procedures
and standards that applied to the keyword search also applied to the selection of extra final results
obtained via this search. During the forward and backward searches, important ideas, citations, and
cross-references frequently appeared, indicating that the review had attained the necessary saturation
[4].
Finally, 13 final results were extracted from the literature basket. In addition, 8 additional final
results were identified during the screening of 624 references and 322 citations. Overall, SLR resulted
in 21 unique final results. The final results were then analyzed from a concept-centric perspective to
extract the use of modern portfolio theory in cryptocurrencies.
3. Discussion and Analysis
This literature review discusses the application of modern portfolio theory to cryptocurrencies. In this
literature series, we identify the significant differences that support cryptocurrency markets from
traditional asset markets, including high price volatility, uncertain trading, and lack of regulation.
These characteristics complicate the application of Modern Portfolio Theory (MPT), but with proper
asset allocation, investors can still achieve a balance of risk and return [3, 5]. A contemporary
portfolio is a collection of securities and other assets that are managed together in order to accomplish
a certain goal. A portfolio can contain any asset, including gold, real estate, stocks, and options.
According to Saksonova and Kuzmina-Merlino, the process of creating and managing a portfolio
entail carrying out the following steps: establishing goals and choosing the best portfolio type,
analyzing the investment’s target, creating the portfolio, choosing and putting into practice portfolio
management strategies, and assessing the portfolio’s effectiveness [6]. Evidence demonstrates that
while creating an investment portfolio, the fundamental guidelines of maximizing profitability and
reducing risk should be followed Yen and Cheng [7]. Therefore, the investing methods and
predetermined objectives should be followed by the portfolio. A growth portfolio, for instance,
comprises of assets that enable the realization of a high growth rate of invested capital and is
characterized by considerable risk [8]. This depends on the investment objectives and the trade-off
between profitability and risk. Since cryptocurrencies are hazardous investments with high potential
for appreciation, the majority of them are probably part of growth portfolios.
Secondly, there is evidence that portfolios should be sufficiently diversified and not depend on one
or two economic factors [8, 9]. The investor must spread his funds among several assets in order to
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231149
211
generate a profit and avoid losing money during the investment process. The basic goal of
diversification is to lessen the danger of a catastrophic loss of money as well as to generate acceptable
and predictable returns. Investors first aim to diversify across several asset classes [10-12]. (e.g.,
stocks, bonds, cryptocurrencies, commodities, financial derivatives, etc.)
MPT emphasizes the importance of diversification for risk reduction and the fact that the
introduction of cryptocurrency introduction of portfolios to increase diversity has also been discussed
in the cryptocurrency literature, and With Bitcoin playing a significant role in return and volatility
spillovers among the main cryptocurrencies, discover evidence of growing dependency across
cryptocurrencies and, as a result, contagion risk [13]. In theory, investors’ ability to diversify their
portfolios may be constrained if they solely invest in cryptocurrencies. The exchange rates of
cryptocurrencies, on the other hand, may be adversely connected when individuals transfer money
between them if there is fierce rivalry among them. Portfolios of cryptocurrencies ought to most likely
be a component of an investor’s overall investment plan. In this case, even if the connection is
substantial, a portfolio of cryptocurrencies would outperform a single coin, at least until one
cryptocurrency achieves enough market domination. Therefore, the primary goal of a portfolio
investment is to gain properties from a variety of investment assets that are not possible to receive
from a single asset. To attain the ideal balance of risk and return, a portfolio has to be built. When the
assets in the portfolio are sufficiently uncorrelated, risk is often minimized. That is, diversification
should result in the portfolio’s overall worth not declining much when the value of its constituent
assets diminishes. Thus, investing in a diversified portfolio might help investors reduce their exposure
to risks associated with particular cryptocurrencies, like as hacking attempts, the failure of significant
exchanges, etc. Overall, in order to select the greatest portfolio, one must optimize it. Investors may
identify the appropriate percentage of assets to deploy to various cryptocurrencies in order to reap the
most rewards by employing portfolio optimization strategies.
4. Conclusion
In conclusion of the discussion and analysis, modern portfolio theory has some applicability in the
cryptocurrency market. The principles of diversification, risk-return trade-off and portfolio
optimization remain applicable to cryptocurrency investments. However, investors must carefully
consider market uncertainties and risks and pay close attention to market dynamics. MPT can be a
useful tool to help investors construct effective portfolios in the cryptocurrency market, but risk
management is still crucial. Cryptocurrency, as an emerging field, brings new challenges and
opportunities for portfolio theory and practice. A variety of difficulties and restrictions arise when
Modern Portfolio Theory (MPT) and cryptocurrency are combined.
Here are some restrictions and ideas for more study: 1) A lack of historical data: To calculate asset
returns, volatility, and correlation, MPT needs previous data. Since cryptocurrencies are still
relatively new, there may not be enough historical data to calculate these values with precision. Future
studies should concentrate on approaches to successfully integrate the limited historical data or
investigate alternate data sources like sentiment analysis and on-chain indicators. 2) Lack of Market
Integration: Compared to traditional financial markets, cryptocurrency markets are more
decentralized and less regulated. The presence of risk-free assets and other MPT presumptions may
be affected by this lack of integration. Taking into consideration variations in market structure,
research might examine how MPT can be used to portfolios that contain both conventional assets and
cryptocurrencies. 3) Extreme price volatility and susceptibility to tail events are two characteristics
that make cryptocurrencies popular. The focus MPT places on volatility as a risk indicator may not
adequately account for the special risk features of cryptocurrency. To better adapt to severe
occurrences, future research may concentrate on adding tail risk factors or creating modified risk
metrics. 4) Behavioral Factors: More so than traditional markets, cryptocurrency prices are affected
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231149
212
by behavioral factors, news emotion, and social media trends. Bitcoin returns, real volatility, surges,
and trade volume may all be predicted by negative sentiment. The integration of behavioral finance
theories into portfolio optimization models to account for market sentiment may be the subject of
future study.
Overall, I may anticipate further study and practice in the future to improve methods and
techniques for investing in cryptocurrencies. Transparency, morality, and strict risk management will
always be essential success components.
References
[1] Letho, L., Chelwa, G., & Alhassan, A. L. (2022). Cryptocurrencies and portfolio diversification in an emerging
market. China Finance Review International, 12(1), 20-50.
[2] Almeida, J., & Gonçalves, T. C. (2022). Portfolio diversification, hedge and safe-haven properties in cryptocurrency
investments and financial economics: A systematic literature review. Journal of Risk and Financial Management,
16(1), 3.
[3] Andrianto, Y., & Diputra, Y. (2017). The effect of cryptocurrency on investment portfolio effectiveness. Journal of
finance and accounting, 5(6), 229-238.
[4] Leedy, P. D. (2019). Practical Research: Planning and Design, (12 ed.). NJ, USA: Pearson Education
[5] Pham, H., Nguyen, B., & Bui, M. (2021). Modern portfolio theory in the age of cryptocurrency. Australian Journal
of Applied Finance (formerly JASSA), (5), 41-48.
[6] Saksonova, S., & Kuzmina-Merlino, I. (2019). Cryptocurrency as an investment instrument in a modern financial
market. Вестник Санкт-Петербургского университета. Экономика, 35(2), 269-282.
[7] Yen, K. C., & Cheng, H. P. (2021). Economic policy uncertainty and cryptocurrency volatility. Finance Research
Letters, 38, 101428.
[8] Mazanec, J. (2021). Portfolio optimalization on digital currency market. Journal of Risk and Financial Management,
14(4), 160.
[9] Charfeddine, L., Benlagha, N., & Maouchi, Y. (2020). Investigating the dynamic relationship between
cryptocurrencies and conventional assets: Implications for financial investors. Economic Modelling, 85, 198-217.
[10] Kumah, S. P., & Odei-Mensah, J. (2021). Are Cryptocurrencies and African stock markets integrated?. The
Quarterly Review of Economics and Finance, 81, 330-341.
[11] Jang, J., & Seong, N. (2023). Deep reinforcement learning for stock portfolio optimization by connecting with
modern portfolio theory. Expert Systems with Applications, 218, 119556.
[12] González, M. D. L. O., Jareño, F., & Skinner, F. S. (2021). Asymmetric interdependencies between large capital
cryptocurrency and Gold returns during the COVID-19 pandemic crisis. International Review of Financial Analysis,
76, 101773.
[13] Platanakis, E., & Urquhart, A. (2020). Should investors include bitcoin in their portfolios? A portfolio theory
approach. The British accounting review, 52(4), 100837.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231149
213
Research on the Problems and Countermeasures Analysis of
Old-age Security for the Elderly in Rural China in the
Context of Population Aging
Le Guan1,a,*
1College of Agriculture Economics and Management, Shanxi Agricultural University, Jinzhong,
Shanxi, 030801, China
a. djones81609@student.napavalley.edu
*corresponding author
Abstract: With the simultaneous process of urbanization and population aging in China, old-
age security for rural senior citizens has once again attracted attention. China’s rural areas are
currently experiencing increased population aging, while social security initiatives for the
aging population are insufficient. Therefore, in the context of population aging, it is
imperative to improve the problem of old-age security for rural senior citizens in China. This
paper mainly uses the literature research method and comparative analysis method, and based
on comparing and analyzing the social pension problems in coastal and mainland rural areas,
it puts forward corresponding countermeasures for China’s actual situation to address the
issue of rural pension security in China. China has transitioned into the phase of population
aging. The research on the rural elderly’s old-age security problems can better identify the
existing problems; the existing problems will be studied and solved to help reduce the burden
of China’s younger generation of old-age pension, which can be stabilized for the elderly in
their old age and is conducive to improving the pioneering ability of the younger generation
and the vitality of the country’s economic development.
Keywords: demographic structure, population ageing, rural China, pension system
1. Introduction
1.1. Background of the Study
Population is one of the major issues among many social problems in the world today. According to
the statistics of the UN, the global population is expected to reach 9-10 billion by the year 2050, and
the population over 60 years old will account for about 20% of the total population [1]. This trend is
mainly due to the continuous development of society, rapid technological advances, and swift
development of medical technology, which have significantly increased the average lifespan of
human beings. While longevity is a positive sign, it also presents several challenges, including
population aging [2]. With the increasing number of senior citizens, the issue of old age security has
emerged as a topic of great concern for governments and international organizations. The complexity
of this issue is not only reflected at the economic level but also involves many areas, such as politics,
culture, and family life. The surge in the elderly population not only poses a huge economic challenge
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231158
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
214
to society but also has far-reaching implications for government policymaking, social resource
allocation, and family structure.
The issue of population aging is particularly significant in a large population country like China.
At present, the old-age security system of China is undergoing improvement and has not yet been
able to fully cover the needs of the elderly. Rural senior citizens are more challenged by old-age
security than their urban counterparts, as they often lack the full range of support and benefits enjoyed
by their urban counterparts. Therefore, it is particularly urgent and necessary to conduct in-depth
studies and research on old-age care for rural senior citizens.
1.2. Purpose and Significance of the Study
The construction of a rural social security system for old-age pensions is an important issue that
affects the life of the country and its people. Currently, China is undergoing rapid economic growth.
The economy and society are developing in a comprehensive and coordinated manner, and the urban
and rural old-age security system is relatively sound, while the rural old-age security system remain
relatively weak. This paper tries to rationalize the relationship between the government, the collective,
and the individual in China’s rural social pension security system by introducing the basic theory of
the pension security system, based on successful experiences at home and abroad, based on which
the social security system in rural areas has been systematically sorted out, to Establish a Pension
Insurance System for Urban and Rural Residents that is more suitable for the actual situation in China.
Rural old-age security is supported by the strong financial power of the state; it is through the
government’s financial income reconfiguration that the rural elderly can maintain their basic
conditions of existence, and even improve the living standards the rural elderly, and then make the
contemporary agricultural economic benefits increase steadily, this is a kind of benign, can give the
farmers incentives and expectations of the new type of social income reconfiguration way. In short,
establishing a rural social security system for the elderly is an important measure for promoting urban-
rural integration and coordinated development, with great practical significance for ensuring the basic
livelihood of the rural population, sustaining the fundamental integrity of rural communities while
gradually closing the gap between them and, in particular, eliminating the problem of poverty among
the rural elderly, and providing them with a certain basic livelihood security.
1.3. Content of Research
This paper includes the following three parts: 1) the status quo of China’s rural old-age security; 2)
the existing problems of rural old-age security in China; and 3) the proposed countermeasures
analysis for problems confronting the rural old-age security in China.
2. Current Situation of Old-age Security in Rural China
2.1. Current Status of Rural Old-age Security
The year 2000 witnessed China’s entry into an aging society. During the Twelfth Five-Year Plan
period, the elderly population experienced a rapid increase. According to the 2020 Statistical Bulletin
on the Development of Social Services, as of the end of 2020, the number of individuals aged 60 and
above reached 255 million, making up 17.8% of the total population. It is projected that by 2050,
individuals in this age group will surge to 478 million, constituting 34.9% of the total population (see
Figure.1) [3]. Population aging seriously affects social stability and economic development. Both
developed and developing countries should pay enough attention to the old-age problem caused by
population aging and build a set of old-age security systems suitable for their national conditions on
this basis.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231158
215
By the end of 2020, the number of people aged 65 and above in China’s resident population will
reach 190.635 million, accounting for 13.5% of the country’s total population (see Figure 2) [3]. In
the period when the birth rate was high, the elderly population almost always relied on the family for
their old age, and at this time, the youth population was large, and they had enough ability, energy,
and time to take care of the elderly in the family, but as China’s population ages and the birth rate
continues to decrease, the ratio of the number of young people in the family to the number of elderly
people has begun to change, and the youth population no longer has enough time, energy, and support
for the majority of the elderly, as it did in the past. The young population no longer has enough time
and energy to support the majority of the elderly as before, resulting in the family model of old age
beginning to shake the trend, resulting in the elderly for elderly service institutions increasing demand
[4]. In addition, due to the increasing number of elderly people who are disabled and living alone, the
demand for specialized, personalized, and comprehensive elderly care services is also increasing, and
the construction of the elderly care service system should also be clear about a regular, rule of law
road of future development. This paper includes the following three parts: 1) the current situation of
China’s rural old-age security; 2) the existing problems of China’s rural old-age security; and 3) the
proposed countermeasures analysis for China’s rural old-age security problems.
Figure 1: The degree of population aging and the birth rate [China Statistical Yearbook (2021)].
Figure 2: Age distribution of the Chinese population[China Statistical Yearbook (2021)].
0.00%
2.00%
4.00%
6.00%
8.00%
10.00%
12.00%
14.00%
16.00%
18.00%
20.00%
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
birth rate Ratio of the population aged 60 and older to the total population
17.95%
63.35%
5.20%
13.50%
0-14 Years old
15-59 Years old
60-64 Years old
Age 65 and over
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231158
216
3. Existing Problems of Old-Age Security in Rural China
3.1. Weak Economic Base of the Rural Elderly
With the rapid development of China’s economy, culture, and society, the demand for factories, office
buildings, and sales venues from manufacturers, residential buildings, recreational venues, and
entertainment venues from people, and libraries, study rooms, and stationery stores from students,
the phenomenon of China’s arable land being occupied has been increasing year by year. The increase
in population makes us have to face the fact that the per capita arable land is decreasing, which also
leads to the contradiction of more people and less land becoming more and more prominent, so that
the farmers who originally made their living by cultivating land have to look for other ways to earn
income to maintain a basic life, and because there is a certain age limit for the position of manual
labor, so that the elderly farmers who are occupied by the land are unable to earn income from manual
labor, and the more stable income they gained from cultivation is no longer available. The more stable
income from farming no longer exists, making it difficult for them to have sufficient funds to pay for
old-age services. The average annual income of older urban persons is 17,892 yuan, with an average
income structure of 15,530 yuan (86.8 percent) for old-age security, 1,223 yuan (6.8 percent) for
market earnings, 382 yuan (2.1 percent) for public transfers, 546 yuan (3.1 percent) for family
transfers, and 210 yuan (1.2 percent) for others. The average annual income of the rural elderly is
4,756 yuan, and the structure of the annual income is as follows: old-age security, 890 yuan (18.7%),
market earnings, 1,520 yuan (32%), public transfers, 784 yuan (16.5%), family transfers, 607 yuan
(12.8%), and others, 955 yuan (20.1%). [5] The composition of the average annual income of the
elderly reflects the current situation in which the urban elderly use guaranteed income as their main
source of income, while the rural elderly use market and other incomes as their main source of income,
and at the same time, public transfers, which are sourced from fiscal payments, have become an
important economic support for the rural elderly. [5] The income gap between urban and rural older
persons is too wide, and in terms of old-age security, rural older persons are far less well off than
their urban counterparts, while rural older persons are far less well off in terms of market earnings
because of their age, social development, and other problems, so that the rural older persons’
economic base is too weak when it comes to old-age issues.
3.2. The Traditional Family Model of Old Age is Unstable
Provinces have responded positively to the policy of family planning by introducing a series of
preferential policies for one-child families, leading to a significant increase in the number of one-
child families, with the result that the number of young people has declined, while the number of
older persons has risen sharply in contrast, resulting in changes in family structure that have had a
serious impact on the traditional way of raising old people in the family. First of all, the aging of
China’s population has been increasing year by year, and the average life expectancy of the elderly
has been lengthened. Along with the increase in age, the income from work has been gradually
reduced; however, the cost of illness has been increasing year by year with the increase in age,
resulting in the elderly not only not being able to continue to save but also consuming the savings
from the past, and in the policy of advocating late marriages, late and preferential births, and excellent
childbearing in China, after their parents have become old, the children are not old enough to support
a family, and they do not need to support a family. Support a family age, do not have the ability to
support their parents, so that the traditional family pension model began to shake [6]; Secondly, the
development of social change accompanied by rapid economic development, young people and
middle-aged people as an essential part of social life, their work, the pressure of life has increased
greatly so that the majority of people can rarely work and family, and because of the implementation
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231158
217
of the policy of family planning caused by the current stage of the implementation of the family
planning policy has resulted in the current stage of the majority of one-child, a child supporting two
elderly people and increase the pressure on young people and middle-aged people, the burden of
family pensions has become increasingly heavy (China’s urban and rural family pensions dependency
ratio has reached 19.70%), from the time, energy, and the economic basis of the traditional model of
family pensions relying on the land is not stable [7]; Finally, the central and western cities in recent
years, due to the geographic location of the city, the development is far behind the coastal cities, the
economic situation is not as good as it used to be. Finally, the development of central and western
cities in recent years due to geographic location and other reasons lagged far behind the coastal cities;
the economic backwardness led to more young people beginning to leave their hometowns to flow to
economically developed areas, the degree of population outflow in central and western China has
increased year by year, the outflow of young people began to settle down in foreign countries, the
establishment of a personal family, which makes it difficult for young people to take care of the
family, parents and the two, resulting in the traditional family model of old age is not solid.
3.3. Inadequate Social Security System for the Elderly
The implementation of China’s new rural social security system has indeed alleviated the rural
pension problem to a certain extent, but the existence and deepening of these problems also require
more in-depth thinking and solutions.
3.3.1. Problem 1: Insufficient Laws and Regulations
China’s rural social security system for the elderly still lacks comprehensive laws and regulations to
govern its operation, which leads to instability and uncertainty in policy implementation. According
to the 2021 data, although China has issued a series of policy documents, there is still no national law
to establish the rights and obligations of rural old-age security. This prevents the rights and interests
of the rural elderly from being adequately safeguarded. It is recommended that national laws and
regulations be formulated to clarify the benchmarks and responsibilities for policy implementation
and to improve the stability and predictability of the rural old-age security system.
3.3.2. Problem 2: Insufficient Incentives
Rural residents’ pensions are usually paid according to the number of elderly people, which may
reduce farmers’ incentives to contribute. According to the 2020 data, average pension levels remain
relatively low nationwide. In some areas, monthly pensions for the elderly only cover basic needs and
are unable to cope with rising healthcare and pension costs. To increase farmers’ motivation,
consideration could be given to determining pension payment standards based on the actual income
and contribution history of individual farmers to better reflect their financial burdens and needs.
3.3.3. Problem 3: Insufficient Subsidies for Collective Farmers
The current pension security policy mainly focuses on urban workers and those who stay in their
hometowns, while the subsidies for collective farm households are relatively insufficient. According
to data from 2022, collective farming households tend to face greater risks in old age because they
have a relatively single source of income, and the inadequacy of the social security system makes it
difficult for them to cope with the rising costs of old age. It is recommended that the government
increase its support for collective farm households’ old-age security, raise the level of their pension
payments, and ensure that their basic livelihoods are adequately protected.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231158
218
3.3.4. Issue 4: Legal Effectiveness and Government Trust
Rural social pension systems are often implemented by local government units but generally lack
sufficient legal force and monitoring mechanisms. This has led to poor and unfair policy
implementation. In addition, the public’s distrust of local government policies has led to an uncertain
attitude towards pension contributions. To address this problem, a more transparent and effective
government regulatory system can be established to increase the credibility and transparency of
government policies, as well as to strengthen policy publicity and training for farmers to enhance
their awareness of their right to information and to defend their rights.
4. Analysis of Countermeasures for Old-Age Security in Rural China
4.1. Promoting Farmers’ Income Increase with Rural Revitalization Strategy
Rural revitalization includes characteristic townships and agricultural supply guarantees, and both are
important components of the rural revitalization strategy [8]. Characteristic townships can make full
use of the preferential policies issued by the state and the resources of the village location to develop
characteristic industries and promote characteristic culture; reasonably improve the industrial support
and accelerate the transformation and upgrading of local sunset industries; actively develop multi-
industry moderate scale operation, which does not destroy the beautiful relationship between man
and nature in harmonious coexistence, but also makes the resources used to achieve the highest degree
of utilization, and promotes the effective agglomeration of various similar industries. A mutually
friendly exchange of technical facilities accelerates the speed and quality of integration between the
upstream and downstream industries; the government, through the understanding of the employment
positions of local enterprises, provides more employment, entrepreneurship, and career opportunities
for rural people; improves the infrastructure, create a beautiful and livable production, living
environment, the formation of urban and rural areas as the core of the urban linkage and common
development [8]. Characteristic townships are a key link to effectively realize rural revitalization and
a reasonable way to promote the development of the three rural areas.
4.2. Advocating Rural Mutual Care for the Elderly
From the perspective of the current development of the practice of mutual aid in old age, the three
main techniques of mutual aid in old age, namely voluntary service, low-pay service, and “time bank,”
have been widely noticed and recognized. Since China began to pay attention to the aging of the
population brought about by the problem of old age has begun to vigorously recommend the elderly
in rural areas for mutual aid in old age, but the actual implementation of mutual aid in old age in
individual provinces did not meet the expected value of the model, the main reason for this is that, in
practice, too much attention to theoretical propaganda and ignored the construction of the rural
ecological environment as well as local customs and cultures, mutual aid in old age is not placed in
the village of the social trust of the village, and there is a lack of practical implementation of mutual
aid in old age technology. The main reason for this is that in practice, too much attention has been
paid to the propaganda of theory and neglected the construction of rural ecological environment and
local customs and culture, and the technology of mutual care for the elderly has not been placed in
the social trust of the villages, and there is a lack of social capital to carry out mutual care for the
elderly[9].
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231158
219
4.3. Improvement of Farmers’ Social Security System for Elderly.
First of all, the farmers’ social pension security legal system should be improved from the law to
clarify the implementation of China’s pension security system and specific methods. The relevant
institutions, departments, and personnel in violation of laws and regulations are to be strictly punished
by the law as a guideline to make the lawless elements hanging [10]. The pension security system
stipulated in the law, in addition to preventing lawbreakers, can also enhance the residents’ trust in
the system. Secondly, we should increase the subsidies for rural residents, expand the scope of
subsidies for old-age security, increase the number of subsidies, reduce the burden of personal
contributions, and encourage farmers to take the initiative to pay to reduce the difficulty of the
government’s collection of old-age pensions. Once again, improve the ability to preserve and increase
the value of pension funds. First, China’s pension has strong liquidity, but its capital accumulation
effect is weak; therefore, the management body of the national pension fund can purchase short-term
treasury bonds or savings to obtain short-term income to ensure that the funds in a fixed period of
time are not in a non-liquid idle state, to enhance the utilization rate of available funds.
5. Conclusion
With the gradual emergence of population aging, the problem of old-age security for the rural old-
age population tends to become more obvious. At the present stage, China’s rural old-age security
system can give most of the rural old-age population a certain degree of protection, but there is still
room for progress, and the country and the government should continue to find problems and solve
them quickly in the process of development. During the Fifth Plenary Session of the 19th CPC Central
Committee, the formal proposal to implement the national strategy for proactively addressing
population aging was put forward. A comprehensive plan for active management of population aging
was outlined, offering a robust foundation for achieving the goal of Socialism with Chinese
Characteristics in the New Era led by President Xi. China’s population aging has intensified in recent
years, with a large number of elderly people in rural areas and problems in old-age security to be
solved. However, due to a certain difference between China’s level of economic and social
development and the process and rate of population aging, resulting in a gap between old-age services
and the people’s expected goals to a certain extent, and an inability to adequately adapt to the needs
of old-age services for the elderly. In conclusion, we should fully respect the objective law of
population development, face up to the reduction of the demographic dividend caused by population
aging in China as an objective inevitable trend, and constantly deepen and optimize the theory,
formulate reasonable policies, find out the countermeasures and implement them effectively,
positively face a series of problems caused by population aging, and appropriately take care of the
economically weak elderly in rural areas, and construct scientific and reasonable rural elderly service
system, and fully solve the problem of farmers’ elderly service needs. Pension service system to fully
solve the problem of farmers’ old age. Of course, this paper is written to receive their cognitive level
and understanding of the constraints of the research process, and there are certain limitations. First,
in the content of the study, this paper is mainly through a large number of literature reviews,
comparative analysis, theory and practice combined with less. Secondly, in the research method, This
paper primarily employs literature review and comparative analysis as its research methods. The two
methods have certain limitations to a greater or lesser extent, and at the same time, this paper also has
subjectivity in the collection of data. In the future, further refinement can be carried out on data
collection as well as the expansion of research methods, which is conducive to the profound study of
the topic.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231158
220
References
[1] United Nations Report: World Population Expected to Exceed 8 Billion This November - Xinhua News (news. cn)
[2] Wang Hua, & Wang Lianggui. (2020). Current Situation of Population Aging and Countermeasures in Xinjiang
Uygur Autonomous Region. Xinjiang Traditional Chinese Medicine, 38(6), 4.
