The financial contagion effects of the global covid 19 pandemic evidence from fintech and traditional financial markets

The financial contagion effects of the global covid 19 pandemic evidence from fintech and traditional financial markets The financial contagion effects of the global covid 19 pandemic evidence from fintech and traditional financial markets The financial contagion effects of the global covid 19 pandemic evidence from fintech and traditional financial markets The financial contagion effects of the global covid 19 pandemic evidence from fintech and traditional financial markets The financial contagion effects of the global covid 19 pandemic evidence from fintech and traditional financial markets The financial contagion effects of the global covid 19 pandemic evidence from fintech and traditional financial markets

i

STATEMENT OF AUTHORSHIP

I declare that this submission is my work and, except where due reference ismade; this dissertation contains no material previously published or written byanother person(s)

This dissertation does not contain material extracted in the whole or partfrom the dissertation or report presented for another degree or diploma in University

of Economics Ho Chi Minh city or any other educational institution

January 2023

Pham Thi Ngoc Dung

First of all, I would like to express my best gratitude to Assoc.Prof NguyenKhac Quoc Bao and Dr Le Dat Chi, who helped me with their valuable mentorshipduring my dissertation stages for many years at the University of Economics Ho ChiMinh City I also would like to thank all the support from my colleagues at theUniversity of Economics Ho Chi Minh City during my study

I would especially like to thank my faculty (Faculty of Finance andBanking) and my university (Ton Duc Thang University) for providing me with anopportunity to pursue a Ph.D In addition, I also express my gratefulness to all of

my colleagues in my faculty who support my work at Ton Duc Thang Universitywhen I conduct my dissertation

Most importantly, this dissertation is dedicated to my family They help me

in providing encouragement as I work to complete my dissertation

Pham Thi Ngoc Dung

TABLE OF CONTENT

ACKNOWLEDGMENT iii

TABLE OF CONTENT iv

LIST OF ABBREVIATIONS vii

LIST OF TABLES viii

LIST OF FIGURES ix

ABSTRACT OF THE THESIS x

CHAPTER 1: INTRODUCTION 1

1.1 Research Background 1

1.2 Research context 3

1.2.1 The adoption of cryptocurrency in Asia emerging countries 3

1.2.2 The impact of the global Covid-19 pandemic on the financial market 6

1.3 Research objectives and scope 11

1.3.1 Research objectives 11

1.3.2 Research scope 12

1.4 The research questions 12

1.5 Methodology 13

1.6 The research contributions 14

1.6.1 Theory contribution 14

1.6.2 Practice implications 18

1.7 Dissertation structure 19

CHAPTER 2: LITERATURE REVIEW 20

2.1 Definitions 20

2.1.1 Definitions of hedge and safe haven properties 20

2.1.2 Contagion Definitions 21

2.2 Theoretical literature on the international propagation of shocks 28

2.2.1 Non-crisis-contingent theories 29

2.2.2 Crisis-Contingent Theories 30

2.3 Literature on the connectedness between Bitcoin, altcoins, and other financial markets 34

2.3.1 The Interdependence between Bitcoin, altcoins, and traditional financial markets 35

2.3.2 The contagion effect between Bitcoin, altcoins, and traditional financial markets 39

2.3.3 The Impact of Covid on the Relationship Between Stock Market and Cryptocurrency Market 43

2.4 The Research Gap 47

2.5 Hypothesis development 54

2.6 Conceptual framework 62

2.7 Conclusions 63

CHAPTER 3: METHODOLOGY 65

3.1 Data set 65

3.2 Identification of the Turmoil Period 67

3.3 Models for testing the hypotheses 69

3.3.1 Hypotheses tested for the safe haven properties of Bitcoin and Altcoins 70 3.3.2 Hypothesis tested for the contagion effect between the the cryptocurrency market and the the stock market 73

3.3.3 Hypothesis tested for the transmission mechanism of the contagion effect between the the cryptocurrency market and the the stock market 80

CHAPTER 4: EMPIRICAL RESULTS AND DISCUSSION 83

4.1 The hedge and safe haven properties of cryptocurrencies during the Covid-19 pandemic 84

4.1.1 The hedge and safe haven properties of Bitcoin 86

4.1.2 The hedge and safe haven properties of altcoins 89

4.2 Empirical results on the contagion effect 96

4.2.1 Contagion effect in terms of a change in time-varying correlation 96

4.2.2 Contagion effect in terms of a change in mean spillover and volatility spillover 99

4.3 The transmission mechanism of the contagion effect between the stock market and the cryptocurrency market during the Covid-19 period 113

4.4 Discussion 119

CHAPTER 5: CONCLUSION 129

5.1 Conclusions 129

5.2 Key contributions 132

5.3 Implications 133

5.3.1 Implications for Investors 133

5.3.2 Implications for Policy Markers 136

5.4 Limitations and recommendations for further research 141

PUBLICATION 143

REFERENCES 144

LIST OF ABBREVIATIONS

Altcoins Alternative coins

ARDL Auto Regressive Distributed-Lag Model

COVID-19 The coronavirus disease

GARCH Generalized AutoRegressive Conditional Heteroskedasticity

Fintech Financial technology

LIST OF TABLES

Table 1.1: The 2021 Global Crypto Adoption Index - Top 15 countries………… 5

Table 2.1: A summary of empirical studies examining the contagion effect between the bitcoin / altcoin market and the traditional financial market……… 50

Table 3.1 : Covid-19 arrival date……… ……… …… 67

Table 3.2: Data Description……… ……… ………… 69

Table 4.1: Descriptive statistics of Return Series (Full sample) ……….….83

Table 4.2: Testing for Safe Haven Properties of Bitcoin……… ……87

Table 4.3: Testing for safe haven properties of Ethereum……… … 90

Table 4.4: T-test for DCC of Bitcoin-Stock pair……… ………… 97

Table 4.5: T-