[3] Statistical Communiqué of the People’s Republic of China on the 2020 National Economic and Social Development
- Government of China (www.gov.cn)
[4] Severe Talent Loss in Elderly Care Institutions? Yunnan Plans to Train 50,000 Elderly Care Nurses by the End of
2022 - Today’s Headlines (toutiao.com)
[5] China Research Center for the Aging - China Urban and Rural Elderly Population Tracking Survey (2010) (crca.cn)
[6] Enhancing the Role of Long-Term Care Insurance in Providing Care for the Elderly - Legal Daily
(legaldaily.com.cn)
[7] Liu Yan. (2020). Analysis of Rural Elderly Care Issues in Liaoning. Agricultural Economics, (7), 2.
[8] Xie Suyan. (2022). Issues and Countermeasures of Characteristic Town Construction under the Rural Revitalization
Strategy: A Case Study of Dalian City. Agricultural Economics, (2), 56-58.
[9] He Xuefeng. (2020). Mutual Aid for the Elderly: A Way for Rural Elderly Care in China. Journal of Nanjing
Agricultural University: Social Science Edition, 20(5), 8.
[10] Guo Fengkai, & Jiang Tianhong. (2020). Research on Risk Control of Rural Fund Mutual Aid Associations in City
W. Rural Science and Technology, 11(34), 3.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231158
221
The Key Functions of Relevance and Relationship Effect in
Marketing: Based on the Analysis before and after
Restructuring of Huiyuan Group
Zongyi Li1,a,*
1Journalism and Communication College, Tianjin Normal University, Tianjin, China
a. yhan@tjnu.edu.cn
*corresponding author
Abstract: As the economy and society develop continuously, the speed of commodity renewal
is accelerated, and the competition between brands is becoming more and fiercer. Fruit juice,
as one of the most popular commodities, is deeply loved by all kinds of people and meanwhile,
it is also a traditional and dynamic industry. Recently, Huiyuan Group, a celebrated Chinese
fruit juice company, has adopted various measures to cope with the fierce competition
effectively in the commodity market, such as broadening marketing channels, strengthening
market research combined with promoting product innovation, and enhancing the platform
interaction in order to maintain customer relations. This paper analyzes the problems existing
before the marketing bankruptcy of Huiyuan Group, and the changes made after the
reorganization, based on the key function theory of “relevance” and “relationship” effect in
4 Rs. The analyzing results find that Huiyuan Group focuses on the “relevance” and
“relationship” effect in the marketing process, broadens the marketing channels such as
TikTok, Xiaohongshu and Weibo, and pays attention to the user platform interaction and
product innovation, all of which bring considerable benefits to the enterprise marketing, and
are conducive to the marketing and maintenance of the enterprise brand.
Keywords: case analysis method, literature analysis method, integrated marketing, 4 Rs
1. Introduction
According to the data of the National Bureau of Statistics, the annual output of fruit juice and fruit
juice drinks in China has exceeded 170 million tons, ranking the fourth largest in the domestic
beverage industry [1]. As the leader of the juice industry, Huiyuan Group, however, did not pay
attention to the application of the two key effects of “correlation” and “relationship” in the marketing
in the early marketing process, resulting in a decrease with fluctuation its net profit before 2017, and
was delisted in 2020. Encountered with the survival crisis, Huiyuan Group carried out innovative
measures related to targeted marketing strategy, and finally reversed the crisis, and earned a lot of
good reputation with increasing profits. By using the literature analysis method and the case study
analysis of Huiyuan Group, the paper innovatively summarizes a set of systematic and efficient
integrated marketing model. The article also provides an advanced integrated marketing strategy
template for other juice enterprises to improve their core competencies.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231163
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
222
By depicting the defects of Huiyuan Group in the early marketing process and its innovative
marketing strategy after restructuring, this paper explores how the relevance and relationship effect
in 4 Rs theory is applied to promote the development in integrated marketing. Finally, this paper
attempts to discover what kind of integrated marketing strategy to can adapt to today’s competitive
market environment and how to promote the benign development of the juice industry.
Under today’s new media marketing environment, employing the theory of “relevance” and
“relationship” effect in 4 Rs into the fruit juice industry research can fill the blank in the theory of
enterprise integrated marketing. And also Huiyuan Group puts the theory into practice, playing a
guiding role in the fruit juice enterprise integrated marketing in the real society, and bringing the
economic and social benefits in the sustainable development of the enterprise. 4 Rs theory is a new
marketing theory including the four-part effect of relevance, reaction, relationship and reward, which
expounds new four elements of marketing. Relevance means the establishment of a relationship with
customers; reaction means improving the market response speed; relationship refers to the
relationship marketing; and reward is the driving force of the marketing [2]. The traditional marketing
strategy is sales-oriented, while 4 Rs focuses on the maximization of customer correlation; the
traditional marketing concerns market share more, while 4 Rs cares more about customers loyalty. In
other words, the traditional 4 Ps, 4 Cs marketing theory is only isolated, with no related effects. The
marketing research based on the theory of 4 Rs focuses on the mutual benefits of the enterprise and
customers. As competition-oriented, it satisfies the customer basic demand, creates demand actively,
and develops innovative integrated marketing strategies, making the enterprise and customers form a
unique sticky relationship and bringing the competitive advantage. At the same time, the theory of
“relevance” and “relationship” effect in 4 Rs has its own deficiencies and defects. For instance,
developing the strategy needs the enterprises to achieve certain objective conditions, or pay the
corresponding cost, which prohibits some enterprises from conducting 4 Rs marketing strategy.
Nevertheless, mastering the 4 Rs marketing theory of relevance and relationship, is of guiding
significance both in innovative enterprise integrated marketing, and in the improvement of the
existing marketing strategy.
2. Case Analysis of Huiyuan Group
2.1. Introduction of Huiyuan Group
Beijing Huiyuan Beverage and Food Group Co., Ltd., founded in 1992, is a large modern enterprise
specializing in fruit juice related products. In 2001, Huiyuan’s sales revenue reached 1.5 billion yuan,
occupying 23% of the juice market. After maintaining its leading position in the industry in the
following years, China Huiyuan Juice Group Co., Ltd., which was split by the company, was listed
on the main board of the Hong Kong Stock Exchange in February 2007. In September 2008, Coca-
Cola announced that it would buy all the shares in Huiyuan Juice for HK $17.92 billion, but the
Ministry of Commerce rejected the deal on antitrust grounds, and the plans failed. During this period,
Huiyuan Group focused on the acquisition strategy, ignoring the update and improvement of its
marketing strategy, which led to its lead in sales gradually narrowed and its debt increased little by
little. From a brilliant market value of over HK $31.3 billion, it plunged until suspension in 2017 and
then it was cancelled by the Hong Kong Stock Exchange in February 2020[3]. In 2021, Shanghai
Wensheng Asset Management Co., Ltd., as a restructuring investor, planned to invest 1.6 billion yuan,
and after the implementation of a series of innovative self-rescue measures of integrated marketing
strategies, Huiyuan Group’s sales performance increased significantly year-on-year, gradually
reducing debt and changing a loss into a profit.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231163
223
2.2. Ignorance of the Marketing in Huiyuan Group before Bankruptcy and Its Consequences
Before its bankruptcy, Huiyuan Group had always adopted the old business and operation model.
Even in the era of mobile Internet, it still maintained the outdated marketing mode. It ignores the
important measures to associate with customers in terms of business and demand, resulting in a slow
market expansion, decreased consumer acceptance, and profits declining. In 2008, when Coca-Cola
acquired Huiyuan, in order to improve the assessed value of the assets, it expanded production blindly.
Meanwhile in order to reduce the cost, it reduced large-scale marketing channels, and therefore the
employees in the group have been reduced from 9,722 in 2007 to 4,935 in 2007, and sales staff also
fell from 3,926 to 1,160, thus a key link between the corporate and customer damaged. This made the
weak connection with customers even weaker, and the product distribution was also blocked,
eventually resulting in years of debt growth. As shown in figure 1, the sales growth of Huiyuan juice
was insignificant, and the growth rate reached the lowest point of 0.46% in 2009. What is more
obvious is that the net profit in Huiyuan began to decline in 2008, and reached the historical lowest
point with minus more than 200 million yuan in 2015.
Figure 1: 2006 Receivables and net profit of Huiyuan Juice from the year to 2017 [4].
Before the bankruptcy, Huiyuan Group indeed seized the opportunity in the juice market, but it
neglected the building of the relationship with customers, lacked good communication with customers
and the understanding of consumers’ needs, and did not foster innovation, thus becoming overtaken
because of lagging its sales [5]. Brands such as Uni-President, Master Kong and Pepsi continued to
pay attention to the relationship between enterprises and customers, taking the lead in the updating
products such as “Fresh orange more”, “Daily C” and “Fruit Colorful”, expanding their production,
and seizing market share constantly. Huiyuan, on the other hand, after the obstruction of the group
acquisition event, maintained the old hierarchical marketing model and the same single variety of the
production. And it did not consider how to continue to expand the juice market, nor did it focus on
maintaining existing relationships with customers. This, therefore, greatly damaged the establishment
of a mutually beneficial relationship between enterprises and customers, and eventually, it lost its
customers, its product market share was shrinking and the loss and debt of Huiyuan increased
dramatically. According to the Figure 2, during the eight years from 2009 to 2017, the debt amount
increased year by year, and the difference was five times earlier.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231163
224
Figure 2: After the failure of Coca-Cola acquisition, Huiyuan Group’s debt changed [6].
2.3. Huiyuan Group’s Emphasizing on “Relevance” and “Relationship” Effect and Its
Achievement
After restructuring in 2021, Huiyuan Group realized that the channels for customers to receive
information in the network media environment had changed, and that it should constantly pay
attention to strengthen the connection with customers, and that it ought to make changes in the single
marketing channel before [7]. The company began to continuously invest in expanding its marketing
channels, increase the audience marketing coverage, strengthen the original media promotion such as
TV advertisements and newspaper reports, and also make efforts in marketing on TikTok,
Xiaohongshu, Weibo and other new online media platforms. Besides, it invited a number of popular
celebrities to stand up for their products and invested in multiple groups, which strengthened the
relationship between the enterprise and customers, and eventually expanding the targeted consumers,
and increasing the sales. Huiyuan Group also developed the marketing strategy of direct selling, so
as to establish efficient relationships with customers at all levels directly. It attracted a large number
of new customers, and make the product market share to increase by allowing customers to know
more about the enterprise and products, and buy these products and goods in a more convenient and
affordable way. After the implementation of the new marketing strategy, the sales of 2021 increased
by 49.7% in 2020, and the gross profit increased by 39.5% in 2021. In addition, the satisfaction index
of the juice industry ranked first. From January to April 2022, the online product revenue of Huiyuan
brand business went up by more than 200%.
4 Rs marketing theory emphasizes the use of “relationship” marketing and the “relationship”
marketing focuses on one-to-one interaction within different ranges. As for Huiyuan Group, it
establishes collaborative and stable relationship with customers, and cooperates between buyers and
sellers in terms of effective innovation behavior of related products [8]. In details, initially Huiyuan
Group increased the relationship with customers, through the comments with potential consumers in
its official media platforms, such as TikTok, and Weibo, and in this way Huiyuan Group can discover
the customer psychology and their purchase motivation. Accordingly, it fostered the product
innovation to meet the personalized needs of customers, and from 2021 to 2023 it developed some
new products such as sea-buckthorn pulp, probiotics plum juice, and plum juice, making the products
younger, more fashionable, more attractive and more personalized [9,10]. Secondly, based upon the
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231163
225
relationship effect, Huiyuan Group revived its juice product marketing through Internet celebrity
drainage and article pushing. In this way, it maintained the relationship with customers and brought
mutual benefits both for Huiyuan Group and the consumers. On October 5, 2022, in Huiyuan brand
live, more than 1.74 million people were watching, and till October 11, 2022, 340000 followers were
increased. Huiyuan also provided the lower price in their live streaming to maintain the benign
relations with customers, and in this way it achieved a win-win situation, with a surge in sales and
profits. And the endeavor in its juice producing and the various remarkable marketing methods such
as the lottery in Weibo and Xiaohongshu raised brand awareness, increased the overall sales, and
maintained the long-term relationship with customers.
2.4. Analysis and Comparison
According to the analysis and research of Huiyuan Group in the given two stages, it is found that the
marketing strategy of Huiyuan Group has changed greatly and brought a relatively contrasting
revenue effect. The Enterprise marketing attaching great importance to the “relevance” and
“relationship” effect in 4 Rs theory will bring considerable sales to the enterprise, thus promoting the
benign development of the enterprise; otherwise, if the enterprise ignores the use of “relevance” and
“relationship” effect, it will face stagnating sales and difficult development.
In the marketing strategy, the disadvantages such as the rare relevance between enterprises and
customers and the single marketing methods will lengthen the distance between the manufacturers
and consumers, resulting in unsalable products. At the same time, enterprises do not pay attention to
the maintenance of relations with a variety of customers, will make the product recognition and sales
decrease gradually, resulting in the decline in corporate profits, debt increase and even the survival
crisis. After the reorganization, Huiyuan Group innovates its marketing strategy, focuses on the
investment of relevance and relationship, and expands marketing channels by focusing on the effect
of “relevance” and “relationship”. It has established the “marketing matrix”, developed and increased
the investment in TikTok Weibo, Xiaohongshu and other platforms, and enhanced the platform
interaction and product innovation, making the marketing approaches more diverse [11,12]. While
improving the relations with customers at all levels, Huiyuan Company also pays attention to maintain
the win-win relationship between the enterprise and customers. Thus, its sales increase year by year,
which means Huiyuan Group not only obtains its economic profits, but also improves its brand
awareness. All these show the importance of the correlation and positive relationship between
enterprises and customers. This new marketing mode is also suitable for other enterprises of
commodity production and sales, especially for those which are innovation-oriented, and are
incapable to fully apply the four aspects of 4 Rs theory due to their limitations of conditions and
strength. This way is an appropriate approach to increase of sales and promote the connection with
users.
3. Conclusion
Based on the marketing analysis of Huiyuan Group before its bankruptcy and after restructuring, this
paper focuses on the importance of “relevance” and “relationship” effect 4 Rs theory in the field of
the enterprise marketing. The research finds that without the “relevance and relationship”, the
marketing development of enterprises cannot be achieved. Nowadays, with the rapid development of
the Internet and the increasing media where customers receive marketing more easily, it is particularly
significant for enterprises to establish good relations with customers in different media, which means
that without good connection, customers cannot be retained, and enterprises cannot survive without
customers. The relationship between the enterprise and customers is also an important part of the
marketing strategy. On the one hand, the benign interactive relationship can promote the enterprise
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231163
226
to develop new products, improve the benefits, and realize the customized product value; on the other
hand, in the conditions where owning stable customers are crucial to lead the role in the market
nowadays, maintaining the stable and healthy relationship has become the “engine” of enterprise
marketing development. In the end, the article analyzes and innovates a marketing strategy, with the
relevance and relationship effect as the core, applicable to the marketing investment of small and
medium-sized enterprises. This strategy helps enterprises to reduce marketing costs, increase the
input-output ratio of industries, and promote the benign development of enterprises.
The study presented in this paper also has some limitations. In the economic ability, not all
enterprises are capable enough to establish relevance with a variety of customers. And since the
commodity market is dominated by the economic benefit, Huiyuan Group, as the lead of the market,
can pay attention to improve the relation with customers to innovate its marketing strategy without
worrying insufficient fund in a short time. So, this makes some small and medium-sized enterprises
with short capital chain or long production cycle unable to adopt this strategy. In the future, under the
application and continuous improvement of the theory of “relevance” and “relationship” effect, it is
expected to explore a more cost-effective, faster and more direct integrated marketing strategy.
References
[1] Li Wei. Research on the marketing strategy of fruit juice drinks. Jilin University, 2021.
[2] He Yunjing. 4 New theory of RS marketing. China SMEs, 2001 (04): 27.
[3] Wang Ying. Case study of the financial distress of Huiyuan Juice. The Chinese Academy of Fiscal Sciences, 2022.
[4] Li Yinfeng. He Miao and Huiyuan are almost overwhelmed, NFC can let consumers fall in love with juice again,
2018.08.03, 2023.08.25, https://m.jiemian.com/article/2356558.html
[5] Ren Xin. Research on the fruit juice marketing strategy of HS Company. Northwestern University, 2019.
[6] Bread finance, 4.2 billion illegal loans, over 10 billion liabilities, facing the delisting crisis of Huiyuan juice
experienced what? 2018.08.01, 2023.08.25, https://baijiahao.baidu.com/s?id=1607525098684178194&wfr=sp
ider&for=pc
[7] Kan Xiong, Liu Yunlong, Teng Tian. Research on integrated marketing strategy of restaurants based on 4 Rs theory.
Price Monthly, 2012, (06): 58-61.
[8] Chen Li, Zhang Lili. Organizational marketing discussion based on 4 Rs theory. Legal System and Society, 2007,
(09): 636-637.
[9] Li Shengyan. Brand communication characteristics and marketing strategies of FMCG enterprises —— Take
Huiyuan Juice as an example. China News Industry, 2017, (06): 10-11.
[10] Song Mingming. Research on the integrated marketing strategy of new media under the background of media
convergence. International Public Relations, 2021, (12): 85-87.
[11] Qiu Caihua. Research on marketing integration strategies of small and medium-sized enterprises in the new media
Era. LaoBrand Marketing, 2021, (02): 107-108.
[12] Liu Mengjie. Marketing integration strategy of small and medium-sized enterprises in the new media era. Shanghai
Commercial, 2022, (04): 48-50.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231163
227
Comparative Study on the Low Birth Rate Process in China
and Japan and the Impact on University Graduates’
Employment
Mingwei Wang1, a, *
1College of Economy and Trade, Shandong Management University, Dingxiang Road, Jinan City,
China
a. julin@ldy.edu.rs
*corresponding author
Abstract: The low birth rate is a global phenomenon that cannot be overlooked, having
profound implications for the socio-economic conditions of many nations and regions. The
declining birth rates in China and Japan have been a significant part of the global discourse
on this topic. This paper analyzes the impact of low birth rates on the employment of
university graduates, encompassing both positive effects such as an increase in employment
rates, and negative effects including a decrease in innovation and increased difficulties in
promotions. The study observes that there is a roughly 15-year gap in the shifts in the low
birth rate phenomenon and the employment conditions of university graduates between China
and Japan. The differences between the two nations mainly manifest in the timeframe and
pace of the declining birth rates and the associated time lags. A commonality is that the
fundamental cause for the low birth rate in both countries is the high cost of child-rearing.
Lastly, the paper proposes suggestions for the Chinese government and university graduates,
drawing upon the countermeasures and experiences of Japan. These suggestions helps
enhance the employment competitiveness of university graduates and promote sustainable
economic and social development, offering valuable references for both governments and
university students.
Keywords: low birth rate, university graduates’ employment, China, Japan
1. Introduction
Globally, the phenomenon of declining birth rates has emerged as a matter that cannot be disregarded.
This trend, characterized by a diminishing fertility rate which subsequently results in a gradual
decrease in the proportion of young population, is closely affiliated with various factors including
economic development, governmental policies, and the dwindling desire for childbearing amongst
the reproductive age groups. The repercussions of declining birth rates exert profound influences on
the socio-economic landscapes of numerous nations and regions.
In this context, the declining birth rates witnessed in China and Japan constitute a significant
segment of the global discourse on this topic. Since 2018, China has experienced a rapid decline in
its fertility rate, marking its entrance into a phase where the issues stemming from a declining birth
rate have become pronounced. Conversely, Japan has endured the predicaments associated with a
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231171
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
228
consistently low birth rate for many years. Historically, China witnessed a period of rapid population
growth over several decades, which brought an influx of inexpensive labor force, thereby fueling a
substantial demographic dividend that propelled immense economic growth. Therefore, the sharp
population decrease in recent years necessitates an urgent discussion on its implications for the
employment prospects of Chinese university graduates and its subsequent impact on socio-economic
development. On the other hand, Japan, a nation perennially grappling with declining birth rates, has
encountered significant impacts on its socio-economic fabric. Notwithstanding, Japan has managed
to stabilize university graduate employment and sustain positive economic growth, fostering an
economic model that coexists with a declining birth rate. The employment scenario for the youth in
Japan has undergone a significant transformation from the early 2000s to the present, with various
similarities and disparities evident in the circumstances of China and Japan across different time
periods.
Regarding the literary discourse on declining birth rates, Brinton et al. attribute the cause to a
myriad of factors including the educational and employment opportunities for women, the distribution
of familial and childcare responsibilities, shifts in the social and political milieu, amongst others.
Additionally, the discrepancy between the desire and actual practice of childbearing stands as a
crucial factor exacerbating this trend [1]. Many scholars have analyzed the implications of declining
birth rates. For instance, Chen posits that it harbors negative repercussions, restraining consumption
growth through avenues like population totals and income distribution, consequently affecting the
healthy and sustainable development of the economy [2]. Conversely, Liu perceives positive societal
influences stemming from declining birth rates, including heightened savings substituting
expenditures on education and healthcare, fostering a high savings rate occurrence [3]. In addressing
this issue, several scholars have proposed pertinent strategies. For example, Tanasa and Serban
suggest that labor market regulations and governmental policy adjustments could to some extent
foster fertility rates, although various political, economic, and cultural factors might impede such
adjustments [4].
On the basis of the previous research, this paper aims to scrutinize the current status of declining
birth rates in China and Japan, and its repercussions on university graduate employment, while
dissecting the similarities and discrepancies observed between the two nations in this regard.
Integrating the experiences of Japan with the prevailing circumstances in China, this study endeavors
to proffer recommendations to the Chinese government and university students grappling with the
declining birth rates, thereby holding intrinsic research value in fostering strategies to enhance
university graduate employment amidst this trend.
2. Background and Causes of the Low Birth Rate in China and Japan
2.1. Background and Causes of the Low Birth Rate in China
2.1.1. Background of the Low Birth Rate in China
According to the National Bureau of Statistics of China, and the China Statistical Yearbook 2022, by
the end of 2022, the total population of China decreased by 850,000 compared to the end of the
previous year. Throughout the year, there were 9.56 million births and 10.41 million deaths, resulting
in a natural population growth rate of -0.60‰. This marked the first time since the establishment of
the People’s Republic of China in 1949 that the population experienced negative growth under natural
development conditions. In 2020, the birth rate in China had already dropped to an ultra-low fertility
rate standard of 1.3[5]. Since the mid-20th century, China’s population had maintained a trend of
rapid growth. However, starting in 2018, China’s birth rate experienced a sharp decline, interrupting
the previously sustained high-speed population growth trend (see Figure 1). Although the population
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231171
229
was growing during this period, the growth rate was slowing down, and the negative population
growth phenomenon in 2022 was a general trend [5].
Figure 1: China’s Birth Rate Statistics.
2.1.2. Reasons for the Low Birth Rate in China
Due to high child-rearing costs, among other factors, the childbearing willingness of couples of
childbearing ages has decreased, leading to a year-by-year decline in the birth rate. The primary
reason for the contemporary low birth rate phenomenon in China is the continuously escalating cost
of raising a child. The steep child-rearing costs have led to a collapse in the willingness to bear
children for many individuals [6]. Such ideas are disseminated among the population of childbearing
age through various channels, including the internet, fostering a greater sense of identification with
the concept of giving birth to fewer but higher quality offspring. Combined with the young people in
China continually seeking higher standards of living, has culminated in an ideology that is
acknowledged by many. Consequently, there has been a larger scale decline in the willingness to have
children. The sharp decrease in the birth rate in recent years is the result of the combined effects of
changing perceptions and the realities of the current situation.
2.2. Background and Causes of the Low Birth Rate in Japan
2.2.1. Background of the Low Birth Rate in Japan
In the post-war period, Japan experienced a surge in population due to the repatriation of soldiers and
individuals from its colonies, and the government’s intention was to support birth rate control.
However, population control in Japan was not implemented through compulsory measures, but
stemmed from the unwillingness of couples of childbearing age to have children due to high child-
rearing costs [7]. In 1989, Japan faced a significant shock with a birth rate of 1.57. It was not an
isolated incident. In 1995, the birth rate plummeted to a historic low of 1.42. In 2005 and 2006, the
birth rates in Japan were 1.26 and 1.32, respectively. Although fluctuations in the birth rate have been
observed, the general trend is a decline (see Figure 2). The consequence of the low birth rate is that
Japan, having experienced negative population growth starting in 2009, continues to progress further
down the path of population decline.
2.2.2. Reasons for the Low Birth Rate in Japan
The reason behind Japan’s low birth rate is the expensive costs associated with raising children. In an
effort to maintain their current standard of living without a decline, families with couples of
childbearing age are unable to accommodate higher child-rearing costs, thus are unwilling to raise
more children [7]. Furthermore, the influence of the economic bubble has deeply ingrained a lower
willingness to have children within the populace.
1.67
1.77 1.81
1.55 1.5
1.28
1.16
1.15
1.25
1.35
1.45
1.55
1.65
1.75
1.85
2015 2016 2017 2018 2019 2020 2021
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231171
230
Figure 2: Japan’s Birth Rate Statistics.
2.3. Comparative Analysis of the Low Birth Rate in Both Countries
2.3.1. Analysis of Time Lag Patterns
One pattern can be observed in the low birth rate processes of the two countries: the significant years
in which both countries entered a deep phase of low birth rate, namely 2005 and 2018, and the years
of negative population growth, 2009 and 2022. Both sets of years have a gap of about 15 years, and
the gap between the two respective years within each country is approximately 4 years. This pattern
also indicates certain similarities in the low birth rate trends between China and Japan.
2.3.2. Differences and Similarities
One of the differences between the two countries is the period when they entered the deep phase of
low birth rate. Taking the point of negative population growth as a benchmark, China achieved
negative population growth in 2022, while Japan reached this point in 2009. China faced a slump in
birth rates in 2018, while Japan encountered a nadir in birth rates around 2005. These are both
significant markers indicating the onset of a deep phase of low birth rate, but they occurred at different
times. In the process of entering this deep phase, China demonstrated a more abrupt transition, while
Japan started experiencing a gradual decline in birth rates from the 1980s, indicating a slower process.
The biggest commonality between the two nations is the reason behind the low birth rate. Although
the causes leading to a low birth rate are complex and multifaceted, a central similarity between China
and Japan is that the primary cause in both countries is the continuous rise in child-rearing costs,
which has resulted in a decline in the willingness to have children among the population of
childbearing age.
3. Comparison of Employment Conditions of University Students in China and Japan at
Different Stages
3.1. Comparison of University Students’ Employment Conditions During the Deep Phase of
Low Birth Rate
This paper considers the population born during periods of higher birth rates. When the university
students among this group reach the employment age (around 20 years old), it corresponds to the
period when their respective countries are entering the deep phase of low birth rate. This study
compares the employment conditions of this group in both countries. Specifically, considering the
population born around 1975-1985 in Japan, who entered the workforce approximately between
1995-2005 and the population born around 1993-2003 in China, who entered the workforce
approximately between 2013-2023.
1.25
1.35
1.45
1.55
1.65
1.75
1.85
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231171
231
3.1.1. Employment Situation of University Students in Japan During the Deep Phase of Low
Birth Rate
During 1975-1985, the birth rate in Japan was between 1.76 and 1.91, a period when the birth rate in
Japan was relatively high, approximately 0.5 higher than 20 years later.
By the late 1990s in Japan, university graduates faced increasingly difficult to secure employment,
and the unemployment rate was high. The skills and attributes necessary for employment were
relatively insufficient [8]. Companies preferred innovative talents with capabilities over mechanical
and submissive workers. During this period, the young people, facing a high-threshold employment
environment, began to lower their expectations for life and accept the reality of unemployment [9].
During this period, Japan’s unemployment rate surged from the usual 1%-2% before the 90s to 5%-
6% around the millennium [10].
3.1.2. Employment Situation of University Students in China During the Deep Phase of Low
Birth Rate
From 1993 to 2003, China’s birth rate ranged between 1.57 and 1.69. Although the one-child policy
suppressed the growth of the birth rate, it remained relatively high, sustaining a rapid population
growth.
When the population born during this period reached employment age between 2013 and 2023, a
situation similar to Japan’s emerged. With the yearly increase in graduates, more individuals flooded
the labor market, intensifying competition for jobs and exacerbating employment difficulties for
graduates [11]. Due to increased competition, the phenomenon of “lying flat” emerged among
Chinese youth, characterized by lowered aspirations for real life, and downgraded consumption and
desires [12].
3.1.3. Comparison Between the Two Countries
From the analysis above, it can be seen that there are many similarities in the employment situation
of university students in both countries when entering the deep phase of low birth rates. Both
exhibited increased difficulties for university graduates in finding employment and lowered life
expectations. Despite a 15-year gap, the manifestations in China and Japan are highly similar during
this period.
3.2. Comparison of Current Employment Situations for University Graduates
3.2.1. Current Employment Situation for University Graduates in Japan
In Japan, the deepening of the declining birth rate trend has led to a continuous decrease in the number
of graduates. The job market generally exhibits a state of supply exceeding demand. Currently, the
employment rate for university graduates in Japan is high, making it easier for them to find jobs.
3.2.2. Current Employment Situation for University Graduates in China
The present employment and labor supply situation in China is in a phase where demand exceeds
supply, with the overall scenario being that it is challenging for university graduates to find jobs, and
companies have more leeway to suppress wages.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231171
232
3.2.3. Comparison Between the Two Countries
Compared to China, Japanese graduates nowadays find it easier to secure jobs with decent incomes
and benefits than their Chinese counterparts of the same age and individual conditions. In contrast,
Japanese companies need to incur higher costs than Chinese companies to hire graduates of the same
caliber.
4. The Impact of Low Fertility Rates on University Graduates’ Employment Opportunities
A standard university student takes approximately 20 years from birth to join the workforce. This
implies that the changes in the societal birth rates that can influence university students’ employment
would be seen about 20 years later. It is noted that the sharp decline in China’s birth rate began around
2018, and the current job market for university graduates in China is still dominated by those born
during the high fertility rates of the 2000s. The decrease in birth rate in China has not yet significantly
impacted the employment of university students, lacking typicality.
4.1. Positive Impact of Decreasing Birth Rates on University Students’ Employment
4.1.1. Positive Impact Evidence
Figure 3 and figure 4 illustrate the employment rates of university students in Japan from 2013 to
2023, simplified for analysis. Considering that individuals generally join the workforce around the
age of 20, we have extracted birth rate data between 1993 and 2003, as illustrated in Figures 3 and 4.
The data exhibits a general decline in birth rates in Japan between 1993 and 2003. Conversely, the
employment rates for university graduates twenty years later, from 2013 to 2023, display an overall
increasing trend. The noticeable dip in the employment rates in 2021 and 2022 is primarily attributed
to the global prevalence of COVID-19, an external factor. Nevertheless, the overarching trend is an
increase. This suggests that the decreased birth rates in Japan would lead to a rise in university
graduates’ employment rates as this cohort enters the working age group.
Figure 3: Japanese Birth Rate Curve.
Figure 4: Japanese University Graduates
Employment Rate Curve.
4.1.2. Positive Effects: Causes
The positive impact of declining birth rates on the employment rate of university graduates can be
attributed to several factors. The primary reason is the reduction in the number of births, which results
in a decreased supply of labor in the job market. With fewer individuals entering the workforce to
1.46 1.5
1.421.43
1.391.38
1.341.36
1.331.32
1.29
1.25
1.3
1.35
1.4
1.45
1.5
93.50%
94%
96%
97%
97.50%
98%
97.50%
98%
96%96%
97.50%
93.50%
94.00%
94.50%
95.00%
95.50%
96.00%
96.50%
97.00%
97.50%
98.00%
98.50%
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231171
233
meet a constant or even increasing demand for labor, competition among university graduates (the
suppliers of labor) decreases, potentially driving up the price of labor (i.e., wages). Consequently,
university graduates are more likely to secure jobs with satisfactory remuneration.
Moreover, there are various other factors contributing to the rise in the employment rate of
university graduates, including government initiatives to boost employment and the emphasis by
universities on career-oriented education. These aspects will be analyzed in greater detail in the
subsequent sections of the paper.
4.2. Negative Impacts of Declining Birth Rates on University Graduate Employment
While the deepening trend of declining birth rates generally appears to promote the employment
prospects of university graduates, there are also negative repercussions that may not be directly
reflected in employment rates.
4.2.1. Decline in Innovation Potential
Young individuals are often regarded as the primary drivers of innovation. A reduced youth presence
in the job market, resulting from declining birth rates, could weaken the overall innovative
capabilities of corporations and society at large. This could diminish the pursuit of novel technologies,
business ventures, or strategic initiatives, thereby affecting the employment opportunities and growth
prospects for university graduates.
4.2.2. Stagnation or Decline in Wages
Though a labor shortage theoretically leads to wage increases, the actual outcome may vary. If
companies decide to cut costs in response to economic slowdowns, or if they pivot towards
automation to reduce their reliance on human labor, wage growth for university graduates could be
curtailed.
4.2.3. Limited Career Progression
The aging population brought about by declining birth rates can result in many senior positions being
held by middle-aged or elderly individuals for extended periods. This scenario can limit the promotion
opportunities and overall career advancement for younger university graduates.
5. Countermeasures, Experiences, and Insights
While a declining birthrate may alleviate the difficulty of university graduates finding employment,
it will ultimately have a general negative impact on the socio-economic fabric, along with a host of
other issues [13]. Consequently, there will naturally be factors that are detrimental to the employment
prospects of university graduates. However, the reality is that Japan has indeed achieved a year-on-
year increase in the employment rate of university graduates under the context of a declining birthrate,
with the satisfaction levels of employed graduates also showing a rising trend. On the other hand,
China is a country on the brink of facing the repercussions of a declining birthrate on the employment
of university graduates. Hence, it is imperative to analyze the strategies and experiences in Japan,
identify the current shortcomings in China, and propose recommendations for the Chinese
government, universities, and students.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231171
234
5.1. Japan’s Experience in Strategies for University Graduate Employment
Japan’s university employment guidance is targeted, offering specialized employment guidance
tailored to students with diverse characteristics. Japan boasts a unique “experiential employment”
system, enabling students to precisely identify roles they are both interested in and competent for.
Concurrently, numerous policies and laws have been implemented to facilitate effective employment
for university graduates [14]. Japan has a comprehensive employment service system that thrives on
the trilateral interaction between the government, higher education institutions, and the broader
society [15].
5.2. Current Shortcomings in China
5.2.1. Shortcomings at the Government and Societal Levels
In China, the efficiency of employment agencies is sub-optimal, with ambiguous delineations of
authority and responsibility. The employment promotion policies and regulations are not well
developed. The social employment service system lacks robustness, and the related departments are
not standardized in their management [14]. China’s relatively closed internet environment and limited
press freedom impede the dissemination of diverse information. This results in a significant
information gap between university students and the real world, further complicating their
employment prospects.
5.2.2. Shortcomings at the University Level
Chinese universities are often deprived of academic independence, with ideological content exerting
considerable influence on higher education. This interference is detrimental to cultivating the critical
thinking skills essential for employment. The feasibility of employment guidance provided by
Chinese universities to students is low, limiting its practical application in facilitating student
employment.
5.3. Recommendations for the Chinese Government, Universities, and University Students
5.3.1. Suggestions for the Government and Universities
(1) Strengthen Cross-Border and International Exchanges: Encourage universities to collaborate with
enterprises from various industries, establishing “experiential employment” to allow students to
immerse in real work environments. It is also recommended to moderately ease internet restrictions
to open more platforms for international exchanges, helping university students bridge the
information gap between domestic and international arenas and foster a global perspective.
(2) Promote Academic Independence in Universities: Transform universities into hubs for
flourishing academic research. Reduce the interference of ideology in university education and
encourage intellectual exchanges of diverse viewpoints. This would foster students’ abilities in
independent thinking and problem-solving.
(3) Enhance Policy Support and Efficiency of Employment Agencies: providing more policies to
support for university student employment. Learn from the efficient management models of Japan to
ensure that domestic employment agencies can provide quick and accurate services for university
students and enterprises.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231171
235
5.3.2. Suggestions for University Students
(1) Focus on Skills Development: In addition to academic and professional skills, it is vital to cultivate
soft skills like communication, teamwork, and critical thinking. Develop the ability to critically
assimilate diverse viewpoints, aligning professional knowledge with practical work experiences.
(2) Build a Global Perspective: Utilize more online resources both domestically and internationally,
participate in foreign courses, or collaborate with student teams abroad to complete projects, thereby
cultivating a global viewpoint.
(3) Enrich Social Practices: Establish interpersonal exchanges with people from diverse
backgrounds outside of the university setting, facilitating interpersonal interactions in the workplace.
Strive to secure more internship opportunities during university years to accumulate experience.
With the deepening of the low birth rate trend, coupled with the continuous advancement of
globalization and technological progress, the future job market for university graduates will face even
more challenges and opportunities. In this scenario, the government, universities, and university
students need to constantly adapt and innovate to address these challenges and opportunities. Future
studies can further explore how to utilize advanced technologies such as big data and artificial
intelligence to enhance the matching and accuracy of employment information; how to effectively
implement employment-oriented reforms in education; and how to effectively promote the
development of employment capabilities among university students.
6. Conclusion
This paper analyzes and compares the impact of the declining birth rate in China and Japan on college
graduates’ employment, reaching several conclusions. Firstly, the positive impact mainly manifests
in the increase of employment rate among college graduates. Secondly, the negative impacts include
a decrease in innovative capability, reduction in wages, and limitations on promotions. The
differences between China and Japan are primarily reflected in the time span and speed of the
declining birth rate, as well as the relative time differences. A commonality between China and Japan
is that the fundamental cause of the declining birth rate in both countries is the high cost of child-
rearing. On the basis, this paper proposes relevant suggestions for the Chinese government,
universities, and college students. It holds research value in terms of how to promote the employment
of college students in the wave of declining birth rate.
However, this paper only conducts research from the perspectives of fertility rate and college
graduate employment rate statistical analysis and qualitative analysis. In the future, empirical tests
can be carried out on a more sufficient data foundation to validate the conclusions more effectively.
References
[1] Brinton, Mary C., et al. (2018) Postindustrial fertility ideals, intentions, and gender inequality: A comparative
qualitative analysis. Population and Development Review, 44, 2, 281-309.
[2] Chen, Xiaoxiao. (2022) “The Impact and Countermeasures of Population Aging and Low Birth Rate.” Economic
Research Guide, 21, 45-47.
[3] Liu, Pengfei. (2022) “Research on the Impact Mechanism of ‘Low Birth Rate’ on Family Saving Rate.” Journal of
Central University of Finance and Economics, 3, 81-92+104.
[4] Tanasa, Serban. (2011) Politics, Culture and Fertility Trends. Dissertation.
[5] Yuan, Xin, Du Peng, Tong Yufen, et al. (2023) “Special Discussion: Negative Population Growth and High-Quality
Population Development in China.” Journal of Beijing Administrative College, 145, vol. 3, 11-25.
[6] Yang, Zhao. (2022) “Viewing the Decline of Fertility Rate in Contemporary China from the Perspective of
Historical Materialism.” Jiangnan Forum, 10, 58-61.
[7] Ueno, Chizuko. (1998) “The declining birthrate: Whose problem?.” Review of Population and Social Policy, 7,
103-128.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231171
236
[8] Nguyen, Danh, Yanagawa Yoshinari, and Miyazaki Shigeji. (2005) “University education and employment in Japan:
Students’ perceptions on employment attributes and implications for university education.” Quality Assurance in
Education, vol. 13, 3, 202-218.
[9] Ma, Ping. (2006) “Employment Situation and Enlightenment of Japanese University Students.” Journal of Social
Sciences in Shanxi Higher Education Institutions, 4, 124-126.
[10] Zou, Ping. (2002) “Changes in Japan’s Employment Situation and Government Countermeasures.” China Labor,
9, 56-58.
[11] Chu, J., Liu, J., & Liu, S. (2020) “Research and Strategy Analysis on the Employment Problems of College Students.”
Open Journal of Social Sciences, vol. 8, 129-138.
[12] Zhang, Quanzhou. (2023) “Generation Mechanism, Concrete Representation and Governance Strategies of ‘Lying
Downism’.” Theory Guide, 464, vol. 7, 74-80.
[13] Aoki, Reiko. (2004) Microeconomics of Declining Birthrate: Review of Existing Literature. Institute of Economic
Research, Hitotsubashi University.
[14] Fan, Huaqiang, Du Le, & Wen Keso. (2022) “Comparative Study of Employment Conditions of College Students in
China and Japan.” Chinese College Student Employment, 1, 24-29.
[15] Xu, Yanchen. (2018) “Japan’s Tripartite Interactive University Student Employment Model and Its Reference.”
Chinese College Student Employment, 20, 34-38.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231171
237
Research on Strategic and Financial Performance of Pop
Mart
Yaqi Guo1,a,*
1International College, Zhengzhou University, Zhengzhou City, China
a. yuanzhishui@hhu.edu.cn
*corresponding author
Abstract: At present, with the continuous growth of per capita disposable income, China’s
consumption mode has undergone a comprehensive upgrade, and consumers have begun to
pursue a more sophisticated lifestyle, and have begun to pay more attention to the spiritual
satisfaction brought by goods. With the development of pan-entertainment market and social
media, the trend culture has been widely popularized, and the trend toy market in China has
ushered in rapid development. More and more enterprises have joined the trend play industry,
and the trend play market is facing fierce competition. China’s per capita GDP continues to
soar, the consumption structure has changed, 90 and 00 have become the main consumer
groups in the consumer market. Trend toys began to rise, Pop Mart as a representative of the
trend toy industry, rose rapidly in a short period of time, and experienced two strategic
transformations, leading the development of the trend play industry. This paper mainly
analyzes the two strategic layout of Pop Mart in this era, that is, the centralized strategy and
the diversified business strategy, and analyzes the impact of its strategic transformation on its
financial performance from four aspects: debt paying ability, operating ability, growth ability
and profitability.
Keywords: trendy toys, Pop Mart, centralized strategy, diversified business strategy, financial
performance
1. Introduction
In the era of China’s economic rise, with the change in consumption structures, the younger
generation has become the main force in the consumer market. They are more enthusiastic about
online consumption and occupy a great online discourse power and traffic highland in the mobile
internet space. They have strong self-awareness, diverse values, rich interest circles, personalized
lifestyle propositions, strong patriotism, and a high acceptance of domestic brands. They are willing
to pay a premium for product design and features. The rise of trendy toys fully reflects this point. Pop
Mart, as a typical representative of the domestic trendy toy industry, can rapidly rise in a short time
and occupy a large market share. Its strategic development and layout have attracted widespread
attention.
leading new trends. Since 2018, it has begun integrated development and achieved unprecedented
results in its performance. So far, it has occupied half of China’s trendy toy industry and is the largest
and most profitable trendy toy enterprise in recent years in China. Pop Mart is a typical representative.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231058
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
238
Studying its two strategic transformations and financial performance will help provide a reference for
trendy toy companies and promote the development of the trendy toy industry.
2. Development Overview of Pop Mart
Beijing Pop Mart Cultural Creative Co., Ltd. was established in October 2010. Pop Mart initially
adopted a similar model of “buyer shop”, using its offline channel advantages to select a variety of
fashion products in the fashion market, and the best-selling IP was Molly. Most of the fashion
products are sold in the form of blind boxes, and the blind box vending machine has been developed
as a radar, by placing low-cost vending machines in major shopping malls to observe sales and predict
whether it is suitable to open a store.
In 2017, after Pop Mart had a certain amount of traffic and fixed fans, it began to hold an
international trend toy exhibition, enhance brand influence, deeply explore artists, and strive to
accumulate core high-quality IP, open up the “IP supply chain - channel” model, and conduct joint
internal incubation with multiple brands. The company has set up an internal creative design and
industrial development team of more than 100 people, and the internal artists are deeply involved in
the commercial design, development and processing of contracted IP, and carry out internal IP
creation while accumulating industry design experience in cooperation. In terms of external mining,
in order to ensure a steady stream of high-quality IP, the company has also set up an internal artist
mining team of more than 20 people, actively looking for different styles of artists with commercial
prospects around the world. In addition to promoting the trend culture, the International Fashion Toy
Exhibition is also one of the most important ways to discover the talents of the company, The
exhibition is held once a year in Beijing and Shanghai, inviting artists and fashion brands from all
over the world to participate in the exhibition, and artists need to prepare limited styles of the
exhibition in advance [1].
In addition, the company also cooperates with colleges and universities to jointly find local popular
game designers. The company’s deep cooperation with artists from Hong Kong, Japan and South
Korea and other places. In order to expand the territory for long-term development, Pop Mart cultural
sea, overseas to adopt a combination of online and offline modes, while building an offline experience
park [2].
3. Development Strategy
3.1. Initial Strategic Analysis (2010-2016)
From 2014 to 2016, Pop Mart’s strategic plan was to focus on trendy toys, sign a certain number of
designers, use offline experience stores as the main sales channel, and sell products in blind boxes.
The company aimed to open as many stores as possible to quickly seize the market and create a
domestic trend for trendy toys.
After receiving its first investment, Pop Mart continued to expand rapidly at a speed of 3-5 stores
per year. The signing of Molly brought huge revenue to Pop Mart, demonstrating the feasibility of
this strategy.
In 2015, there were already more than 20 stores across the country. At the same time, Pop Mart
continued to sign designers, introduce new products, buy out popular IP licenses, and produce a
variety of hand-painted figurines. With the help of Molly, Pop Mart attracted a large group of loyal
fans, and the introduction of new products also met the needs of more consumers, increasing product
differentiation to some extent and expanding both store and consumer scale.
Due to the “giant country effect” in the Chinese market, homogeneity and large-scale replication
of enterprises have broad market space. Pop Mart’s offline physical store model once mature can be
quickly copied in similar spaces across the country, which not only reduces the complexity of scaling
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231058
239
but also achieves scale effects, improves operational performance, and reduces cost structures. This
laid a solid foundation for future trendy toy operations.
Since 2014, Pop Mart has been influenced by sales models such as Japanese “fukutsu” and
“nugunade” (similar to blind boxes), creating a blind box sales model. In this model, consumers can
choose to purchase a series of products, but which specific product they will receive is not shown on
the packaging. Only when they open the package can they see it. This marketing strategy brings
tremendous surprises to consumers. This novel sales model has never appeared before in China and
has attracted many consumers since its launch. Pop Mart promotes small and low-priced blind boxes
as the main product first, attracting fans’ attention while expanding the fan base through the surprise
and social attributes of blind boxes. At the same time, Pop Mart continuously expands the number
and scale of IPs. Different IPs bring different emotional experiences to consumers through their
original design, achieving the bundling of consumers and Pop Mart IPs. When consumers are familiar
with IP images are combined with blind boxes, giving the original IP a sense of mystery and novelty,
providing consumers with a more satisfying and joyful shopping experience. Various IPs also
continuously launch new series of blind box products, continuing the emotional connection between
consumers and products and causing consumers to enthusiastically engage in purchasing behavior.
This provides a good foundation for further expansion and development.
3.2. Strategic Transformation (2017-2022)
3.2.1. Supply Analysis
It should Reduce costs and increase efficiency and open up the ‘IP-Supply Chain-Channel’ model.
By being the first to break through the “IP-Supply Chain-Channel” model, Pop Mart helped establish
scale advantages. The continuous demand for large orders provided a foundation for the company to
enhance its ability to manage factories. Based on the popularity of trendy toy exhibitions and online
pre-sale data, the company has strengthened its ability to predict new product sales momentum.
Pop Mart has strong anti-risk supply capabilities and based on its increasingly mature supply chain,
Pop Mart’s top IP launches 1-3 new products every quarter. During the COVID-19 pandemic in 2020,
the company had a small order volume, and with the market recovery after the epidemic, some of the
company’s products were out of stock. In late 2020, the company signed a supply chain information
technology company, Fule Technology, and with the company’s continuous optimization of the
supply chain, the production cycle for incremental orders has been shortened to about three months.
In May 2023, the company launched a total of 15 new blind boxes, reaching the highest level in nearly
a year. The increasingly perfect supply chain system provides a guarantee for the company’s new
product launch and sales.
3.2.2. Core Competition
The company’s rich IP reserves and mature commercial operation have built a good, closed loop. The
rich IP reserve is the cornerstone of IP operation. Similarly, benefiting from Pop Mart’s mature IP
commercial platform operation, Pop Mart has a head start in obtaining high-quality IP. Original IP
incubation takes a long time and has high risks. Designers or studios tend to hand over IP to
manufacturers who have matured and agentd IP. Designers are concerned about the commercial
failure of their creations and undermine their creative confidence. They are also concerned about the
excessive commercialization of IP and losing artistic integrity. The company’s “making stars” ability
has been tested by many blockbusters, and with ample funding, the company’s pursuit and respect
for art are also consistent with designers.
The company has rich IP reserves, far higher than its competitors in China. As of 2023H1, the
company operates a total of 93 IP characters, including 12 self-owned IPs purchased and internally
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231058
240
studio-hatched, represented by Molly, Yuki, etc.; 25 exclusive contracted external artists’ IPs,
represented by PUKIY, SATYR ROORY, etc.; and cooperate with well-known third-party IP
suppliers to re-create classic IP images, represented by Mickey, Little Yellow Man, Hello Kitty, etc.
According to the statistics of Forrester Sully, in 2019, the number of IPs operated by Pop Mart was
the same as that of the American company Hasbro, which had a century-old history, far ahead of
Japan’s Bandai and Europe’s LEGO.
3.2.3. Brand Strategy
It should deeply mine the artists and collaborate internally and externally to enhance brand influence.
As of H1 2020, the company has maintained close contact with over 350 artists worldwide and
developed deep collaborations with 25 of them through licensing or intellectual property transfer
agreements. Most of the collaborating artists originally worked in visual arts and design and received
awards and recognition around the world in various art fields, including painting, sculpture, crafts,
clothing design, and so on, and have a large fan base.
In terms of internal incubation, the company has formed a team of over 100 internal creative
designers and an industrial development team. Internal artists are deeply involved in the commercial
design and development of signed IPs, accumulating industry design experience through
collaboration while creating internal IPs. As of H1 2023, the internal creative team has jointly
incubated 9 original IPs, including Yuki, BUNNY, and Little Sweet Potato.
For external exploration, to ensure a continuous stream of high-quality IPs, the company has also
established a team of over 20 internal artist explorers to actively seek out artists with different styles
around the world. The international trend toy exhibition is not only a way to promote trend culture
but also one of the most important ways for the company to discover new talents. The exhibition is
held in Beijing and Shanghai each year, inviting artists and trend toy brands from around the world
to participate. At the same time, artists need to prepare limited-edition designs in advance. By
observing the foot traffic at the exhibition booth, sales at the exhibition, and feedback from the
exhibition, the company can anticipate the trends and preferences of players and discover new artists.
To promote the trend of Chinese culture, the company has teamed up with the Central Academy
of Fine Arts to hold a month-long lecture series on “trend, toy, and design”.
In terms of third-party cooperation, the company collaborates with Disney, Minions, and other IPs
that have a wide fan base. This may help broaden the user base. According to the Baidu Index, users
who searched for “Pop Mart” keywords in May were mainly female users, with 52% of users aged
between 20 and 29 years old. The third-party IPs have a higher proportion of users in other age groups
than the company itself. At the same time, IPs such as “Naruto” have more male players than female
players, accounting for over 60%. Cooperation with third-party IPs can help the company broaden
the age and gender brackets of its users. At the same time, the company will launch collaborations
between its own/exclusive IPs and third-party IPs to help convert players into “Pop Mart” brand fans.
3.2.4. Development Strategy
Similar to China’s operational strategy, Pop Mart has adopted a multi-channel operating model that
combines online, offline stores, and robot stores for its overseas expansion. The company has opened
its own online stores for global users and has also established official flagship stores on Amazon and
AliExpress. In terms of regional distribution, Pop Mart has dozens of offline stores in Japan and
France and has also opened 2-3 physical stores in other regions such as South Korea and Singapore.
In August 2021, Beijing Pop Mart Theme Park Management Co., Ltd. was established, with
business operations including urban park management, amusement parks, catering management,
sports brokerage services, performance venue operations, etc.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231058
241
In cooperation with Beijing Chaoyang Park, Chaoyang Park will authorize Pop Mart to use the
“European Style” project and its surrounding streets and forests in the park. Located on East Fourth
Ring Road in Beijing, it was founded in 1984, with a north-south length of 2.8 kilometers and an east-
west width of 1.5 kilometers. The planned total area is 288.7 hectares, of which 68.2 hectares is water
area and 87% is green space.
It is expected that the park will be completed and opened in September 2023. The reference model
may be more like the Sanrio theme park in Japan, with IP image as the theme of scene experience,
combined with tickets, catering, and peripheral derivative to realize IP fans’ conversion, which is
helpful to maintain users’ enthusiasm and stickiness for IP. For Pop Mart, which has reached a certain
scale of IPs and continues to improve its commercialization capabilities, it will be a cross-field
attempt. The value of the IP hatch and output has been verified by Disney’s business model.
4. Financial Performance Analysis
4.1. Operational Capability Analysis
Overall, Pop Mart’s operational capability can be divided into two stages. The first stage was from
2016 to 2018. During this period, Pop Mart experienced its second strategic planning, expanding its
scale, signing many IPs, forming core advantages [3], and carrying out cultural exports. It can be seen
that after this change, Pop Mart’s operational capability continued to improve year by year, with the
total asset turnover rate increasing from 1.13 in 2016 to 2.28 in 2018; the inventory turnover rate
showed the same changing pattern during this stage; the accounts receivable turnover slightly
decreased in 2017 and began to rise after 2018, indicating that Pop Mart’s capital operation capability
continued to strengthen during this stage, and it had better control over the use of capital. The second
stage was from 2019 to 2022, which was greatly affected by the epidemic. Based on a diversified
strategy, a small adjustment was made to slow down the expansion of store openings[4]. The
operational capability showed a downward trend, as shown in Figure 1.
Figure 1: Operational capacity indicators of Pop Mart from 2016 to 2022.
The total asset turnover rate decreased from 1.58 in 2019 to 0.52 in 2022, which was basically in
line with the industry average; the accounts receivable turnover rate showed the same downward
trend in this interval, but compared with the total asset turnover rate and inventory turnover rate, the
decrease was not significant, indicating that Pop Mart maintained strong bargaining power with
upstream and downstream partners; the inventory turnover rate decreased to 2.27 in 2022, mainly due
to the impact of the epidemic, some offline stores were unable to operate normally, inventory
0
2
4
6
8
10
12
14
2016 2017 2018 2019 2020 2021 2022
Inventory turnover Accounts receivable turnover rate Turnover of total assets
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231058
242
management and delivery became more difficult, and some inventory accumulated. The company
began to conduct promotional activities to adjust [5].
4.2. Analysis of Solvency
During the period from 2014 to 2016, Pop Mart was in its early stages of entrepreneurship and
implemented a centralized strategy, expanding its scale. The trend of pop culture and the emergence
of the toy industry provided favorable conditions for Pop Mart as a well-known offline experience
store. The current ratio, quick ratio and cash ratio were all in an upward trend, and the asset-liability
ratio and equity multiplier were relatively low, indicating that the enterprise had sufficient liquidity
and strong solvency, and the property rights ratio was low, indicating that Pop Mart was adopting a
low-risk, low-return financial structure at this time, which was in a stable expansion stage [6].
From 2016 to 2019, Pop Mart implemented a diversified competitive strategy. The current ratio,
quick ratio and cash ratio showed significant fluctuations, with a sharp decline in 2016. The asset-
liability ratio, equity multiplier and equity ratio all showed a significant increase. This was mainly
due to the large amount of capital needed, such as signing a large number of IPs and building
experience parks, with high costs that move some liquidity, resulting in lower solvency and higher
financial risks for the enterprise.[7] After three years of development and precipitation, Pop Mart
experienced an upturn in 2019 and reached its highest value in 2020, with the current ratio, quick
ratio and cash ratio approaching 10, and the asset-liability ratio, equity ratio and equity multiplier
decreasing. The solvency of the enterprise was strong. However, due to the impact of the epidemic,
the solvency of the enterprise began to decline in 2020. Although it decreased to 4-6 in 2022, it
remained higher than the values from 2017 to 2019 and showed a stable trend [8]. This indicated that
the implementation of strategic transformation brought excellent development prospects to the
company, improved solvency while increasing operating efficiency, and buffered the impact of the
epidemic, as shown in Figure 2 and Figure 3.
Figure 2: Short-term debt-paying ability indicators of Pop Mart from 2014 to 2022.
0
1
2
3
4
5
6
7
8
9
10
2014 2015 2016 2017 2018 2019 2020 2021 2022
Current ratio Quick ratio Cash ratio
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231058
243
Figure 3: Long-term debt-paying ability indicators of Pop Mart from 2014 to 2022.
4.3. Growth Capability Analysis
Pop Mart was loss-making from 2015 to 2016. However, after implementing a diversified business
strategy and an integrated industry chain business model based on blind boxes, Pop Mart achieved a
leap forward in net profits and rapid growth. The net profit growth rate in 2018 was as high as
6,238.85%, and the net profit growth rate steadily declined from 2018 to 2020 [9]. According to the
financial report, as shown in Figure 4. Pop Mart believed that the reason for the decline in net profit
growth rate was that the product process and production quality level had been further improved,
increasing product costs. In summary, although Pop Mart’s total asset growth rate and net asset
growth rate showed a year-on-year increase trend after implementing diversified business strategies
and a full industry chain layout in 2017, the revenue growth rate exceeded 200% from 2018 to 2019,
and the net profit growth rate boomed in 2018. However, due to the impact of the epidemic, the
revenue growth rate sharply declined in 2020, and the net profit growth rate slowed down in 2019
and only grew at a rate of 16.05% in 2020. From 2020 to 2022, Pop Mart remained stable and
continued to diversify its strategies, expand its business scope, and buffer the impact of the epidemic.
Figure 4: Development capacity indicators of Pop Mart from 2015 to 2022.
0.00%
20.00%
40.00%
60.00%
80.00%
100.00%
120.00%
140.00%
160.00%
180.00%
200.00%
2014年2015年2016年2017年2018年2019年2020年2021年2022年
Asset-liability ratio Equity multiplier Equity ratio
-1000.00%
0.00%
1000.00%
2000.00%
3000.00%
4000.00%
5000.00%
6000.00%
7000.00%
2015 2016 2017 2018 2019 2020 2021 2022
Growth rate of main business income Net profit growth rate Growth rate of net assets
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231058
244
4.4. Profitability Analysis
In 2016, both the sales gross profit margin and net asset yield were negative. However, after
implementing a diversified business strategy, the sales gross profit margin showed a rapid and stable
growth trend. This was mainly due to the emerging trend of the toy industry, although performance
showed a growth trend, the overall toy industry had not fully emerged, and the audience was limited.
The cost of signing IPs such as Molly was high [10]. However, after implementing diversified
business strategies in 2016, Pop Mart further broadened the audience and toys truly entered the public
view. At this time, Pop Mart, as the industry leader, and the novel sales model of blind boxes brought
tremendous profits. Since 2017, Pop Mart’s sales net margin, sales gross profit margin, net asset yield,
and total asset yield have been on the rise. According to Pop Mart’s 2020 financial report, Pop Mart’s
independent product gross profit margin decreased from 71.2% in 2019 to 68.7% in 2020, mainly due
to the improvement of product process quality, more complex product processes, and rising raw
material costs resulting in a decrease in gross profit margin. Since then, Pop Mart has been in stable
profitability from 2020 to 2022. However, optimizing the cost-income structure is essential for the
future development of the enterprise. Based on strong profitability in the industry, Pop Mart can bring
scale effects through diversified operations and full industry chain coverage business models,
effectively improve operating efficiency, improve cost structure, and enhance the company’s ability
to control costs and expenses.
5. Conclusion
Pop Mart has undergone two strategic transformations. The first was from 2014 to 2016, when Pop
Mart stopped selling other daily and miscellaneous products and began to mainly produce IP blind
box products with Molly as the focus, implementing a concentration strategy to attract a large number
of customers. Although the company was still at a loss, its performance showed a small upward trend
and its debt-paying ability was relatively strong. From 2016 to 2022, Pop Mart implemented
diversified business strategies, no longer satisfied with existing products. The trend of toys became
popular in the public eye. Pop Mart used its diversified business strategies to stand out from the fierce
industry competition and gain an advantageous position, becoming the pioneer and leader of the toy
market. Its financial situation improved significantly compared to before the implementation of
diversified strategies. Pop Mart developed corresponding strategies based on different environments,
coupled with its novel marketing model, allowing it to expand rapidly in a short period. However,
Pop Mart still needs to consider its long-term development issues due to the lack of background
stories for its IPs, which may result in insufficient user stickiness [11].
References
[1] Hu Maoli. Review of Enterprise diversification Strategy research [J]. Finance and Accounting
Bulletin,2011(36):86-88.
[2] Yang Xingquan, Li Wencong, Yin Xingqiang.Research on the “dual” impact of diversification on enterprise
innovation [J]. Journal of Finance and Economics,2019,45(08):58-71.
[3] Gao Hui, Lin Huang.Research on the impact of diversification strategy on performance of real estate listed
companies [J]. Modern Management Science,2021(02):79-86.
[4] Han Yu. Financial Risk Analysis of Enterprise diversification -- A case study of letV [J].National Circulation
Economy,2019(25):50-53.
[5] Hong Daolin, Xiong Dehua.Diversification and corporate Performance analysis of Chinese Listed companies: An
Investigation based on Endogeneity [J].Financial Research,2006(11):33-43.
[6] Hu Xin. Research on Diversification Strategy and its impact on Financial Performance -- Taking Shennan Stock as
an example [J].Modern Audit and Accounting,2020(07):16-18.
[7] Yang Jun, Zhao Jixin, Li Yuhang.Research on the impact of diversified business Strategy on corporate financial risk
[J]. Journal of Finance and Accounting,2020(14):78-81.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231058
245
[8] Yang Luping. Analysis on the impact of enterprise diversification on performance: A case study of tobacco
enterprises [J].Modern Business,2020(21):103-104.
[9] Yao Jun, Lv Yuan, LAN Hailin.An empirical study on the relationship between diversification and economic
performance of listed companies in China [J]. Management World,2004(11):119-125+135.
[10] Zhu J. Diversification Strategy and business performance of Chinese listed companies [J]. Economic
Research,1999(11):54-61. (in Chinese)
[11] Du Qian. Research on financial Crisis based on letV’s diversification Strategy [D]. Anhui University of
Technology,2018.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231058
246
Strategies and Practices for Dealing with Inflation in New
York
Ying Wang1,a,*
1School of Business, Stevens Institute of Technology, Hobokken, NJ,07030, USA
a. ywang407@stevens.edu
*corresponding author
Abstract: New York City has experienced severe inflation in the aftermath of the COVID-19
pandemic, with significant negative impacts on the lives of its citizens. This study explores
inflation in New York City, its causes, effects, and potential remedies. By analyzing data such
as the CPI index and cases from relevant U.S. government websites, this paper explored
inflation trends and their impact on the cost of living and socioeconomic factors. It is worth
noting that inflation leads to an increase in the cost of living, a decrease in the purchasing
power of money, and an increase in crime rates. The author also draws on successful anti-
inflation strategies elsewhere to propose solutions for New York. Provides valuable reference
for New York policymakers. The research in this article has important practical significance
for the country to address inflation issues and grasp the direction of macroeconomic
development.
Keywords: New York, inflation, CPI, crime rate, socioeconomic factors
1. Introduction
Studying and addressing the effects of inflation in New York are manifold and far-reaching. First of
all, as a global financial and business center, New York's inflation may have a huge impact on the
U.S. and global economies, triggering social instability such as wealth inequality and rising
unemployment. Therefore, through in-depth research on inflation in New York, policymakers can
more accurately formulate corresponding economic and social policies to slow down or avoid its
negative impacts. By scientifically and comprehensively studying inflation in New York and
proposing targeted solutions, we can provide valuable reference for inflation problems in other
regions and even the world. Additionally, for New Yorkers, understanding this phenomenon can help
them plan their personal finances and lives more effectively. Finally, this research will also provide
interdisciplinary research materials and theoretical support for economics, sociology, political science
and other disciplines.
According to a May 2023 study based on a simple quantitative New Keynesian model, the leading
cause of inflation in New York in recent years is mainly attributed to two key factors: the impact of
oil prices and loose monetary policy [1]. These two not only directly push up the prices of goods and
services, thereby increasing the cost of living, but also indirectly affecting consumption and
investment behavior. Another Research took advantage of the very close direct correlation between
the price of coffee (a commodity) and the Consumer Price Index (CPI) in New York (Pearson's
correlation coefficient was 0.61), indicating that changes in commodities prices are reflected in New
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231172
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
247
York's Consumer Price Index (CPI)—an effective indicator of inflationary [2]. The conflict between
Russia and Ukraine and the Covid-19 pandemic has fueled the worst inflation in nearly four decades
in the U.S. and other developed countries, pointing to misallocation of fiscal relief funds and rising
fiscal deficits as the main factors behind the rise in inflation, which rose by as much as 7.7 percentage
points [3]. Moreover, the data from the World Bank confirms this correlation through Engel Granger's
cointegration analysis. The results suggest that investing mainly in renewable energy technologies
may be able to deal with New York’s severe inflation problem to a certain extent [4].
This study will illustrate the basic situation of inflation in New York from more aspects, such as
official data from the US government website to explain the basic situation of inflation in New York,
and then use cases to describe the negative impact of inflation in New York, and then propose
relatively feasible solutions based on the problems and reference materials. To a certain extent, it will
help New York policymakers solve the current inflation problem.
2. Data Illustration
Figure 1 is the region's Consumer Price Index (CPI-U) increased 0.4% in July 2023, according to the
U.S. Bureau of Labor Statistics, and has increased 3.2% overall over the past 12 months [5]. The
growth has been driven largely by rising housing prices, but there has also been an increase in food
prices, particularly the cost of eating out. In relative terms, energy prices fell by 11.5%, especially
gasoline prices, which dropped by 21.7%, but electricity costs increased by 8.1%. Prices for other
consumer goods such as new and used cars and clothing also rose. Taken together, the data paint a
complex and diverse picture of inflation, involving price changes in housing, food, energy and other
daily consumer goods, providing a reference for us to study inflation in the New York area.
Figure 1: Over the year percent change in CPI_U, New York -Newark Jersey City; NY-NJ-PA, July
2020-July 2023.
Figure 2 shows that the PPI index in New York indicates that current inflation levels are relatively
high, especially after COVID-19 restrictions are lifted in 2022 [6]. During this period, the PPI for
food, energy and social services continued to be high, implying that inflationary pressures remained
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231172
248
high. Although the government took a series of monetary and fiscal measures to ease inflation in 2023,
the results were not obvious, and the inflation level still had not returned to the state before the
outbreak.
Figure 2: PPI for Final demand, 1-month percent change, seasonally adjusted from 2020-2023.
From a macro perspective (see figure 3), the overall U.S. MPI (Material Price Index) can show the
overall current inflation situation in the whole United States [7]. In November 2022, the MPI
increased by 0.6%. This is the first increase since early October 2022. Six of the ten subcomponents
rose. Despite last week's price gains, the MPI remains 35% below its all-time high set in early March.
However, commodity prices as measured by the MPI are still 29% higher than pre-pandemic levels
in January 2020. This situation fully demonstrates that the overall inflation situation in the United
States remains severe.
Figure 3: Materials Price Index from 2014-2023 from S&P Market Global market intelligence.
In addition to some basic data that can reflect the current status of inflation in New York, some
daily life cases can also reflect the current status of inflation in New York.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231172
249
3. Case Description
Since the end of the COVID-19 pandemic, inflation has become a growing problem in New York
City, especially among New Yorkers, whose daily lives have been severely affected. As can be seen
from the following two cases, the impact of inflation has penetrated deeply into the daily lives of New
York citizens.
3.1. Residents Under Multiple Pressures of Highest Inflation in Four Decades
In a news update on July 13, 2022, CBS New York (CBS2) revealed the impact of US inflation
reaching a 40-year high, especially on New York City residents [8]. According to the U.S. Department
of Labor, consumer prices rose 9.1% year-over-year, the fastest increase since 1981. This upsurge
raises many socioeconomic questions: how supermarkets respond to rising costs, how residents make
ends meet under the pressure of soaring rents and food prices, and how these pressures exacerbate
social inequality. Several New York residents and experts were quoted in the news, including West
Side residents Kevin Salazar and Ana Gonzalez, as well as New York City broker Bill Kovalchuk
and financial policy expert Michelle Bang Germany, they analyzed this issue from daily life and
professional perspectives respectively. The nonprofit City Harvest reported a 36% increase in food-
insecure residents than before the outbreak. This series of changes and shocks provides a
comprehensive but condensed overview of the inflation situation.
3.2. Inflation and Office Vacancies in New York
Another report in June 2023 stated that New York City’s inflation rate was the highest in the past 40
years, reaching 6.0% [9]. These problems put enormous pressure on individual consumers, businesses
and even the entire urban economy. Inflation in particular, with food prices up 8.4% over the past 12
months, energy prices up 4.4%, and so-called core inflation (which excludes food and energy) also
rose 5.7%. That leaves nonprofits like Citymeals on Wheels facing a 33% increase in food costs.
Declining office occupancy also results in about $12 billion in lost economic activity annually,
including reduced spending at restaurants, transportation and other areas. City Comptroller Brad
Lander's latest report also shows that the purchasing power of the current $15 minimum wage will
drop to just $13 due to the effects of inflation.
4. Analysis on the Problems
4.1. Reasons of Inflation Happens in New York
4.1.1. Wars and Regional Conflicts
The conflict between Russia and Ukraine has been going on for more than 500 days. In addition to
causing widespread damage to warzone economies, it also hurt the economies of other countries, the
United States being one of them. Judging from New York's current high inflation rate and the trend
of sharply rising energy prices, this conflict is one of the main factors leading to New York's recent
inflation. As a major global energy exporter, Russia's actions in its war with Ukraine could destabilize
energy supplies, especially given that Ukraine is one of the key corridors for natural gas transportation.
This instability has the potential to drive up global energy prices, further fueling inflationary pressures
in New York. Geopolitical uncertainty can also make Wall Street investors worried about the global
economic outlook, leading to financial market volatility, currency depreciation and rising inflation.
As the war expands, disruptions in supply chains or the availability of certain key commodities, such
as food, metals or French fries, could cause prices for those goods to spike, pushing up inflation in
New York. The U.S. government's economic and monetary policy responses, such as whether to raise
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231172
250
interest rates or implement fiscal stimulus, also affect the final level of inflation. With the war still
ongoing and the future of Russia and Ukraine difficult to accurately predict, a combination of these
factors is likely to affect inflation in New York to varying degrees.
4.1.2. Global Pandemic Issues
Although New York City lifted the restrictions on the new crown epidemic in the second half of 2022,
the COVID-19 epidemic has had a multifaceted impact on New York City's economy in the past two
years, which has led to an exacerbation of the recent inflation problem. During the outbreak, many
businesses had to temporarily close or reduce production, disrupting their supply chains and creating
severe shortages of goods and services, driving up prices. During the Pandemic, many local workers
in New York were sick or needed to be quarantined, resulting in a severe labor shortage. Such a
situation results in companies having to raise wages for active workers and increase business costs,
and these costs are ultimately passed on to consumers through increased prices of goods and services.
The New York City government has also increased expenditures in response to the epidemic, such as
issuing relief funds and purchasing and importing masks, protective clothing and medical equipment.
These additional financial expenditures may increase the supply of US dollars, thereby triggering
inflation. COVID-19 has also changed people's consumption habits. New Yorkers' increased demand
for medical supplies, toilet paper, and food may also lead to higher prices. To sum up the overall
situation, the central bank may adopt monetary easing policies, such as lowering interest rates or
increasing money supply, to stimulate economic recovery, but this may also increase inflation. In the
past decade, due to the high degree of financialization and the hollowing out of the real economy, as
well as the high degree of globalization of the supply chain, it has been difficult for the United States
to experience serious inflation. However, the COVID-19 crisis and the out-of-control epidemic have
directly led to the collapse of the overall supply chain in the United States. For example, key logistics
personnel such as port dock workers and long-distance transport drivers are often short of or go on
strike due to the epidemic. All this has become a driving force for New York's recent key factor in
inflation.
4.1.3. Manufacturing Outflow
New York's manufacturing outflow is very serious. Most of the low-end goods in New York are
mainly imported from China, Mexico and other places. Manufacturing outflows, also known as
“deindustrialization,” may have been intended to reduce production costs, but when such capital
outflows affect key industries or necessities, supply chains can become destabilized, driving up
commodity prices. At the same time, as the epidemic sweeps the world, product manufacturing
countries such as China and Mexico are also facing serious supply chain and transportation collapses.
Manufacturing outflows could also exacerbate the U.S. trade deficit and put downward pressure on
the dollar. Due to the high cost of domestic labor in the United States, a large number of cheap goods
are dependent on imports. The rising cost of imported products will directly lead to increased inflation
in New York. Due to the long-term outflow of manufacturing and the impact of the epidemic, rising
unemployment in New York City will reduce the purchasing power of consumers. While the
government may counteract this effect through fiscal and monetary policies such as increasing public
spending or lowering interest rates, it is also highly likely to stoke inflation. Capital may flow from
the real economy to the financial market, triggering asset bubbles and economic instability. The
outflow of manufacturing will not only affect New York, but also the overall U.S. supply chain, while
the COVID-19 epidemic and any instability in the global economy may in turn affect inflation in New
York.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231172
251
4.2. Analysis of the Problems Caused by Inflation in New York
4.2.1. Increase in Consumer Goods Prices
As noted in the first case, New York residents are experiencing the highest increase in consumer
prices since 1981, rising 9.1% year-over-year. This not only sharply increases the cost of daily life,
but also further exacerbates the problem of social inequality between rich and poor. For example,
many people are finding it increasingly difficult to afford basic living expenses as food and rent prices
soar. The report of the non-profit organization "City Harvest" further shows that due to the increase
in food prices caused by inflation, residents buy more cheap food, and the proportion of residents
facing food insecurity has increased by 36% compared with before the outbreak.
4.2.2. Office Spaces Vacancies
High inflation has directly led to higher rents for shops and offices in New York City, which has
directly led to a sharp decrease in the occupancy rate of office space in New York City, resulting in
an annual loss of about $12 billion in economic activity, including restaurants, transportation and
other areas. expenditure.
4.2.3. Decreased Purchasing Power of Money
High inflation puts enormous pressure on the entire urban economy. In the above cases, City
Comptroller Brad Lander's report shows that the purchasing power of the current $15 minimum wage
will drop to just $13 due to inflation.
4.2.4. Increase Crime Rate
High inflation will lead to greater financial pressure on relatively poor families and individuals in
New York. When they struggle to maintain basic living through legitimate means, some people may
turn to illegal or immoral activities to obtain income, such as robbery, drug dealing, theft, marches,
and demonstrations.
5. Solutions on the Problem
5.1. Government Regulation Solves the Increase of Consumer Goods Prices
The inflation problem New York currently faces is similar to what China experienced in the late
1940s: consumer goods prices’ rose sharply and the real purchasing power of the currency weakened.
In Zhang, Shan Xi's research article, he describes in detail how China's new government quickly
responded to this economic challenge in the 1950s [10]. At that time, China's consumer goods prices
soared to unprecedented heights. In order to stabilize the economy, the new government decided to
take a series of measures: first, it concentrated most of the national revenue to the central government
to ensure that financial resources were used to the maximum extent; second, important materials were
concentrated and targeted to those in shortage. regions; in addition, the new government has also
significantly reduced unnecessary spending, cleaned up obsolete warehouse inventories, strengthened
tax policies, issued public bonds to raise more funds, and emphasized the importance of savings in
government spending. This series of measures significantly stabilized and lowered prices in just one
year. Compared with New York, these suggestions are instructive: The New York City government
can consider more centralized management and distribution of critical supplies to ensure that they are
used rationally and effectively where they are most needed to slow down price increases caused by
shortages; merchants should be Encourage the liquidation of obsolete or low-selling inventories to
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231172
252
release more liquidity; the government can increase fiscal revenue by raising taxes or issuing public
bonds; finally, considering the current high price environment, the government should manage and
economize expenditures more carefully in order to Ensure adequate resources are available to deal
with economic problems and stabilize consumer goods prices.
5.2. Taking Roosevelt’s New Deal as a Reference
For the situation of office jobs and store vacancies caused by inflation, there was a reference solution
during Roosevelt’s New Deal in 1933. In the article State Capacity and Economic Intervention in the
Early New Deal, Theda Skocpol and Kenneth Finegold talked about how the 1933 U.S. Roosevelt's
New Deal addressed such problems during the Great Economic Crisis in New York [11]. Roosevelt's
New Deal took a variety of measures to address the problem of store closings. One of the key policies
is the establishment of the National Recovery Administration (NRA), which aims to restore store
operations by establishing industry regulations for various industries, including wages and labor
conditions. In addition, the NRA also encourages businesses to comply with minimum wage
standards to increase hiring rates. Another key measure was the establishment of the Public Works
Administration (PWA), which funded and managed various infrastructure projects such as roads,
bridges, and school construction. These projects provide more business opportunities for stores and
create more jobs. Roosevelt's New Deal also implemented the Social Security system, which helped
reduce the financial pressure on the elderly and unemployed, thereby reducing the impact of economic
instability on stores. Roosevelt's New Deal in 1933 helped stores overcome the difficulties during the
Great Depression through a series of policies and measures. Corresponding to today's inflation issue
in New York, it is also of great reference value. First of all, the New York government can consider
formulating or adjusting industry regulations, funding infrastructure projects, and implementing new
safety measures to help current New York shops and offices survive inflation. Period, maintain
economic stability and sustained growth
5.3. Monetary Policy Solve the Problem of Low Purchasing power
After Japan's economic bubble burst in 1990, the country suffered a sharp decline in the purchasing
power of its currency. Krugman's article delves into Japan's economic stagnation and liquidity trap at
the time [12]. Although Japan suffered primarily from the effects of deflation, its currency's low
purchasing power was very similar to New York's today, so New York could learn from Japan's
strategy at the time. The Bank of Japan has adopted a zero or ultra-low interest rate policy and
expanded the money supply by purchasing government bonds and other assets. In addition, the
government has promoted a large amount of fiscal stimulus, such as infrastructure investment, tax
breaks and subsidies. In order to achieve long-term economic growth, Japan has carried out structural
reforms in many areas, also adopted an export-oriented growth strategy, and signed free trade and
investment agreements with many countries. In order to break the long-term deflationary mentality,
the Bank of Japan set an inflation target of 2% and purchased high-risk assets such as stock exchange
funds. These measures helped restore Japan's currency purchasing power to a large extent. Today,
New York can learn from Japan’s post-1990 economic bubble strategy. Specific recommendations
include: adopting loose monetary policies and lowering interest rates through the Federal Reserve
Bank to stimulate investment and consumption; promoting large-scale public investment such as
infrastructure construction to promote employment and economic growth; implementing in areas
such as the labor market, agriculture and corporate governance Structural reforms; encourage New
York businesses to expand exports to enhance their global competitiveness; set clear inflation targets
to manage public expectations; and support financial institutions to purchase riskier assets to further
stimulate the economy. Although the situation between New York and Japan is still different today,
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231172
253
these measures will help New York under inflation solve its low purchasing power problem to a
certain extent.
5.4. Lower the Crime Rates During Inflation Requires Combination Strategies
Gary Becker detailed the relationship between inflation and crime in his 1968 article, and Robert
Merton detailed the effects of inflation in his research and discussed how social inequality and
educational inequality affect crime rates [13,14]. It can be concluded from the history of crime control
in the United States that addressing high crime rates during periods of inflation often requires a
combination of strategies. From an economic policy perspective, controlling inflation is key, and is
usually achieved through monetary policy and fiscal policy. Lowering inflation not only relieves the
economic stress associated with it but can further reduce crime rates. In addition, since high
unemployment is often associated with high crime rates, creating more job opportunities is also an
effective means to alleviate this problem. From a social policy perspective, education and training are
particularly important. By providing more and better educational and job training opportunities to
low-income and high-risk groups, such as educating citizens about the current drug abuse landscape
in New York, the probability of crime related to drug abuse can be fundamentally reduced. At the
same time, it is also necessary to strengthen the construction of the social welfare system. This
includes providing food stamps, unemployment benefits, and housing subsidies to relieve economic
stress and further reduce crime caused by New York's economic distress. Overall, these economic
and social policies need to work together to form a comprehensive strategy to effectively address high
crime rates in times of inflation.
In Table 1, NYPD data are also shown in the inflationary era [15]. While crime numbers and rates
remain high, extremely violent crimes such as murder, rape, and felonious assault have declined. The
reports issued by these government departments and independent institutions can clearly show the
relationship between inflation and crime rates, helping policymakers adjust their methods of
controlling crime problems according to the current inflation situation to strive for maximum social
stability.
Table 1 Index Crime Statistics in May 2023 from NYPD website.
May 2023
May 2022
+/-
% change
Murder
32
48
-16
-33.3%
Rape
124
157
-33
-21.0%
Robbery
1351
1520
-169
-11.1%
Felony Assault
2350
2384
-34
1.4%
Burglary
1127
1278
-151
-11.8%
Grand Larceny
4257
4183
74
-1.8%
Grand Larceny Auto
1369
1033
336
32.5%
Total
10610
10603
7
0.1%
6. Conclusion
After the epidemic is over, New York faces a serious inflation problem, which not only reduces
citizens' living standards but also aggravates the crime rate. The goal of this study is to provide New
York policymakers with practical recommendations for addressing this complex issue. We first
conducted preliminary data analysis using professional economic indicators such as CPI and PPI to
fully understand the inflation situation in New York. Then, through practical cases, it further
highlights how inflation penetrates the daily lives of New York citizens. Putting this information
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231172
254
together it can be concluded that inflation not only causes the cost of living to soar, but also closes
shops, reduces the purchasing power of money, and increases crime rates. The study recommends
that New York policymakers should prioritize several aspects: first, control inflation through fiscal
and monetary policies; second, create more job opportunities to alleviate the link between high
unemployment and high crime rates; third, strengthen The social welfare system, including the
provision of food stamps, unemployment benefits, and housing subsidies, these suggestions can help
New York slow the impact of inflation to a certain extent and enable citizens to better cope with the
rising cost of living. However, the study has its limitations because it focused primarily on New York
City rather than the United States. Future research will need to explore solutions to the inflation
problem at a broader geographical and socioeconomic level.
References
[1] Luca Gagliardone, (2023). Oil Prices, Monetary Policy and Inflation Surges, Working paper, Retrieved from:
chrome-
extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.nber.org/system/files/working_papers/w31263/w3126
3.pdf.
[2] Vochozka, M., Janek, S., & Rowland, Z. (2023). Coffee as an Identifier of Inflation in Selected US Agglomerations.
Forecasting, 5(1), 153–169.
[3] Cline, William. (2023). Fighting the Pandemic Inflation Surge of 2021-2022, Economics International Inc., Working
Paper, Retrieved from https://ssrn.com/abstract=4408811
[4] Dinçer, H., Yüksel, S., Çağlayan, Ç., Yavuz, D., & Kararoğlu, D. (2023). Can Renewable Energy Investments Be a
Solution to the Energy-Sourced High Inflation Problem?. In U. Akkucuk (Ed.), Managing Inflation and Supply
Chain Disruptions in the Global Economy, 220-238
[5] Bureau of Labor Statistics. (2023). Consumer Price Index - New York Region, Retrieved from
https://www.bls.gov/regions/northeast/news-release/consumerpriceindex_newyork.htm, last accessed 2023/9/9.
[6] US Bureau of Labor Statistics. (2023). PPI for Final Demand, 1-month percent change, seasonally adjusted - New
York Region. Retrieved from https://www.bls.gov/charts/producer-price-index/final-demand-1-month-percent-
change.htm
[7] Michael, Dell, Weekly Pricing Pulse: Commidity price growth returns, Nov 17th, 2022, Retrieved from
https://www.spglobal.com/marketintelligence/en/mi/research-analysis/weekly-pricing-pulse-commodity-price-
growth-returns.html
[8] Rozner, L., & Brennan, D. (2022). New York City inflation hits new 40-year high. CBS New York. Retrieved from
https://www.cbsnews.com/newyork/news/new-york-city-inflation-new-high/
[9] David, G. (2023). New York inflation and office occupancy. The City. Retrieved from
https://www.thecity.nyc/economy/2023/2/14/23600299/new-york-inflation-office-occupancy
[10] Zhang, S.X, (1997) Inflation through the ages and its Management Strategies. Chinese Culture Forum,1997(1), 50-
54.
[11] Skocpol, T., & Finegold, K. (1982). State Capacity and Economic Intervention in the Early New Deal. Political
Science Quarterly, 97(2), 255–278.
[12] Krugman, P. R., Dominquez, K. M., & Rogoff, K. (1998). It’s Baaack: Japan’s Slump and the Return of the Liquidity
Trap. Brookings Papers on Economic Activity, 1998(2), 137–205.
[13] Gary, S Becker. (2023). Crime and Punishment: An Economic Approach, The University of Chicago Press Journals,
76(2)
[14] Merton, R. K. (1938). Social Structure and Anomie. American Sociological Review, 3(5), 672–682.
[15] New York City Police Department , NYPD Announces Citywide Crime Statistics for May 2023, Retrieved on
https://www.nyc.gov/site/nypd/news/p00083/nypd-citywide-crime-statistics-may-2023
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231172
255
The Impact of Macroeconomic Factors on Non-
Performing Loans of Commercial Banks: An Empirical
Study Based on 2022 Provincial-level Data in China
Yixiang Li1,a, Diyi Gao2,b, Shu Fang3,c, Xuesong Chen4,d, and Dan Wan5,e,*
1Beijing Information Science and Technology University, Beijing, 100192, China
2Zhengzhou Foreign Languages School, Zhengzhou, 450001, China
3Beijing University of Technology, Beijing, 100124, China
4University of Science and Technology Beijing, Beijing, 100083, China
5Nanchang University, Nanchang, 330031, China
a. lyx010618@163.com, b. diyi_gao@163.com, c. fangshu@bjut.edu.cn,
d. chenxs@ustb.edu.cn, e. wandan@ncu.edu.cn
*corresponding author
Abstract: The level of macroeconomic development varies, resulting in different non-
performing loan ratios for commercial banks. Based on the relevant data of macroeconomic
indicators from 31 provinces and cities in China in 2022, this paper empirically studies the
relationship between regional gross domestic product (GDP), average urban employment
salaries, the proportion of the tertiary industry, per capita consumer expenditure, the number
of recipients of unemployment insurance, and the number of recipients of old-age insurance,
and the non-performing loan ratio using a multiple regression model. A general model is
established and tested to determine the final model. The empirical research results indicate
that the proportion of the tertiary industry and the number of recipients of unemployment
insurance have a positive impact on non-performing loans, while average urban employment
salaries, regional GDP, and the non-performing loan ratio have a negative impact. Finally,
based on the conclusions drawn from this study, relevant policy recommendations are
proposed to provide guidance for commercial banks in China to mitigate the risks of non-
performing loans.
Keywords: commercial banks, non-performing loan ratio, empirical research
1. Introduction
In recent years, with the increasing instability of the global economy, the scale of non-performing
loans in commercial banks has received more and more attention. On March 10, 2023, Silicon Valley
Bank, which had been named the best bank in the United States by Forbes for five consecutive years,
declared bankruptcy, becoming the second-largest bank bankruptcy event in U.S. history. This event
triggered a series of market chain reactions and led to a high level of concern from domestic and
foreign banks regarding non-performing loans in commercial banks. Non-performing loans in banks
are an important indicator for measuring the operating quality of commercial banks and the stability
of a country’s financial system. They reflect the efficiency of credit fund allocation in commercial
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231198
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
256
banks, and a high level of non-performing loans will increase the risks of bank crises and financial
crises. Macroeconomic factors are one of the important factors affecting non-performing loans in
commercial banks. If a region is in good development, it indicates that residents or enterprises have
strong debt repayment capabilities, and the possibility of non-performing loans will decrease. In
China, according to data from the China Banking and Insurance Regulatory Commission, by the end
of 2022, the balance of non-performing loans in the banking industry’s financial institutions reached
3.8 trillion yuan, an increase of 169.9 billion yuan from the beginning of the year, and the non-
performing loan ratio was 1.71%, which exceeded expectations in previous years. This indicates that
there are risks in the credit asset quality of Chinese commercial banks, which poses hidden dangers
to the smooth operation of the Chinese economy in the future. Therefore, based on provincial-level
data in China in 2022, this paper conducts an empirical study to explore the impact of macroeconomic
factors on non-performing loans in commercial banks and provides suggestions on how to prevent
and respond to the risks brought by macroeconomic fluctuations to commercial banks.
2. Literature Review
(1) Macroeconomic Factors
Bing Xie studied the impact and contribution rate of macroeconomic factors on the non-
performing loan ratio of commercial banks and found a negative correlation between macroeconomic
factors and non-performing loans [1]. Fang Xiao, from the perspective of specific provinces, took
Guangdong Province as an example to study the impact of economic downturn on the operations of
commercial banks [2]. Hongbo Liang studied the relationship between macroeconomic uncertainty
and credit risk and found that when macroeconomic uncertainty increases, the non-performing loan
ratio significantly rises, causing banks to tighten credit supply and significantly increase credit risk
[3].
Xueting Wang conducted an empirical study on the influence of leverage ratio and capital
adequacy ratio on the non-performing loan ratio of commercial banks and the moderating effect of
macroeconomic factors. The study found that macroeconomic factors have a moderating effect on the
impact of leverage ratio and capital adequacy ratio on the non-performing loan ratio [4]. Zeyu Zhou
used quarterly data on macroeconomics and financial markets to empirically study the significance
of five influencing factors on the non-performing loan ratio of various types of commercial banks [5].
Tao Wu selected specific banks and researched the influence of macroeconomic factors on non-
performing loans in a provincial branch of JS from the three dimensions of regional economic
conditions, inflation level, and industry structure [6].
(2) Commercial Bank Factors
Changfeng Fang studied the factors influencing the performance of commercial banks and found
that the macroeconomic environment has a significant impact on the average performance of Chinese
commercial banks [7]. Chan Liu focused on the significant relationship between finance and
economic fluctuations and, through constructing a micro-mechanism model, empirically studied the
impact of commercial banks’ pro-cyclicality on the macroeconomy. The study found a positive
correlation between the growth rate of loan balance and GDP growth rate, indicating that large
commercial banks have counter-cyclical effects on the macroeconomy [8]. Xihe Liu conducted
research on the macroeconomic effects of commercial bank deleveraging based on commercial banks,
government financing platforms, industrial and commercial enterprises, and household sectors. The
study found that the greater the degree of deleveraging, the greater the impact on the macroeconomy
[9].
Jia Li compared commercial banks in 16 different countries and found differences in the degree
and direction of the impact of different macroeconomic factors on non-performing loans in
commercial banks across countries [10]. Jijie Wei explored the main influencing factors of the non-
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231198
257
performing loan ratio of commercial banks in four regions of China: Northeast, East, Central, and
West. The study found that regional factors do not have a significant impact on the non-performing
loans of commercial banks [11]. Huiyue Zhao combined micro-indicators of banks with
macroeconomic indicators to analyze the relationship between bank management and economic
operations [12].
The above literature provides a research foundation for this paper. Scholars have focused more on
the cyclical fluctuations of macroeconomics, with fewer studies on the empirical research using
macroeconomic indicators related to residents’ production and daily activities. There is relatively little
research literature on the factors influencing the non-performing loan ratio under the new economic
situation. This paper conducts research on the non-performing loan ratio of commercial banks in 2022,
analyzes the factors influencing the non-performing loan ratio of commercial banks, establishes a
model, and draws corresponding conclusions to provide policy recommendations for the stable
development of commercial banks.
3. Empirical Analysis of Non-performing Loans in Commercial Banks
(1) Sample Selection and Data Source
Macroeconomic indicators reflect the economic situation, including main indicators such as Gross
Domestic Product (GDP), inflation and deflation, investment indicators, consumption, finance, fiscal
indicators, etc. This paper studies the relationship between macroeconomic factors and non-
performing loan ratio. The generation of non-performing loans is closely related to residents’
production and living activities. Therefore, this paper selects 31 provinces and cities in China in 2022
as the sample. The dependent variable is the non-performing loan ratio. By reviewing relevant
literature and combining with the innovative points of this paper, the initial control variables are
determined as regional GDP, average urban employment salary, the proportion of the tertiary industry,
per capita consumption expenditure of residents, the number of recipients of unemployment insurance
for residents, and the number of recipients of old-age insurance for residents. These six variables will
be tested and the final model will be determined in the subsequent empirical research.
The initial data analyzed in this paper comes from the “China Regional Financial Operation Report”
published by the People’s Bank of China and the National Bureau of Statistics. We selected data from
31 provinces and cities in China in 2022 for analysis, with a sample size of 31. To eliminate the
dimensionality difference between different variables, this paper standardizes the data by taking the
logarithm of both the independent and dependent variables.
(2) Model Specification
This paper selects 31 provinces and cities in China in 2022 as the sample to analyze the impact of
macroeconomic factors on non-performing loans in commercial banks through data analysis. To
reduce the absolute differences between data and avoid the influence of individual extreme values, so
as to better explain the economic significance of the results, this paper takes the logarithm of all
variables. To optimize the model, this paper conducts tests for multicollinearity, heteroscedasticity,
and serial correlation for all variables. The preliminary multiple regression model in this paper is as
follows:
Where represents the non-performing loan ratio, is the constant term,
represents regional GDP, represents average urban employment salary, represents the
proportion of the tertiary industry, represents per capita consumption expenditure of
residents, represents the number of recipients of unemployment insurance for residents, and
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231198
258
represents the number of recipients of old-age insurance for residents.
(3) Empirical Analysis of the Model
1. Multicollinearity Test
Using the Klein discriminant method, the correlation analysis between the six selected variables
is conducted to exclude the problem of multicollinearity in the model. The results are shown in Table
1, indicating the presence of multicollinearity in the model.
Table 1: Klein Discriminant Method Results.
GDP
AS
PTI
PCE
NUI
NEI
GDP
1.000000
0.004034
0.070474
0.500766
0.880460
0.796755
AS
0.004034
1.000000
0.811488
0.734880
0.117770
-0.529865
PTI
0.070474
0.811488
1.000000
0.693932
0.238682
-0.371497
PCE
0.500766
0.734880
0.693932
1.000000
0.619650
-0.076697
NUI
0.880460
0.117770
0.238682
0.619650
1.000000
0.584300
NEI
0.796755
-0.529865
-0.371497
-0.076697
0.584300
1.000000
Multicollinearity issues can lead to ineffective model results. Below, we will address the issue of
multicollinearity by conducting linear regressions of each explanatory variable on the dependent
variable. This will provide the coefficient of determination R2 for different models, as shown in Table
2.
Table 2: Linear Regression Results of Explanatory Variables on Dependent Variable.
Variable
R2
GDP
0.095563
AS
0.257123
PTI
0.024302
PCE
0.108195
NUI
0.015320
NEI
0.000010
Based on the coefficient of determination results in Table 2, the variables will be gradually added
to the regression model. Following the criterion of minimizing the AIC, the final model includes the
explanatory variables: average urban employment salary (AS), regional GDP (GDP), the proportion
of the tertiary industry (PTI), and the number of recipients of unemployment insurance for residents
(NUI). The regression results of the corrected model are shown in Table 3.
Therefore, the corrected multiple regression model is as follows:
Table 3: Regression Results of the Corrected Model.
Variable
Coefficient
Prob.
C
32.70687
0.000000
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231198
259
Table 3: (continued).
AS
-3.410106
0.000000
GDP
-0.672077
0.000500
PTI
3.272313
0.003900
NUI
0.365527
0.009100
2. Heteroscedasticity Test
The White test is used to determine whether there is heteroscedasticity in the model. Partial results
are shown in Table 4.
Based on the results of the White test in Table 4, . According to the White test, at
the level, the critical value from the chi-square distribution table is X20.05(14)= 23.685. By
comparing the calculated X2 test statistic with the critical value, we accept the null hypothesis and
reject the alternative hypothesis. Additionally, since the p-value corresponding to is greater than
0.05, it can also be concluded that there is no heteroscedasticity in the model.
Table 4: White Test Results.
F-statistic
3.205068
Prob. F (14,41)
0.0141
Obs*R-squared
22.85161
Prob. Chi-Square (14)
0.0627
Scaled explained
SS
15.85290
Prob. Chi-Square (14)
0.3224
3. Serial Correlation Test
The coefficient of determination R2 for this regression equation is 0.677, indicating a good model
fit. Furthermore, all four variables have passed the T-test, indicating their significance in the
regression. For a sample size of 31 and a significance level of 5%, and considering a model with 4
explanatory variables, consulting the DW statistic table, it is found that dL=1.160 and dU=1.735. The
DW value for the model is 1.829. Based on the analysis, the DW value falls within the range of (dU,
4-dU), which is between 1.735 and 2.265. Therefore, the null hypothesis H0 is accepted, indicating
that the equation is not autocorrelated. This indicates that the model results are valid and further
analysis can be conducted.
(4) Model Empirical Results
According to the regression results of the model, the coefficients between the proportion of the
tertiary industry and the number of recipients of unemployment insurance benefits and the non-
performing loan ratio are positive, indicating that both factors have a positive impact on non-
performing loans. According to the dimensionless regression results, for every one unit increase in
the proportion of the tertiary industry, the non-performing loan ratio of commercial banks will
increase by 3.272 units. For every one unit increase in the number of recipients of unemployment
insurance benefits, the non-performing loan ratio of commercial banks will increase by 0.366 units.
In 2022, due to the impact of the COVID-19 pandemic, regions with a higher proportion of the tertiary
industry will be more strongly affected by the pandemic, leading to higher non-performing loan ratios
in regions with a higher proportion of the tertiary industry. The number of recipients of unemployment
insurance benefits reflects the living conditions of local residents. When their lives are affected, their
ability to repay debts naturally decreases, resulting in an increase in the non-performing loan ratio.
The coefficients between the average salary of urban employees, the regional GDP, and the non-
performing loan ratio are negative, indicating that both factors have a negative impact on non-
performing loans. For every one unit increase in regional GDP, the non-performing loan ratio of
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231198
260
commercial banks will decrease by 3.410 units. For every one unit increase in the average salary of
urban employees, the non-performing loan ratio of commercial banks will decrease by 0.672 units.
Both indicators reflect the development of the regional economy. In regions with better development,
the living standards of residents improve, naturally reducing the possibility of non-performing loans.
In conclusion, all four variables are consistent with their practical significance. Overall, the
average salary of urban employees has the greatest impact on the non-performing loan ratio, followed
by the proportion of the tertiary industry, regional GDP, and the number of recipients of
unemployment insurance benefits.
4. Conclusion and Recommendations
(1) Conclusion
1. The proportion of the tertiary industry has a positive impact on the non-performing loan ratio of
commercial banks. Generally, the higher the proportion of the tertiary industry, the higher the non-
performing loan ratio of commercial banks.
2. The number of recipients of unemployment insurance benefits has a positive impact on the non-
performing loan ratio of commercial banks. Generally, the higher the number of recipients of
unemployment insurance benefits, the higher the non-performing loan ratio of commercial banks.
3. The average salary of urban employees has a negative impact on the non-performing loan ratio
of commercial banks. Generally, the higher the average salary of urban employees, the lower the non-
performing loan ratio of commercial banks.
4. Regional GDP has a negative impact on the non-performing loan ratio of commercial banks.
Generally, the higher the regional GDP, the lower the non-performing loan ratio of commercial banks.
(2) Recommendations
1. Optimize industrial structure and improve loan structure
Local governments should accelerate the upgrading of the industrial structure in their regions,
optimize the industrial structure, promote the transformation and upgrading of traditional industries
and manufacturing industries, support the development of local emerging industries and industries
with local characteristics, attract high-tech industries, and control the proportion of the tertiary
industry reasonably. Play a regulatory role, establish a platform for enterprise experience exchange,
and provide financial support to enterprises. While complying with macroeconomic policies,
commercial banks should optimize their loan approval system, actively provide financial support to
small and medium-sized enterprises, optimize the structure of fund allocation, and reduce the
possibility of non-performing loans.
2. Promote coordinated economic development and increase regional GDP
The level of regional economic development is an important factor affecting the development of
the banking industry. Provincial governments in economically backward regions should learn from
the development paths of developed regions, adapt measures to local conditions, attract investment,
stimulate consumption, broaden import and export trade, and vigorously develop the regional
economy. Improve the income distribution system, ensure the reasonableness of income levels in
various industries, and thereby increase the local employment rate and per capita salary level. Pay
attention to macroeconomic indicators in the local area, formulate corresponding financial regulatory
policies, and ensure the standardized and efficient development of the banking industry.
3. Improve the financial regulatory system and establish risk prevention mechanisms
Different levels of regional economic development require different regulatory goals. Local
governments should formulate reasonable financial regulatory goals based on actual conditions,
reduce stock and control increment. Commercial banks should adapt to the trend of interest rate
marketization, accelerate business transformation, increase investment in financial technology,
improve service efficiency and quality, reduce dependence on loans as the main source of profit,
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231198
261
develop diversified sources of profit, and reduce the risk of non-performing loans caused by the
pressure of loan profitability.
References
[1] Xie, B. (2009). Analysis of macroeconomic factors influencing non-performing loans of commercial banks. Finance
Theory and Practice, 30(06), 22-25.
[2] Xiao, F. (2009). The impact of macroeconomic downturn on the operation of commercial banks and
countermeasures: a perspective from Guangdong. Southern Finance, (03), 63-66.
[3] Liang, H. B., & Liu, Y. L. (2012). Empirical study on the relationship between credit risk of commercial banks and
macroeconomic uncertainty in China. Financial Theory and Practice, (03), 81-84.
[4] Fang, C. F., & Liu, S. L. (2011). Impact factors of commercial bank performance: industry structure, governance
structure, and macroeconomic environment. Financial Forum, 16(06), 9-17. DOI: 10.16529/j.cnki.11-
4613/f.2011.06.002.
[5] Liu, C. C. (2019). Macroeconomic fluctuations, financial regulatory requirements, and pro-cyclicality of
commercial banks. Finance and Economics, (08), 20-27. DOI: 10.19622/j.cnki.cn36-1005/f.2019.08.004.
[6] Liu, X. H., & Wang, J. J. (2019). Macroeconomic effects of commercial bank deleveraging. Nankai Journal
(Philosophy, Social Sciences, and Economics Edition), (02), 168-179.
[7] Wang, X. T., Chen, D. P., & Liao, H. F. (2021). The influence of leverage ratio and capital adequacy ratio on non-
performing loans of commercial banks: the moderating effect of macroeconomic conditions. Hainan Finance, (08),
3-15.
[8] Zhou, Z. Y. (2021). The impact of macroeconomics and financial markets on the non-performing loan ratio of
commercial banks. (Doctoral dissertation). Hunan University. DOI: 10.27135/d.cnki.ghudu.2021.003546.
[9] Wu, T. (2021). Research on the impact of macroeconomic factors on the non-performing loan ratio of JS Bank
provincial branches. (Master’s thesis). Inner Mongolia University. DOI: 10.27224/d.cnki.gnmdu.2021.001435.
[10] Li, J. (2015). International comparison of factors influencing non-performing loans of commercial banks based on
a macro perspective. Times Finance, 24, 68-70.
[11] Wei, J. J. (2021). Research on regional differences in the non-performing loan ratio of commercial banks. Investment
and Entrepreneurship, 32(02), 22-24.
[12] Zhao, H. Y., & Shen, H. Y. (2023). Research on the influencing factors of non-performing loan ratio of commercial
banks. Guandong Academic Journal, (01), 51-61. DOI: 10.19470/j.cnki.cn22-1417/c.2023.01.014.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231198
262
The Sino-US Trade Conflict’s Impact on Semiconductor
Stock Index: China’s Case
Zhenxiang Liu1, a, *
1China Shenzhen chenghan school (international curriculum experimental campus), Shenzhen,
518116, China
a. 201004010218@stu.swmu.edu.cn
*corresponding author
Abstract: On march 22, 2018, U.S. president trump signed a trade memorandum with China
at the White House and restricted Chinese companies investment and acquisitions in the
united states. The Sino-US trade war officially started. This trade war lasted for more than a
year. China has imposed tariffs on each other six times, which has had a great impact on many
fields. This paper background is at the point in time when the Sino-US trade war has just
ended. The purpose and content of this paper is related impact of the China and US trade war
on my country’s semiconductor industry. This article used to analyze the impact of the trade
war through the previous data area, and this paper will use the previous data to predict the
future semiconductor trend through the ARIMA model (with the Sino-US trade war as the
node), and predict that the Sino-US trade war will have little impact on the daily, some impact
on the weekly, and a great on the monthly. Based on our results, people can speculate that
from a long-terms perspective, and choose to invest in the semiconductor industry.
Keywords: China, trade war, US, chip industry, ARIMA model
1. Introduction
2016 Trump during the presidential campaign, trump claim basis USA trade law section 201 and
section 301 to imposing tariffs on China. He hopes to reduce USA trade deficit and boost the economy
[1]. Along with rise of the China economy, USA economic policy toward China has undergone drastic
changes under recent president. During the Clinton administration, China and US was “strategic
partnership”. During the Bush administration, the US household investment developed rapidly, (come
to recession during 2008). A huge amount of desire on borrowing come to place and China, on the
same time, was experiencing a dramatic development after the policy of reform and open. Therefore,
China is willing to loan out money to exchange for greater return in the future, so that President Bush
claimed that relationship between China and US was “responsible stakeholder country”. During the
Obama administration China and US was “mutual respect, mutual benefit and win-win partnership”,
challenge change have taken place in Sino-US relation [2]. In August 2017 Trump according US trade
law section 301 to investment about China infringes us intellectual property exhibition [3]. This move
is considered the first direct trade measure in the US and China trade war.
On March 22, 2018, US president Trump signed a trade memorandum with China at the white
house and increased tariffs on some products from China. This represents the start of the US and
China trade war, this trade war dispute has resulted in repeated tariff increase.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231200
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
263
On the December 1, at the G20 summit in Argentina, Chinese president Xi met with us president
Trump and reached a proposal for a 90 day truce [4]. The traded war in the 2019 caused by Huawei
being included in the list of controlled entities, mainly occurred in the two fields. The first one
industry is China has advantages, for instance light industry, the second industry is China don’t have
advantages, such as the semiconductor industry. US sanctions against China are mainly focused on
the semiconductor industry [5].
Some scholars discuss and analyze and discuss the direction of the current round of trade friction
between China and US from the perspective of going beyond the trade relationship between China
and the US and even beyond the two economic relationships between the two countries. Goals1, to
ease the trade deficit Goals. Goals 2, to combat China’s advanced manufacturing and high-tech
industries. Goals 3, require China’s to fully open its financial industry [6].
In this Sino-US trade war, the area where China is taxed include: semiconductors and some high-
end products. basically, no impact. For example, in the short term, the United States has an impact
on China’s shipping industry. The main impact on China’s shipping industry is that the trade war’s
psychology impact on us shippers may affect the promotion of the hot season [7].
From the perspective of the forestry industry, trade protectionism will not only not make the timber
harvesting and processing industry in the United States grow stronger, but will accelerate the
shrinkage of th timber industry, which is contrary to the trump administration’s “manufacturing return”
concept [8].
From the perspective of the semiconductor industry, the United States is trying to use the “trade
war “to attack the developing of my country’s high-tech industry. In the long run, it will not be
effective at all, but in the short terms, it will have some adverse effects on my country’s technological
developing, which is in line with our prediction [9].
2. Research Design
2.1. Data Source
This article uses the search engine Investing [10] to obtain and search the index of chip industry and
this paper obtain opening price, closing price stock yield and expense. And process it and divide it
into daily weekly monthly and intercept the data from 2010 to 2020.
2.2. Weak Stationarity Test
A stationarity test is required for this model in order to identify whether the model is smooth enough
so that a forecast method can be decided (The assumption of the model is not smooth initially). In
this article, the manipulation on the whole data set is taking the first and second order difference so
that the construction of a smooth model is completed. The result of ADF test on STATA is represented
below, in table 1, no matter the daily, weekly, and monthly data, the p-value of them is 0 after the
first and second order difference, which implied that the null hypothesis about the data set is not
stationary can be rejected. In conclusion, the entire data frame passes the stationarity test after the
first and second order difference.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231200
264
Table 1: Weak stationarity test.
t
p
Daily
Raw
-2.864
0.1745
1st order difference
-26.753
0.0000
Weekly
Raw
-2.980
0.1376
1st order difference
-11.864
0.0000
2nd order difference
-20.616
0.0000
Monthly
Raw
-2.677
0.2456
1st order difference
-6.549
0.0000
2nd order difference
-9.358
0.0000
2.3. ARIMA Specification
ARIMA model is the combination of AR and MA model, first, AR model is an autoregressive model,
which is a prediction method that use historical information to predict the future performance of data.
It requires the data frame to meet some specific smoothness tests, so that past historical data can
reliably estimate horizontal data. The first order form of autoregressive model can express as:
(1)
Among them,
reference to observed value at time t, c is constant, is the autoregressive
coefficient, is the observed value at time T-1, and is the error value.
When the autoregressive model is of order p, the model can be expressed as:
(2)
Among them ,, is the autoregressive coefficient.
Second, the MA model, in the way opposite with AR model, doesn’t rely on the historical data. It
is defined based on the white noise sequence (a sequence that the past and future data is not correlated
and only fluctuated around a specific mean value) The basic ideas of MA model is: most of time series
should be relatively stable. On the stable basis, the tag value at each point in time is fluctuated in a
determined pattern by various unforeseen contingencies in the past period of time. That is within
period of time, the time serious should fluctuate around a certain mean value. The label at the time
point of the sequence will move around a certain mean value.
In summary, the ARIAM model is to speculate the future through past data.
MA model can be expressed as:
(3)
The ARIMA model can be expressed as:
(4)
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231200
265
3. Empirical Results and Analysis
3.1. Order Identification
PACF of stationary time series is a function of ACF, and PACF is introduce through ACF.
(5)
(6)
(7)
(8)
ACF is independent of any particular time t and is a function of the time interval h Expressed:
(9)
Among them, h is the interval, describing the correlation between the variable sequence and the
sequence of its own interval time h. covariance of the two sets of vectors. the
standard deviation of the two sets of vectors.
PACF determines the order of p (the order of AR), and the ACF determines the MA (q) Order.
Daily, the data is entered after 10 and falls into the shaded part. Weekly the data is entered after 1
and falls into the shade part. Monthly, the data is entered after 1 into the shade part (see Figure 1).
PACF
ACF
Daily
Weekly
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231200
266
Monthly
Figure 1: ARMA (p, q) identification.
(Photo credit: Original)
3.2. Prediction Result and Interpretation
Table 2, the original hypothesis of this test is: there is no autocorrelation in the sequence (that is, the
sequence is white noise), such as prob<0.05, the original hypothesis is rejected. It can be seen from
the figure that daily meets the null hypothesis, monthly does not meet the null hypothesis, and not
meets weekly.
Table 2: Residual test.
Model
Portmanteau (Q) statistic
Prob > chi2
Daily-ARIMA(10,1,10)
2192.6054
0.0000
Weekly-ARIMA(8,2,1)
52.1470
0.0945
Monthly-ARIMA(6,2,1)
36.8376
0.6134
This paper select 3 times period daily, weekly, monthly for analysis. From the overall trend of the
figure, figure 2 daily keeps rising steadily, even more than expected growth, and figure 3 weekly and
short-terms (almost 6 weeks) keep rising even higher than expected it is expected to growth more,
and the short-terms (more than 7 weeks) actually growth is less than the expected growth (Figure 4).
Figure 2: Actual value and fitted value, daily.
(Photo credit: Original)
2000
2050
2100
2150
2200
2250
2300
2350
2400
Actual value Fitted value
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231200
267
The actual results of figure 3 monthly are two different from the forecast result, indicating that the
US and China trade war has a great impact on the valuation of China’s semiconductor industry.
Figure 3: Actual value and fitted value, weekly.
(Photo credit: Original)
Figure 4: Actual value and fitted value, monthly.
(Photo credit: Original)
This paper is about the influence of the Sino-US trade war on the China’s semiconductor industry.
In very short-terms (daily), basically no impact. In the short-terms (weekly)have small impact, but
there is no effect if it is less than 7 weeks and it will be affected if it exceed 7 weeks (but not more
than 7 weeks). long-terms(monthly) have huge impact (more than 60 percentage). During the period
1700
1800
1900
2000
2100
2200
2300
2400
2500
2600
2018-01-12
2018-01-19
2018-01-26
2018-02-02
2018-02-09
2018-02-16
2018-02-23
2018-03-02
2018-03-09
2018-03-16
2018-03-23
2018-03-30
2018-04-06
2018-04-13
2018-04-20
2018-04-27
2018-05-04
2018-05-11
2018-05-18
2018-05-25
Actual value Fitted value
1700
1800
1900
2000
2100
2200
2300
2400
2500
2600
2700
Sep-17 Oct-17 Nov-17 Dec-17 Jan-18 Feb-18 Mar-18 Apr-18 May-18 Jun-18 Jul-18 Aug-18
Actual value Fitted value
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231200
268
of US and China trade war, the main means of US sanction against China were tax increase and
restriction on the export of the semiconductor industry.
China
US
Figure 5: Industry structure of China and US.
(Photo credit: Original)
There is a huge distinguish in industrial structure of Sino-US (see Figure 5). Proportion of the
tertiary industry in US is much higher than that of the primary and secondary industries. In China,
although the tertiary industries occupy a dominant position in the industry structure, compared with
the United States, the secondary industry is two times that of the United States. The proportion of
primary industry in the United States is one-seventh that of China. The primary industries refer to
agriculture, the secondary industry refers to industries such as automobiles and pharmaceuticals, and
the tertiary industries such as automobiles and pharmaceuticals, and the tertiary industry mainly
includes finance, education, telecommunication, retail entertainment medical care etc. [11].
China’s economy has developed enormously in just a few years, Sino-US relations have undergone
tremendous changes, basically it can be roughly divided into four stages.
The first stage was between 1979 and 1988, when Sino-US relations were relatively stable; the
second stage was from 1989 to 1994, when Sino-US relations deteriorated sharply; the third stage
was from 1995 to 2000, when Sino-US relations were relatively stable. After stability, the fourth stage
is from 2001 to 2005. although there are still frictions in bilateral relations in this stage, Sino-US
relations are improving [12].
4. Conclusion
The object of this paper is to predict the stock index of the semiconductor industry without the Sino-
US trade war. The research background of this paper is that the Sino-US trade war has affected many
industries in China, and the United States had a huge influence on China’s chip industry. After
forecasting, this paper found that there is no daily impact, and weekly impact, but there is no complex
impact, and there is a great monthly impact, but there is no complex impact, and there is a great
monthly impact. Because there may be some problems with the ARIMA model, this paper suggests
that latecomers in the academic community can use the LASSO model and the neutral net model to
predict.
This paper has several suggestions for policy makers and investors. Investors,some time ago due
to the Sino-US trade war, China’s financial market suffered the most losses. This article recommends
investors investing in the semiconductor industry. Investors who want to hold chip industry stocks in
7.90%
40.50%
51.60%
primary industry secondary industry teriary industry
0.90%
19.10%
80.00%
primary industry secondary industry teriary industry
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231200
269
the short term may consider it more, in the short terms, this impact will not disappear unless China
can make a huge breakthrough in the semiconductor industry in the short term. However, if investors
who want to hold stocks in the semiconductor industry in the long term can choose to buy, people
can trust the relevant government departments, there will be some policy support for companies like
HUAWEI that are trying to break the US blockade of the semiconductor industry in some ways.
Policy makers suggest using some financial means to save the financial market of the
semiconductor industry and increase support for the semiconductor industry. Policy support and R&D
investment to avoid being locked in by foreign countries.
References
[1] JY Chen. (2018). The background,reasons,nature and China’s countermeasures of the Sino-US trade war. Journal
of wuhan university (philosophy and social sciences edition) 71(5), 72-81.
[2] C Gao. (2014). Changes in peripheral structure and strategic adjustment in the context of china’s rise. International
Economic Review, 2, 33.
[3] RX Cao. (2020). Analysis of “special section 301” related issue in the Sino-US trade war. Heilong jiang finance,
(2), 11-14.
[4] MN Zhang. (2019). Reshaping Sino- US relations in “ trade war ” and economic and trade negotiations finance
and economics , (2), 1-1.
[5] ZW Tang, YX Li , & LP Zhang . (2021). The identification method and breakthrough path of “stuck neck “technology
under the background of Sino-US trade friction -- Taking the electronic information industry as an example. Science
and technology progress and countermeasures, 38(1), 1-9.
[6] SA Huang. (2018). China cannot passively open up its finance- strategic thinking based on the Sino-US trade war.
Journal of Dongbei University of finance and economics, (3), 3-5.
[7] Y Chen, DJ Wang , HY Su , HF jiang , & X Zhang . (2019). The impact of the Sino-US trade war on forest product
trade and countermeasures. Forestry Economic Issues, 39(1), 1-7.
[8] T Zhang. (2018). The impact of the Sino-US trade war on the shipping industry . China water transport, (4), 12-13.
[9] RK Zhang. (2019). The United States’ technological blockade against China in the context of the trade war and my
country’s response strategy. Science and Technology China, (8), 1-4.
[10] Investing homepage. https://cn.investing.com
[11] LH Zhang. (2013). The current situation of China’s industrial structure and its adjustment ideas- based on a
horizontal comparison of the industrial structures of China and the United States. Economic Perspective , (1), 29-
30.
[12] J QI. (2008). Analysis of the evolution trend of Sino-US economic and trade relations. Contemporary Economic
Research, (4), 47-49.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231200
270
Application of Various Mechanisms on Students School
Choice Problem
Zhenhao Cai1, a,*
1School of Mathematics, Renmin University of China, Beijing, China
a. 2022201257@ruc.edu.cn
*corresponding author
Abstract: School choice mechanisms are used to determine how to assign students to the
public schools. These mechanisms take into account the students’ preferences order and the
schools’ priority order to calculate an assignment that meets the needs and preferences of
both students and schools. This paper sketches the school choice problem and its real-life
condition in some cities, discusses the theory background of school choice problem and its
specialty as a one-side matching problem. Also, this paper introduces the Defer Acceptance
(DA) mechanism, the Top Trading Cycle (TTC) mechanism, the Boston (BOS) mechanism
and the Serial Dictatorship (SD) mechanism. The paper analyzes four different mechanisms’
desirable properties, including strategy-proofness, stability and Pareto efficiency. Meanwhile,
the flaws of the mechanisms are introduced. At last, based on the analysis and the purpose of
improving students’ welfare, the paper raises some suggestion for the government, schools
and parents, ensuring the students a transparent, fair and equal environment of choosing their
schools.
Keywords: school choice, matching mechanisms, matching theory
1. Introduction
In many countries, public school systems have historically operated as monopolies, with the goal of
fulfilling the objectives set by legislatures and educational institutions. However, educational
policymakers in countries like the America have increasingly recognized the need for educational
reforms, particularly in response to the school choice debate initiated by economist Milton Friedman
in 1955 [1]. School choice refers to a range of programs that aim to empower parents to select the
schools their children attend. Under traditional school choice systems, the mechanisms assign
students to schools according to their residential districts, regardless of individual preferences or
school quality. Wealthy parents already have a choice in the schools their children attend since they
can afford to relocate to a place with better educational options or enrol them in a private school, but
in the United States, access to these options is constrained by a lack of financial resources. School
choice policies aim to enhance education quality by expanding options, particularly for parents
without the financial capacity to exercise choice, and broaden students’ access to high-quality
education, no matter what their socioeconomic background are. Advocates of school choice believe
that granting autonomy to public schools can lead to better educational outcomes by allowing schools
to tailor their approaches to meet the needs of students and parents.They also argue for providing
parents with more information to make informed choices.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231202
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
271
However, the impact of these market-based reforms on social capital and overall educational
welfare is still debated among scholars.Social capital refers to the networks, relationships, and trust
that contribute to positive social and educational outcomes.Some argue that school choice can
enhance social capital by empowering parents and increasing their involvement in their children's
education. Others suggest that it may lead to increased segregation and reduced support for the public
school system. In short, based on laws of different states and countries, there are several ways for
students to be admitted to colleges or universities, accordingly, in the language of economics,
different matching systems should be applied. Thus, present work is going to compare different
matching mechanisms, especially in the context of school choice model, and analyze the pros and
cons of them in various real-world situations, and put forward some suggestions for the school,
parents and the government in order to offer students a better welfare.
2. Mechanisms on School Choice Problem
2.1. Background Description
In education, school choice is literally one of the most heated topic. The school choice model is a
one-sided matching where only one of two sets of agents have preferences over the other and act
strategically, while the other set of agents do not have preferences and do not act strategically. This
model is commonly used for allocating seats in public schools, centralized university admissions, and
the allocation of vacant house, based on priorities [2]. In this model, the welfare of students is the
primary concern, and priorities, such as neighborhood proximity or sibling enrollment, are
predetermined and not strategically reported. Due to the limitations of assigning each student to their
top choice school, mechanisms need to be designed to allocate students to schools in a fair and
efficient manner [3].
In this problem, Students (denoted by i) and schools (denoted by s) are the two sets of agents. Each
student has a preference list (which is often strict in theory) over the public schools, and there’s a
maximum quota (denoted by qs) of the number of available seats for each school. A matching in this
context is a mapping that determines students are assigned to a particular school or just stay
unassigned. It is important to note that a school is only mapped to a student if the student chooses the
school and is chosen by the school, and the maximum quota (qs) is the constraint that a school’s
capacity of students.
Before introducing the important mechanisms in literature, the definition of some properties of the
matching outcome need to be explained. The core concept of the school choice literature is
elimination of justified envy. No unassigned pair of a student and a school (i, s), where student i
prefers school s to his or her present assignment meanwhile he or she has a higher priority compared
to an student in school s, exists. It is obvious that in the context of school choice problem the concept
elimination of justified envy is similar to the concept stable in college admission problem[3]. The
definition of a feasible matching is that every student whom a school assigns is acceptable to the
school. What’s more, the definition of an individually rational(IR) matching is that no student would
rather be unassigned than accept his or her present assignment, and a matching is considered non-
wasteful when every student prefers his or her present assignment to some other school with at least
one available seat. A matching is stable if it is non-wasteful, individually rational, and eliminates
justified envy [4]. Besides, a matching ν is Pareto dominated by matching µ if µ assigns every student
a weakly better match and a strictly better match for one student at least. The definition of a Pareto
efficient matching is that it is not Pareto dominated by any other matching. Since only the students’
welfare is considered in the context of school choice problem, the student-optimal (stable) matching
is not Pareto dominated by any other (stable) matching. The strategy-proofness of a mechanism is
that no student can gain a better assignment, that is, be admitted to a school with higher priority, from
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231202
272
misreporting. If misreporting makes him or her better off compared with being truthful, then the
mechanism is vulnerable to a student’s strategic manipulation.
In the mechanisms to be described, the algorithm receives preference lists of schools’ ranking from
the students and the number of available seats of each school. The students’ have distinct rankings
without ties. In the event of ties, the mechanism breaks them arbitrarily. Using this information, the
mechanism then generates a matching that assigns students to schools according to their preference
list and the maximum quota of the schools. It aims to optimize the allocation of students to schools
according to their stated preferences and the constraints imposed by the school capacities. In order to
simplify and unite the notation, suppose a school choice problem (i, s, qs, P (priority), PL (Preference
List)) is given in the description of following mechanisms [5].
2.2. The Student-Proposing Defer Acceptance (DA) Mechanism
Step 1. Every i make a proposal to the school which is his or hers top choice. Each school s places
the first qs applicants and also the acceptable ones on its hold list, and rejects the rest. Decrease a
school’s quota by one if it accepts a student.
Step k. A student who is rejected by a school at step k-1 can make a proposal to their next favorite
school. Each school s places the first qs applicants and also the acceptable ones on its hold list, and
rejects the rest. Decrease a school’s quota by one if it accepts a student.
End. The mechanism stops operating when all students are accepted, or all schools have reached
their maximum quota. Leave the remaining students unassigned.
The DA mechanism is one of the most classic mechanism in matching theory. Due to its stable
outcome and strategy-proofness, DA mechanism is applied in many real-life problems, including
school-choice problem. Although the stable match that is best for all the students, and they are the
ones whose welfare should be cared about, DA mechanism does not possess Pareto efficiency due to
the mutual exclusivity between efficiency and stability [6]. The student-proposing Defer Acceptance
(DA) mechanism has another name, student-optimal DA mechanism. According to its name, it
eliminates justified envy, and any other mechanism that possesses the property of elimination of
justified envy is Pareto dominated by this [7].
The DA mechanism is a widely used algorithm for matching students to schools based on their
preferences. It involves a series of proposals and rejections until a stable matching is reached.
Kesten’s mechanism (EADA mechanism) introduced efficiency adjustments to the DA mechanism,
which in turn were further simplified by Tang and Yu while maintaining the same matching outcomes
[8, 9]. The modification by Tang and Yu demonstrates that it is possible to enhance the efficiency of
Kesten’s algorithm. The improved algorithm provides a more efficient and simplified approach for
studying the school choice problem. In general, the Defer Acceptance (DA) mechanism is a well-
behaved mechanism [3].
2.3. The Top Trading Cycle (TTC) Mechanism
Step 1. If a student s has a top choice among the schools, point from he or she to that school. If not,
the student points to himself or herself, indicating a preference to remain unassigned. Every school
point to the highest priority student for the school. This creates a set of student-school preferences.
There must be at least one cycle (a student pointing to himself or herself is also considered one).
These cycles represent potential assignments. For each cycle, the school offers a seat to the student
points to it, or a student leaves unassigned if he or she is pointing to himself or herself. If there are no
more available seat of a school, remove that school from consideration [2].
Step k. For each unassigned student, if he or she has a top choice among the remaining schools,
point from he or she to that school. If not, the student points to himself or herself, indicating a
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231202
273
preference to be unassigned. For each remaining school, the highest priority student for that school is
pointed from it. There must be one cycle at least (a student pointing to himself or herself is also
considered one). These cycles represent potential assignments. For each cycle, the school offers a seat
to the student points to it, or a student leaves unassigned if he or she is pointing to himself or herself.
If there are no more available seat of a school, remove that school from consideration.
End. The mechanism stops operating when all students are accepted, or all schools have reached
their maximum quota. Leave the remaining students unassigned.
The Top Trading Cycle (TTC) mechanism is a popular one used in school choice problems to
allocate students to schools. It is a strategy-proof mechanism, meaning that it incentivizes truthful
reporting of preferences from both students and schools. Overall, the TTC mechanism offers a
valuable combination of Pareto efficiency, strategy-proofness for students, and adaptability to handle
type-specific quotas when necessary. By incorporating the quotas into the algorithm, the mechanism
can still maintain both constrained strategy-proofness and Pareto efficiency. This ensures that the
assignment remains fair and efficient while adhering to the specified quotas.
By using the TTC algorithm, school choice systems aim to achieve efficient and fair allocations
while respecting the preferences of both students and schools.
2.4. The Boston (BOS) mechanism
Step 1. The mechanism only considers the 1st choice of each student. For each school s, find the
students that put the school on his or her 1st choice. The school s assigns seats to those students in
turn based on their priorities at s until all of the qs seats of the school s have been assigned, or, there
is no other student who has put the school on his or her 1st choice and is acceptable to s.
Step k. For all of the remaining students, the mechanism only considers the kth choices of them.
For each remaining school s, find the students that put the school on his or her kth choice. The school
s assigns seats to those students in turn based on their preference rankings at s until all of the qs seats
of the school s have been assigned, or, there is no student left who has put the school as on his or her
kth choice and is acceptable to s [10].
End. The mechanism stops operating when all students are accepted, or all schools have reached
their maximum quota. Leave the remaining students unassigned.
The Boston mechanism, also known as the “Boston mechanism with tie-breaking,” aims to assign
students to their top choice schools as much as possible. However, it is not considered strategy-proof
because students may misreport their preferences strategically to increase their chances of getting
their top choice school [10].
This issue arises when the mechanism uses tie-breaking rules to resolve situations where multiple
students have the same top choice school. In such cases, students may strategically manipulate their
preferences to improve their chance of being assigned to their top choice school.
As a result, the Boston Public Schools guide advises parents to consider strategic preference
submissions. This means that parents may be advised to list their preferred schools in a way that
maximizes their chances of being assigned to their top choice school, even if it may not accurately
reflect their true preferences.1
This strategic behavior can undermine the fairness and reliability of the assignment process, as it
may result in students not being assigned to their genuine top choice schools. Researchers and
policymakers have been working on developing alternative mechanisms that are both strategy-proof
and efficient to address these concerns.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231202
274
2.5. The Serial Dictatorship (SD) Mechanism
Step 1. Students are matched at the top of the priority order with his or her top choice school. After
the student is assigned, he is removed from the priority order, and the respective school’s maximum
quota is decreased by one. If there are no more available seat of a school, remove that school from
consideration.
Step k. Based on the school priority list, the highest priority remaining students are assigned to the
top-ranked school on her preference list that still has available capacity.
End. The mechanism stops operating when all students are accepted, or all schools have reached
their maximum quota. Leave the remaining students unassigned.
The SD mechanism is the only mechanism presented that has all the desirable properties:
elimination of justified envy, Pareto efficiency, and strategy-proof [11]. However, it can only be
applied to some of the one-side matching problems with universally accepted, open and transparent
priority order. In this context, schools are not strategy agents, but just choose students based on their
home address, which school his or her siblings are attending, and so on. Thus, even though SD
mechanism could generate a theoretical superior outcome, it is barely applied in real-life school-
choice system. Nonetheless, if there is a clear priority order, for example, score of college admission
exam, it is absolutely feasible and fair.
3. Conclusion and Suggestions
Despite there are various mechanisms, all of them have advantages and disadvantages. However, in
the real-life situation, the actual system may be far more complicated than just applying a mechanism
on students’ preference list. It could be a hybrid mechanism, considering many associated factors.
Thus, for the sake of improve the welfare of the students, the government should take a global view
of the location of indigenous schools, the overall preference of students, the “under-demanded schools”
and so on. Besides, governments are supposed to support research and initiatives concerning school-
choice mechanism innovation and improvement. Allocating sufficient funding to ensure quality
education for all students, including resources for extracurricular activities, mental health support,
and professional development for teachers. Enhance transparency by providing clear and easily
understandable information about the school selection process, eligibility criteria, and available
options. Ensure that students and parents have access to comprehensive information about schools,
including their performance, curriculum, extracurricular activities, and support services.
For the parents, they should establish open communication channels with the school, actively
engaging in regular discussions about their children’s progress and stay informed about their child's
education, including school policies and resources, to actively advocate for their best interests.
For the schools to be picked, fair is of primary importance. They need to provide equal
opportunities for all students, regardless of their background or abilities. Academically, design
mechanisms that promote fairness and equity in school assignments, minimizing biases and
discriminatory practices and regularly review and evaluate the allocation mechanisms to identify and
rectify any unfairness or unintended consequences. Recognize that different regions or communities
may have unique needs and preferences when it comes to school choice mechanisms. Thus, the
process of designing mechanisms can be adapted and customized accordingly. Consider allowing for
flexibility in the allocation process, such as accommodating late enrollments or transfers, addressing
special needs, or making provisions for changes in student preferences.
References
[1] Zweifel, A. (2010). Matching mechanisms in theory and practice, Munich, GRIN Verlag
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231202
275
[2] Hakimov, R. and Kübler, D. (2021). Experiments on centralized school choice and college admissions: a survey.
Experimental Econonomics, 24, 434–488
[3] Abdulkadiroglu, A. and Sonmez, T. (2003). School Choice: A Mechanism Design Approach, American Economic
Review, 93(3), 729-747
[4] Abdulkadiroglu, A. and Sonmez, T. (2013). Matching Markets: Theory and Practice, Advances in Economics and
Econometrics, 1, 3-47
[5] Abdulkadiroglu, A. and Andersson, T. (2022). School Choice. NBER Working Paper No. 29822
[6] Roth, A. E. (1982). The economics of matching: Stability and incentives. Mathematics of operations research,
7(4):617–628.
[7] Gale, D. and Shapley, L. S. (1962). College Admissions and the Stability of Marriage. American Mathematical
Monthly, 69(1), 9–15
[8] Kesten, O. (2010). School Choice with Consent, The Quarterly Journal of Economics, 1297-1348.
[9] Tang, Q. and Yu, J. (2014). A New Perspective on Kesten’s School Choice with Consent Idea, Journal of Economic
Theory, 154, 543—561.
[10] Kojima, F., and Ünver, M. U. (2014). The “Boston” school-choice mechanism: an axiomatic approach. Economic
Theory, 55, 515-544.
[11] Abdulkadir, E., and Peshko, O. (2020). School Choice Problem: Mathematical Formulation and Solution
Algorithms. Proceedings of the International Scientific Conference – Digital Economy and Society 2020 (DES2020).
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231202
276
Research on Marketing Strategy of Lawson Convenience
Store Brand from the Perspective of New Media
Jinsheng Tan1,a,*
1School of Modern Languages and Communication, University of Bottra Malaysia, Kuala Lumpur,
43400, Malaysia
a. rahwati@upm.edu.my
*corresponding author
Abstract: Lawson is a well-known convenience store brand, and marketing certainly uniquely
shapes Lawson’s brand image. Whether new media marketing, as a new marketing method,
plays a great role in Lawson’s brand-building process is what people need to study. This paper
mainly takes Lawson’s publicity methods on new media platforms, such as live broadcasting
and online animation platforms, as examples. This paper also focuses on the analysis of
Lawson Convenience Store’s disregard for new media marketing, the adverse consequences
brought by its overemphasis on animation, as well as its over-belief in the limitations of
offline marketing. To solve these problems, the suggestions given in this paper mainly include
recruiting many young people who know the Internet, communicating with large enterprises
with experience in new media marketing, and opening a new online marketing platform. After
analyzing this paper, Lawson should be able to improve the level of new media marketing,
which has a demonstration effect on the whole convenience store industry.
Keywords: network platform, animation, young customers
1. Introduction
1.1. Research Background
Lawson Convenience Store is a well-known business in Japan. The development in China is also
perfect; in order to better occupy the convenience store market, Lawson has begun to expand, trying
to establish its field in the central region of China. However, with the management of Lawson
Convenience Store for so long, does it still meet the current development stage? Nowadays, the essay
will look at Lawson Convenience store marketing strategy analysis, from what aspects to improve
the management? And what can Lawson’s marketing approach inspire other businesses?
About 27% of customers said the products of Lawson are slightly more expensive, especially
convenience products [1]. Ryohei Yoshida, general manager of Lawson’s China business promotion
headquarters, said the number of stores in China exceeded 5,000 [2]. Imagine that if Lawson used an
efficient advertising method, consumers would even be willing to spend money in Lawson
convenience stores. If the price goes up, The scale of Lawson convenience stores in China is not only
the current more than 5,000 stores. In addition, can good publicity make consumers feel that spending
a little more money in a Lawson convenience store is worthy of the value of the Lawson brand so as
not to arouse consumer anger over high prices? Is Lawson’s experience typical and universal among
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231089
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
277
many convenience store brands? Therefore, analyzing Lawson’s marketing methods can help us make
fewer detours in new media marketing methods and avoid making the same mistakes as Lawson has
made.
As people all know, the accurate positioning of convenience stores has always been to consider
the convenience requirements of customers and do a good job of the usual operation services; this
kind of thinking sometimes makes convenience stores stand still. However, the full range of technical
professional operation services is the key. Still, the novel and interesting marketing promotion
violence can maintain the mystery of customers and satisfaction and stimulate consumption impulse.
Is there such a problem at Lawson Convenience Store? How can other brands avoid it?
In recent years, more and more companies have given great importance to the operation of new
media platforms to achieve effective marketing. Looking back, although Lawson is also involved in
social network service platforms, he does not attach great importance to the role of new media, lacks
understanding, and does not understand the crucial effect of Internet media in marketing promotion
and brand culture construction.
1.2. Literature Review
While people searched for marketing strategy information for Lawson Convenience Store, some
articles were very enlightening. For example, Pu saw the link between Lawson’s marketing approach
and its international approach, especially the great success of anime, which combined marketing
strategy and localization with young people [3]. In addition, Jiang also found that Lawson attaches
great importance to brand effect, especially good at linking with popular animation such as ‘Love
Live’, and launched related derivative products such as backpacks, water bottles, etc., and then
achieved the purpose of promoting the brand through media publicity [4]. Lu also discovered the role
of new media in Lawson’s brand building, analyzed new media and big data together, and concluded
that new media played a crucial role in Lawson’s transformation of its service and sales [5]. Liao
discovered the meticulous marketing methods of Lawson Convenience Store. Lawson focuses on the
preferences and consumption concepts of young people in their 20s and actively creates an anime-
style publicity method. Lawson found that young people are a huge potential market, so in the
marketing process to cater to young people’s preferences. Lawson even promotes itself through online
games or anime [6].
As seen above, most scholars have studied popular cultural means catering to young people, such
as anime and games, in Lawson’s marketing style. Only a few people could connect Lawson’s
marketing methods and high technology, such as big data analytics. Lawson’s online marketing and
offline publicity cooperation and the effect are also few people to analyze.
1.3. Research Framework
First, this essay will outline the current situation of new media marketing at Lawson Convenience
Store. Secondly, this essay will analyze a series of problems in Lawson’s marketing. This part will
be composed of three parts: analysis of current situation, analysis of problems, and analysis of causes
of problems. Third, this article will give reasonable suggestions for the problems analyzed and give
solutions or optimization schemes according to the different problems. This section will give each
specific solution for each specific problem. Each solution strategy forms a text. At the end of the
paper, a summary of the paper’s content and the significance and enlightenment of the research will
be written down in this essay.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231089
278
2. Case Description
Lawson pays special attention to the publicity method of linkage with famous IP in the marketing
method. The collaboration of themed shops through 2D cultural IP has become one of the most
effective marketing methods for brands to appeal to young audiences. Since 2012, the company has
been developing themed convenience stores in Japan in tie-ups with Ultraman, Detective Conan,
Naruto, and Teddy.
Lawson does not like to promote himself on new media platforms such as TikTok directly. Lawson
uses the most common marketing method to link with famous animation and game IP offline, then
publish it to online social platforms. A typical example is the online promotion of a newly opened
anime co-branded store.
Lawson also launched a theme store with the animation app Bilibili. At the same time, Bilibili also
plans to cooperate with Lawson in multiple channels in 2018, giving more possibilities for cross-
brand cooperation. Among them, introducing more stores in the Jiangsu, Zhejiang, and Shanghai
areas has also been included in the plan of cooperation between the two sides [7].
It is worth noting that Lawson not only focuses on offline and online publicity but Lawson also
opened online sales services. Selling online through the Internet is also a marketing tool. Lawson was
a pioneer in this regard.
The sales of Lawson convenience stores on the online platform also maintained a high growth rate.
Taking Meituan as an example, more than 4,000 Lawson convenience stores are currently online on
the platform, covering more than 69 cities across the country, and the order volume from January
2022 to July 2022 increased by 183% compared with the same period of 2021 [2].
As of July 25, 2022, Lawson announced that the number of stores in China has exceeded 5,000.
Yoshida Ryohei revealed that by 2025, the company’s goal of opening stores in China is 10,000 [2].
Public reports show that in recent years, the proportion of convenience stores opening a related
online business has shown a gradual and steady upward trend, and the epidemic has further catalyzed
online business development. Online + offline to become the new leading model.
3. Analysis of the Problem
3.1. Lawson New Media Marketing Status Analysis
Lawson doesn’t pay enough attention to new media operations compared to offline publicity. Most
of Lawson’s marketing methods are offline publicity, such as new store opening activities,
convenience stores selling products of the brand, and so on. Lawson hasn’t used a lot of new media
to market her brand.
From a series of examples mentioned above, Lawson’s new media marketing methods are mainly
reflected in the cooperation with well-known games and animation and the launch of online games
and animation-themed activities on online platforms to market its brand. The specific ways are to
promote the newly opened theme convenience store of Lawson and popular animation interaction on
Bilibili, open an online purchasing platform, and live broadcast large-scale animation-themed
activities sponsored by Lawson. Through this series of promotional activities, Lawson Convenience
Store will certainly attract the attention of many young people who love animation culture. Lawson
is trying to create a consumer environment for anime immersive experiences and transfer this sense
of animation atmosphere to online sales. This is a major feature of Lawson’s current new media
marketing strategy.
Lawson has noticed emerging marketing strategies that combine offline and online campaigns and
the benefits of linking up with popular new things like anime and new media. Some results have been
achieved. However, on the whole, Lawson’s network marketing scale is small at this stage, and there
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231089
279
is no stable new media platform, but it only publishes videos on multiple platforms, and there is no
fixed partner. Lawson’s new media marketing is still in its infancy. This is the general state of
Lawson’s new media marketing.
3.2. Analysis of Problems in Lawson’s Marketing Methods
In recent years, more and more companies have placed great importance on operating new media
platforms to achieve effective marketing. Looking back, although Lawson is also involved in social
network service platforms, he does not attach great importance to the role of new media, lacks
understanding, and does not understand the crucial effect of Internet media in marketing promotion,
publicity planning, and brand culture construction.
Perhaps mainly because consumers buy goods directly at convenience stores, Lawson may think
that online publicity does little to expand the influence of the brand. Compared with online publicity,
expanding the number and scale of convenience stores may play a more effective role in publicity
and expanding the consumer market. But then came the problem: the rapidly expanding convenience
store scale did not bring Lawson high returns. Lawson China only reached 2,000 stores in January
2019, less than two years to achieve the goal of 1,000 new stores. However, the rapid expansion has
not masked a dilemma: Lawson has yet to make a profit in the 24 years it has been in China [8].
Focusing on just one way of doing business often leads to failure.
Lawson has also launched an online app to sell his products, and the app has registered nearly 15
million users. But Lawson went no further by developing his independent online platform to promote
himself. Lawson chose other well-known social platforms such as SNS, Weibo, WeChat, etc. As a
result, Lawson’s fans are too scattered to come together and form a cohesive force. Therefore, the
result is that the total number of fans on Weibo and WeChat is less than the number of registered
users of the Lawson app (5.2 million on Weibo and 1.11 million on WeChat). This shows that there
is a huge problem with Lawson’s Internet propaganda. In addition, nearly 9 15 million users registered
in Lawson’s app are in Jiangsu, Zhejiang, and Shanghai. Lawson still has a long way to go in
expanding its regional influence.
3.3. Cause Analysis of the Problem
As a Japanese brand, it is natural for Lawson to associate its brand with anime. The advantage of this
is that the brand image can be more distinct. But the downside is that it greatly reduces the number
of regular customers. Lawson’s target audience is naturally dominated by young people who love
animation. This greatly limits the role of Lawson’s online marketing. Second, Lawson may have put
a limited amount of money into the opening of the new store and no more money into the online
publicity, let alone the formation of a complete publicity chain.
People can see that in recent years, Lawson has also begun to try to enter the web platform.
However, as long as Lawson’s action on the network platform is to open an official app as a platform
for selling products, it is not a new media marketing platform in the real sense. This may be due to
Lawson being content to open official accounts on other social platforms. In short, Lawson’s online
marketing is now in a fragmented state. Many people believe that the real reason for this status quo
is that Lawson does not have a deep understanding of new media marketing methods and responds
too slowly to the rapid development of new media.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231089
280
4. Suggestions
4.1. Lack of Interest in New Media Marketing
The fundamental problem of Lawson’s new media marketing is that Lawson does not attach
importance to new media marketing.
From the examples above, Lawson has no interest in new media marketing. This will be the
fundamental question in Lawson’s propaganda strategy. The fundamental solution to this problem
was to make Lawson aware of the enormous potential of online media. Take TikTok, today’s largest
self-media platform, as an example. Global monthly active users are more than 500 million, daily
active users 150 million, deeply loved by young users. The number of users continues to grow, and
the user time is also gradually increasing [9]. That is to say, if you choose TikTok as a publicity
platform, you have at least 150 million potential customers. Relying only on the previous approach
to expanding the number of convenience stores neither advertised cheap on online platforms nor
advertised wide range on Tiktok. These are the things Lawson needs to pay attention to and think
about.
How to change Lawson’s backward propaganda concept is not very difficult. Lawson needs to get
many young people into Lawson’s propaganda department. Contemporary college students love the
Internet, and many are happy to use it to show their views. When these thoughtful, willing young
people are brought into the company and given the power to develop its online publicity efforts, they
will surely come up with many novel ideas and put them into practice. In this way, the company’s
online publicity achievements must be greatly improved.
Secondly, more and more enterprises are beginning to realize the great promotion role of new
media for the company’s publicity work. Many companies are using online platforms to announce
new products and ideas. Companies can do many things through online platforms that they previously
had to do separately, such as offline conferences, fan meetings, the release of new policies, and so on.
Moreover, compared with traditional offline publishing, these works can be spread quickly on the
Internet, with very high timeliness and a large audience range. That is to say, there are many
advantages and advanced nature of new media that people have not yet explored. If Lawson can
communicate and cooperate with a large number of well-known brands, he will surely realize the
great potential of new media and gain valuable experience.
4.2. Unrigid about the Identity of Anime
As a typical Japanese convenience store, Lawson has a close relationship with anime. Lawson has
also linked his brand to anime, creating an anime image of the brand. This distinctive brand image
has brought Lawson a stable group of young customers and has also attracted a lot of animation
companies willing to cooperate with Lawson. However, over-emphasizing animation features will
narrow the customer base, which is not conducive to expanding potential customers.
If Lawson wants to expand the brand’s influence, he does not need to give up the original brand
positioning - animation. Lawson can consider adding other popular factors in the new media
marketing based on animation. Combining with animation can consolidate old customers and
positively impact other customer groups. Some people think combining two seemingly unrelated
things - anime and other popular culture- is difficult, but this is overthinking. Lawson could try to
link the latest pop culture to his original anime persona, such as organizing an online campaign. Or
launch a joint gift with the brand, and set how much money consumers can get the gift and so on. As
long as Lawson is willing to make efforts to study the combination of new culture and new media, it
will certainly achieve results.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231089
281
Finally, and perhaps in a new venture, Lawson could look for a new brand positioning while
promoting online. New media marketing is new, and there are many new ideas from the new media
operation. Lawson can choose to build an online sales platform using the Internet platform. Lawson
is likely to become the new online shopping platform. Lawson can even open up new ideas and build
the online marketing platform into a pop culture platform with a completely different style from the
original animation culture, which will enrich the brand image of Lawson and break consumers’
inherent concept of Lawson, but will not hurt the original animation brand image. In this scenario,
the Lawson sales platform on the Internet is a new brand image, which can be based on other popular
cultures, such as hip-hop culture. It’s like Disney, dominated by fantasy style, has a Marvel franchise
subsidiary dominated by hero culture. Marvel is a subsidiary of Disney, which produces most of its
films [10]. Lawson’s web platform could take a new path, emulating Disney’s Marvel franchise.
Thinking outside the box is always good for brands.
In summary, Lawson’s biggest obstacle to reaching new potential users is sticking to the anime
image. The way to break this image is not to abandon the original brand positioning but to establish
a new online image that does not conflict with the original brand positioning.
5. Conclusion
5.1. Key Finding
In summary, Lawson’s main problem is that he does not pay attention to the publicity role of network
platforms and is too rigid in the brand image of animation. Lawson can solve these problems by
recruiting a large number of Internet-savvy young people, communicating with other well-known
brands, and opening accounts on new online platforms.
5.2. Research Significance
This article finds the reason for Lawson’s long history of low publicity and offers a new approach for
other brands with similar problems.
In addition, Lawson Convenience Store is one of the most developed and well-known big brands
among convenience store brands. However, as a first-class convenience store brand like Lawson lags
behind well-known brands in new media marketing, the convenience store industry generally has
problems like Lawson. The study of Lawson’s Internet marketing of convenience store brands
positively impacts promoting the convenience store brand effect.
References
[1] Lawson convenience store marketing strategy analysis, from what aspects to improve, 2022.08.04, 2023.08.17,
https://www.yoshu.com.cn/news/show-4104.html
[2] Jiang, Y. M.(2022). Experience and enlightenment of Lawson Convenience Store’s business strategy in China for
Chinese enterprises. Science and technology Innovation and Productivity (05),30-32.
[3] Pu, J. X.(2017). Research on International Business Strategy of Lawson Convenience Store in Japan (Master’s
Thesis, Heilongjiang University).
[4] Wang, Y. F.(2020). Exploring the personalized marketing strategy of convenience stores under the background of
new retail. Special Zone Economy (07),148-150.
[5] Lou, F. F. (2022). Research on ZJ Convenience Store Service Marketing Strategy (Master’s Thesis, Yunnan
University of Finance and Economics).
[6] He, Y. M. (2018). Creating the value of Lawson Scene: Enabling, Platform and Cooperation. Shanghai Commerce
(03),19-20
[7] Sun, D. G. (2016). Research on Retailer Reverse Marketing channel Optimization (Ph. D. Dissertation, Tianjin
University of Finance and Economics).
[8] Li, R. Y. & Tang, L. C. (2022). Exploring the brand design Value of chain convenience stores -- A case study of
Lawson Convenience Store in Haikou City. China Packaging (09),65-67.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231089
282
[9] Liu, B. T. (2018). On the development status of Lawson Convenience Store in China. Chinese and Foreign
entrepreneurs (16),22.
[10] Li, Y., Xing, W. L., Cao, H., Wang, F. Y. & Wang, S. Y. (2018). Research on the new convenience store problem
from a comparative perspective: A case study of Lawson and Su Guo. Market Week (10),33-34.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231089
283
Stock Price Prediction for Technology Company
Yuchen Wang1,a,*
1Graduate School of Arts and Sciences, Columbia University, New York, US
a. yw3890@columbia.edu
*corresponding author
Abstract: Individuals aim to develop accurate models for stock prices to make informed
decisions as investors, so they can determine opportune moments to purchase and sell stocks
for maximizing profits. This paper select Apple stock from yahoo finance range from Aug
1st 2013 to Aug 1st 2023, and then forecasting it’s future 30 days stock price. This study
contain four models, which are XGboost, linear regression, K-Nearest Neighbors (KNN) and
Long Short-Term Memory (LSTM). Those models are all fit the train and test data and then
draw a visualization plot. For selecting the best model, this paper use root mean squared
error(RMSE) metrics and mean absolute percentage error(MAPE) and got the best model are
Linear Regression and LSTM. Depending on the mechanism of four models, LSTM can
treated as the best one to predict future stock price. In future study, researchers can use LSTM
model more to predict stock price for other companies in order to get the best result.
Keywords: LSTM, prediction, RMSE, visualization
1. Introduction
Stock market prediction is important. It offers a range of advantages for investors and businesses
alike. These benefits include the potential for profitable trading decisions, effective risk management
by anticipating market downturns, informed strategic planning, and optimized portfolio construction.
Accurate predictions also enable a competitive edge, drive research and development of innovative
strategies, provide insights into market sentiment, and inform long-term financial goals.
Technology company’s stock price is sensitive to many factors like constant innovation, uncertain
profit prospects, susceptibility to market expectations and sentiment and so on. So how to predict
their stock price is the essential part in investment activity.
In the past, many researchers using single machine learning method like decision tree [1-3], linear
regression [4-7], support machine vector [7-10] and so on to predict the stock price. There prediction
is easy, but it might not perform very accurately for prediction part. Thus, this paper uses XGboost,
KNN, linear regression and LSTM to analysis the issue so as to fill the potential research gap.
2. Data
The data collected for this research is from Yahoo finance. The Stock data is “AAPL”. The reason
for the selection is shown below. Apple stands as a renowned global technology powerhouse,
celebrated for its groundbreaking hardware and software offerings. Among its array of products are
the iconic iPhone, iPad, Mac, and Apple Watch, complemented by an array of software and services.
Apple also conducts research and development in fields such as artificial intelligence, augmented
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231103
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
284
reality, and autonomous driving. The data range is from August 1st, 2013, to August 1st, 2023, which
is a ten-year range. The reason for this range is that using long range data can reduce the percentage
of outlier, making the price prediction more accurately. For the summary chart, close price is the only
value the study need. The mean closing price is 71.00. The min price is 16.08, and the max price is
196.45, standard deviation is 52.92, 25%, 50%, 75% quartile are 28.48, 44.52 and 125.30 (See Table
1).
Table 1: Summary Statistics for AAPL Closing Price.
Summary Statistics
Close
1
count
2516.0000
2
mean
71.003268
3
std
52.920136
4
min
16.075714
5
25%
28.479375
6
50%
44.517500
7
75%
125.297499
8
max
196.449997
From the data visualization plot, the information suggests that the min price is in 2014 and the max
price is in 2023. This infer that AAPL is developed dramatically in the past ten year as the closing
price is rising (See Figure 1).
Figure 1: Closing Price for AAPL.
3. Method
3.1. XGboost
XGBoost is based on the concept of gradient boosting, which involves sequentially training a series
of weak learners like decision trees and combining their predictions to improve the overall model's
accuracy. XGBoost employs an efficient gradient optimization algorithm to minimize the loss
function during the training process. This optimization approach accelerates convergence and
enhances training efficiency.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231103
285
XGBoost is often used for predicting stock prices due to its ability to handle complex relationships,
capture non-linearity, and effectively incorporate various features that influence stock price
movements. Its ensemble learning approach, regularization techniques, and handling of missing data
make it suitable for the dynamic and noisy nature of financial data.
The steps for XGBoost for stock price prediction:
Data Splitting: Divide the dataset into three distinct sets: training, validation, and testing.
Model Selection and Training: use hyperparameters to tune the model in order to optimize model
performance. Common parameters include learning rate, maximum depth, number of boosting rounds,
and regularization terms. Utilize the training data to train the XGBoost model, followed by validation
using the validation dataset to assess its performance.
Hyperparameter Tuning: hyperparmaters have many combinations. In order to find the best one,
the study need to use some functions like grid search or random search.
Evaluation: Use some metrics like RMSE. Those can assess whether the model is fitted.
3.2. KNN
The KNN algorithm is a straightforward and intuitive machine learning technique employed for tasks
such as classification and regression, even extending to predicting stock prices. KNN's predictions
hinge on how closely input data points resemble established labeled data points present in a training
dataset. KNN is a relatively simple algorithm to understand and implement. It doesn't involve
complex mathematical concepts or intricate hyperparameter tuning. And it has the ability to determine
whether the data relationships are linear. Additionally, KNN have the ability potentially adapt to
changing market conditions as it considers recent similar data points for predictions.
Here are steps in predicting the stock price:
Prepare the dataset with features and the target variable. Then make the dataset separated to train
set and test set.
Choose one of the K values that can represents the neighbors’ number that will be considered when
making predictions. This value is typically chosen through cross-validation.
Identify one of the training points, which has the smallest range to the testing point.
For regression purposes, compute the mean of the target values belonging to the KNN. In the case
of classification tasks, identify the most frequent class within the KNN.
By applying some metric like RMSE, the study can assess the model performance.
3.3. Linear Regression
Linear regression is the tool to many things like predicting the stock price because of its simplicity,
ability to identify trends, and interpretability. It serves as a baseline model and provides insights into
the relationships between predictor variables and stock prices.
Using linear regression for predicting stock prices is another approach, though it comes with its
own set of considerations and challenges. Linear regression assumes a linear relationship between the
input features and the target variable, which in this case would be the stock price. Here's how it might
approach using linear regression for stock price prediction:
Here are steps in predicting the stock price:
Separating the Data: Dataset is separated to training, testing and validation sets to train the study,
tune hyperparameters, and evaluate performance.
Model Selection and Training: choose linear regression as the modeling technique. Train the model
using the training data. This technique can lower the error for the predicted VS actual values.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231103
286
Feature Selection: given that linear regression assumes a linear relationship, it's important to select
features that are more likely to exhibit linear correlations with the target. Feature selection techniques
can help in choosing the most relevant features.
Evaluation: assess the model's performance using appropriate evaluation metrics like MAE, MSE,
RMSE and R-squared.
Prediction: apply the trained linear regression in order to have a accurate predicted result in the
future data.
3.4. LSTM
LSTMs are effective for dealing with sequential data like stock prices because it can capture long-
range dependencies, allowing them to consider historical data's influence on future prices. Also, it
can identify and adapt to both short-term fluctuations and long-term trends. Lastly, it trained on a
large dataset can capture general financial market patterns for specific prediction tasks.
The LSTM cell has several components in the whole working phase:
1. The input gate (i_t) assesses the importance of incoming data in relation to the existing cell state.
2. The forget gate (f_t) decides which information in the previous cell state should be discarded or
ignored.
3. During the cell state update (g_t) phase, new potential values for the cell state are calculated
based on the input.
4. The cell state (C_t) is then modified using the combined influence of the input gate, forget gate,
and cell state update.
5. The output gate (o_t) determines how much the current cell state contributes to the final output
of the LSTM.
6. The hidden state (h_t) is computed using the output gate and the updated cell state, resulting in
the new representation of the LSTM cell (Details are shown in Figure 2).
Figure 2: LSTM Model.
4. Result
4.1. XGboost
In XGboost model, ['High', 'Low', 'Open', 'Volume'] is the input variable, and the [‘Close’] is the
target variable. Then the study split the data to 80 percent of training data and 20 percent of testing
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231103
287
data. Afterwards, the training process involves utilizing the XGBRegressor function. This involves
fitting the model using the training data, allowing the algorithm to learn from the provided
information. Subsequently, the trained model is used to make predictions on the test data, enabling it
to generalize its learned patterns to new, unseen data. The prediction is stored as y_pred. To check
the model accuracy, the study uses rmse. And the rmse is 39.30 (See Figure 3).
Figure 3: Actual Close Price VS Predicted Close Price in XGboost Model.
4.2. KNN
The dataset is separated from the feature variables ('High', 'Low', 'Open', 'Volume') into X and the
target variable ('Close') into y. Then the study defined a TimeSeriesSplit object to perform time series
cross-validation with two splits, using the StandardScaler to standardize the feature variables in both
the training and testing sets. By performing hyperparameter tuning using cross-validation with the
number of neighbors (K) as the hyperparameter, the study plotted the error (mean squared error) for
different values of K to help us find the best K value. After that, GridSearchCV can be used, and the
best K value is 46. Then the best_knn model is using this value, and then the study fit the best_knn
model on the scaled training data. Then the best_knn is used to predict our scaled data. The best_knn
model plotted a graph representing actual values VS predicted value for test data. After that, the
RMSE is calculated to be 23.25 (See Figure 4).
Figure 4: Actual Close Price VS Predicted Close Price for KNN Model.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231103
288
4.3. Linear Regression
For data preprocessing, the future price is calculated using historical price data and stored it in the
data, then it dropped the last 30 rows as the data don't have future data to validate these predictions.
In linear regression, the study makes ['High', 'Low', 'Open', 'Volume'] to input variable and ['Predicted']
as target variable. Then the data is split to train data and test data then standardized it. After that, in
order to train the model, the model used the function LinearRegression() to instantiate the model, and
it trained the model using scaled data. For prediction, the "model.predict()" function takes the scaled
test set features 'X_test_scaled' as input and returns the predicted target variable values (in this case,
future prices) based on the trained model's learned patterns. The graph for predicted value vs actual
value is plotted. The root mean squared error for 15.21 (See Figure 5).
Figure 5: Actual Close Price VS Predicted Close Price for Linear Regression Model.
4.4. LSTM
LSTM has many steps within the model. The first step is same as previous mode for splitting the data,
then use function to convert data to LSTM format. Generally, the model use time steps equal to 100.
Next, the model is defined with three layers and one dense layer. The optimizer was set as 'Adam',
the loss function was set as 'mean_squared_error', and the evaluation metric was set as Mean Absolute
Percentage Error (MAPE). In this training, the epochs equal to 10 and the batch size equal to 32 and
the training progress is displayed (verbose=1). The val_mape value for the last epoch is 5.9502, which
is an acceptable value. After that, the model is iterated over the parameters to find the best model.
Next, the best model can use to predict the X_train and X_test by using predict function and then
transform back to the original form. The RMSE for y_train and y_test is 60.84 and 160.69. For AAPL
stock prediction, a visualization plot is made. The train predict price and the test predict price has
little difference from the actual price but not a lot (See Figure 6).
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231103
289
Figure 6: Actual Price VS Predicted Price for LSTM.
5. Conclusion
Among the models employed, both LSTM and linear regression exhibited superior performance in
terms of accuracy. Linear regression showcased commendable predictive power, displaying a lower
value for root mean squared error. This translated to a remarkable accuracy level in the outcomes. In
contrast, the K-Nearest Neighbors (KNN) and XGboost model yielded a comparatively higher root
mean squared error than Linear regression, thereby falling short in accuracy. XGboost, on the other
hand, demonstrated the weakest performance, yielding a high root mean squared error of 15.21 and
an underwhelming return.
This study face limitations including their struggle with non-linear and non-stationary stock data,
the challenge of incorporating market sentiment and news, susceptibility to overfitting due to limited
data, inability to grasp causality, and difficulty adapting to abrupt market shifts.
References
[1] Wang, Y., and Guo, Y. (2020). Forecasting method of stock market volatility in time series data based on mixed
model of ARIMA and XGBoost. China Communications, 17(3), 205-221.
[2] Vuong, P. H., Dat, T. T., Mai, T. K., and Uyen, P. H. (2022). Stock-price forecasting based on XGBoost and LSTM.
Computer Systems Science & Engineering, 40(1).
[3] Han, Y., Kim, J., & Enke, D. (2023). A machine learning trading system for the stock market based on N-period
Min-Max labeling using XGBoost. Expert Systems with Applications, 211, 118581.
[4] Stock Prediction Using Linear Regression. (2020) Retrieved from https://medium.com/analytics-vidhya/stock-
prediction-using-linear-regression-cd1d8351f536
[5] Panwar, B., Dhuriya, G., Johri, P., Yadav, S. S., and Gaur, N. (2021). Stock market prediction using linear
regression and SVM. In 2021 International Conference on Advance Computing and Innovative Technologies in
Engineering, 629-631.
[6] Gharehchopogh, F. S., Bonab, T. H., and Khaze, S. R. (2013). A linear regression approach to prediction of stock
market trading volume: a case study. International Journal of Managing Value and Supply Chains, 4(3), 25.
[7] Ravikumar, S., and Saraf, P. (2020). Prediction of stock prices using machine learning (regression, classification)
Algorithms. In 2020 International Conference for Emerging Technology, 1-5.
[8] Kurani, A., Doshi, P., Vakharia, A., and Shah, M. (2023). A comprehensive comparative study of artificial neural
network (ANN) and support vector machines (SVM) on stock forecasting. Annals of Data Science, 10(1), 183-208.
[9] Sheth, D., and Shah, M. (2023). Predicting stock market using machine learning: best and accurate way to know
future stock prices. International Journal of System Assurance Engineering and Management, 14(1), 1-18.
[10] Panwar, B., Dhuriya, G., Johri, P., Yadav, S. S., and Gaur, N. (2021). Stock market prediction using linear
regression and SVM. In 2021 International Conference on Advance Computing and Innovative Technologies in
Engineering, 629-631.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231103
290
Applicatoin of TTCC Algorithm in House Reallocation
Market in China
Hanyu Xu1, a, *
1St. Andrews College, Cambridge, England, United Kindom
a. hanyu.xu@standrewscambridge.co.uk
*corresponding author
Abstract: In recent years, many low-income families have been suffering house allocation
problem. The current welfare house allocation mechanism is of low effiency. Despite Chinese
government has been working on improving the current mechanism to solve this proplem,
there are still many people which try to cheat in the house allocation market. Most of the
people could not even truly express their preference through the current mechanism. It is
important to find a more suitable mechanism which can help low-income familes reallocate
their welfare housing. This article tries to make anaysis about applying Top Trading Cycles
and Chians mechansim (TTCC) in the house reallocation market. This mechanism can lead
to a more efficient result. Besides, this article expalins why TTCC is better than the current
mechanism by studing the feature of the result from TTCC. The results show that TTCC can
be used to in house reallocation market with a strcit ranking among low-come familes and
there requires a strcit background investigation organization as well.
Keywords: TTCC mechanism, house lottery mechnaism, waitlist option, chain selection rule
1. Introduction
As the urban area in China keeps increasing, more and more families from rural area have moved to
the cities and the growth of the population in urban area has caused many problems. The demand of
living house is much higher than the actual number of the available ones in China. The price of one
commodity house in the central city can sometimes exceed one billion. Consequently, instead of
buying house in the urban area, 33.9% of the low-income families choose to live in rental houses [1].
However, the rent is too high that most of those families could not afford it at all.
To solve this problem, Chinese government has built up many new houses since 2015 [2]. Those
houses are welfare homes for those low-income families. They can rent those houses at low prices.
This plan has brought significant effect to the low-income population living issue. Until now, with
the help of welfare housing, 88.4% of the low-income families only pay 300 yuan per month [1]. The
rent is far below the market price in central city.
Nevertheless, after the development of those low-income families in these years, the current houses
are not suitable for those people anymore for many reasons, like changing in workplaces or the
increasing number of family members [3]. The Ministry of Construction in China has announced that
Chinese government would not plan to add the number of welfare housing [2]. Consequently, it is
important to reallocate the welfare housing and study for a suitable house allocation mechanism.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231204
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
291
Compared with the existing mechanism (which is a house lottery) [4], Top Trading Cycles and
Chains (TTCC) mechanism, which is a variant of Top Trading Cycles (TTC) by David Gale can lead
to a Pareto efficient, strategy-proof, and individual rational result [5]. The application of this
mechanism is based on a strict preference, the house reallocation can satisfy this demand.
This article focuses on making descriptions of the current house allocation mechanism in China
(house lottery), and application of TTCC in the house reallocation market with an example.
2. House Lottery Mechanism
The most cities in China chose to apply house lottery mechanism in the first round of house allocation.
Every family which is in the lottery mechanism will be allocated a randomly welfare house. Many
problems have been caused by using this mechanism. In China, lottery mechanism can be used in
many markets such the lottery in the welfare housing allocation market [6], school district housing
market [7] and it is also applied as a policy for primary school students entering [8].
Overall, this mechanism ignores the preference of low-income families. The position of the rental
housing was too far to their workplace. For some families, the area of the house is too small which
lead to a result that many families could not live in it at all. For example, the occupancy rate in
XiangLin ShiJi HuaFu, a living community, in Shenzhen was only 40% when it was allocated as a
welfare house [2]. Some children in those families were forced to transfer to another school because
of the lottery mechanism. Lottery mechanism is of low efficient, and it is hardly to satisfy the demand
of mostly low-income families.
3. TTCC Mechanism
Top Trading Cycles and Chains mechanism (TTCC) which was proposed by Roth, Sonmez, and
Unver in 2004, was applied in the kidney exchange market. Briefly, TTCC mechanism is the
extension of TTC mechanism. The differences between TTCC and TTC are the implement of TTCC
is adding a specific chain selection rule and ‘waitlist option’ when applying TTC algorithm.
Housing allocation problem and kidney exchange problem are regarded as two different problems
[9]. Every patient in the kidney exchange market would never escape from the mechanism because
they only survive with at least one healthy kidney which means every patient would stay in TTCC
mechanism until they receive a suitable kidney. However, in the house reallocation problem, these
low-income families have the same feature because they are living a non-satisfactory condition and
they always want to move to better and more suitable houses. Even the house is a welfare one, those
low-income families still need to pay for the rent. Consequently, comparing with escaping from the
algorithm, those low-income families prefer to choose to keep their house in trading, stay in the
‘waitlist option’ as they would not reject to have a better life. Further, in the normal house market,
there is no role for money trading between the low-income families. The supplier of the welfare
housing is Chinese government.
Before starting to explain the application of the TTCC mechanism officially, it is important to
introduce ‘waitlist option’ and the chain selection rule firstly which can facilitate the understanding
of TTCC mechanism.
3.1. Waitlist Option (W-Chain)
Waitlist option (w-chain) is an ordered chain with a starting house and an ending tenant. The starting
house always point to their current owner and the owner can point to their best choice house according
to a strict preference list. In TTC mechanism only exist complete cycles because every tenant will
escape from the mechanism easily if they find all other houses are unacceptable. However, in the w-
chain if there are no acceptable house for the ending tenant. Instead of escaping from the algorithm,
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231204
292
she will ask waitlist option for help. Compared with the cycle, a pair (hm, tm) can be in the part of over
one w-chains. A pair can only be part of one cycle because every cycle will be removed from the
algorithm immediately when it is formed. Besides, the length of an ordered chain is not always fixed,
a w-chain can be added into the tail of another w-chain.
In Example 1, every tenant has been assigned a house which has the same index. A list of the w-
chain in Figure 1 has been shown below, with W1= (h3, t3), W2= (h2, t2, h3, t3), W3= (h1, t1, h2, t2, h3,
t3), W4= (h4, t4, h2, t2, h3, t3), and W5= (h5, t5, h3, t3).
Figure 1: W-chain in Example 1.
3.2. The Chain Selcetion Rule
It is clearly that there are more than one w-chains in Example 1. Each w-chain can intersect with
others. How to choose an efficiently w-chain from these figures has been a problem.
The chain selection rule that will be applied in this mechanism is choosing a maximal w-chain. In
this way it can benefit the most people at the same time. If there are more than one the longest chain.
This mechanism will choose the one with higher priority tenant inside. If the highest priority tenant
is in more than one the longest chain. The choose the second longest chain. This step is repeated until
complete finding the chain, and every w-chain is the shortest. Those chains can be combined together
as the tail of w-chain is flexible, considering preference list. Nevertheless, when this situation still
happens after this step, each tenant will point to the waitlist option waiting for the next round of
allocation. The selected w-chain will be kept until the last step of the algorithm is completed. All the
pair inside of the chain is going to be assigned the house they point because all the houses that the
tenant point to are acceptable for them. In the end, only the head of the w-chain point at the waitlist
option, and they will own the highest priority in the next round of house reallocation.
3.3. The Application of TTCC Mechanism
Step 1: Firstly, all the remaining houses always point at their paired tenants. Tenants can point at the
houses which they preferred.
Step 2:
Step 2-1: Every tenant can point at their most preferred houses, so there are at least one cycle or
chain exist [7].
Step 2-2: If there is a cycle formed, then keep it. Whenever a cycle has formed, all the house and
tenants are removed in the cycle and every point at their most preferred house again.
Step 2-3: Step 2-1 and 2-2 are repeated until there are no further chains available.
Step 3:
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231204
293
Step 3-1: If there is no any other house and tenant left anymore, then the algorithm end (everyone
has assigned an acceptable welfare house).
Step 3-2: If there are some chains exist at Step 2-1, the algorithm then can apply the chain selection
rule to select the most suitable chain.
The chain is going to be removed from all the w-chains for the moment until the last step of the
algorithm. The selected chain is not fixed, the head of this chain do not have to point to w when it is
selected. Other pairs can be added to the tail of the selected chain.
Step 4: Whenever a chain has been chosen, another cycle is formed in the meantime [10]. Then
agents can repeat Step 2 and 3. At the end of this step, there is no any pair remaining. The head of the
selected chain should be assigned the waitlist option and the algorithm end.
3.4. A Modelling Example by Using TTCC Mechanism
Before beginning the explanation about the Example 2, the TTCC mechanism must base on a strict
preference. Table 1 shows a preference list for all the tenant in this example. Assuming that the tenant
with the smallest has the highest priority. From the left to the right side, with the decreasing of the
index of tenant, the priority is decreasing as well.
Table 1: The preference list of the tenant.
t1
t2
t3
t4
t5
t6
t7
t8
h3
h1
h1
h6
h2
h1
h4
h3
h2
h3
h8
h5
h4
h3
h2
h1
h4
h8
h5
h7
h8
w
h1
h2
h7
h3
h5
h4
h5
h1
h6
h3
h8
h2
h5
h6
h8
w
Firstly, every tenant can point at their preferred house, and both carry a paired-houses which is
used to be traded. The paired house always points at their current owner.
Figure 2: The first round in Example 2.
According to the Figure 2, there is a cycle been formed which is C1= (h1, t1, h3, t3). The cycle then
should be erased in this figure and t1 receive h3, t3 receive h1, and then Round 1 ends.
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231204
294
Then, the process goes to the second round, that every non-removed point at the remaining house
which they preferred the most.
According to Figure 3, there is a new cycle which is C2= (h2, t2, h8, t8). Then, the same step in
round one is repeated, removing the whole cycle and assign h8 to t2, h2 to t8, then round 2 ends.
Figure 3: The second round in Example 2.
In the third round, there is no cycle anymore and according to the Figure 4, two chains have been
formed. The first one is W1= (h5, t5, h4, t4, h6, t6) and the second one is W2= (h7, t7, h4, t4, h6, t6)
Now the algorithm starts the chain selection rule. W1 and W2 are both the longest chain in the
Figure 4. Compare with t5 and t7, t5 has the smaller index. Consequently, W1 should be removed from
the algorithm for a moment.
The remaining pair (h7, t7) can be added into the tail of W1. According to the Table 2, h5 is
acceptable to t7 and round 3 end. Now W1 is the only chain in the whole algorithm.
Figure 4: The third round in Example 2.
According to the Figure 5 and Table 2 below, the mechanism can allocate h6 to t4, h4 to t5, h5 to t7
there is no more acceptable house for t6, then t6 must point to w and ask the waitlist option for help,
which means waiting for the next round of house reallocation and the algorithm ends.
In the application of TTCC in the reallocation of welfare housing market. There first need to be a
strictly supervisory institution. Some of the low-income families might not need to change there
currently living places like others. Just for moving to a better place. Then they ask for joining the
algorithm. Second, for all the tenants who are already in the reallocation plan list. Chinses government
need to rank them from high to low. For example, the family with the most membership can be
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231204
295
assigned the smallest index. Only with this rank list, during the application of TTCC mechanism, the
chain selection rule can be carried out.
Figure 5: The last round in the Example 2.
Table 2: Matching results in the end.
t1
t2
t3
t4
t5
t6
t7
t8
h3
h8
h1
h6
h4
w
h5
h2
4. Conclusion
This article first explains briefly about the house lottery mechanism, then making a briefly description
about the TTCC mechanism. Compare the differences between these two mechanisms, it is clear to
see that TTCC mechanism is much better than another one. Many families which are not poor at all
try to save money and receive the best welfare housing by giving money to the government employer.
It is difficult to satisfy the demand of most tenants. TTCC mechanism can lead to a result with many
advantages. The output of the TTCC mechanism has been proved to be Pareto efficient, strategy-
proof, and individual rational. After explaining the reason of applying TTCC mechanism in the house
reallocation problem, the paper focus on the detailed of TTCC. The detailed description of TTCC is
split into three parts. The first part is the waitlist option in TTCC, which can help those families stay
in the mechanism, giving them hope to receive a more suitable housing. The second part is explaining
the chain selection rule. There is a specific chain selection rule in this problem, and it is aimed to
benefit the most people in the shortest time. With these two parts, it is easier to understand how TTCC
works.
There are limitations about using TTCC mechanism. On the one hand, like TTC mechanism, it
must need a strict preference when carrying out. On the other hand, people cannot lie in the
mechanism at all. Every tenant must stay honest. Chinese government has to examine whether those
families are literally living a hard life like they described. In the future, how to rank the priority in
the chain selection rule for those tenants can be studied, and there are many standard need to be
showed in the future and which kind of low-income people can join this mechanism. It also needs a
strict standard.
References
[1] Liao, A. (2018). The Housing’Status and Problems of Low-income Families -- An Analysis Based on the Monitoring
Data of the Economic Status of Low-Income Families in 2017, China Academic Journal Electronic Publishing
House, 1006-2815, 04-0078-08
[2] Zhao, F. (2016). Study on Reallocation of Public Rental Housing- Based on Matching Mechanism, Centrel China
Normal Unibersity
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231204
296
[3] Liu, Y. (2014). Analysis on Distribution of Affordable Housing Situation,Problems and Countermeasures, Journal
of North China University of Science and Technology(Social Scienc.,e Edition), 14(03),5-8
[4] Chen, J. and Bi, J. (2012). Discussion on the Distribution of Affordable Housing in China, Economic Research
Guide, (02),96-97
[5] Krishna, A., and Wang, Y. (2007). The relationship between top trading cycles mechanism and top trading cycles
and chains mechanism. Journal of Economic Theory, 132(1), 539-547.
[6] Liu, F., (2011). Ranking Model of Low-renting Public Housing Match Rent Queuing and Application, Shanxi
Architecture, 37(04),246-247
[7] Zhang, X. (2018). Negative Influence of “Lottery”Policy for Pupils from Primary School to Go to Junior Secondary
School and Improvement, Journal of Teaching and Managemen, (06) 8-9t
[8] Li,S. and Lan, Q. (2013). Comparison and Revelation of License Plate “Lottery”Policy and the Auction Mechanism,
27(03),47-51+59
[9] A. Abdulkadiroglu and T. Sonmez. (1999). House allocation with existing tenants, Journal of Economics Theory 88,
233–260.
[10] Roth, A.E., Sonmez, T. and Unver, M. U. (2004). Kidney exchange, The Quarterly Journal of Economics. 119 (2),
457–488
Proceedings of the 2nd International Conference on Financial Technology and Business Analysis
DOI: 10.54254/2754-1169/56/20231204
297
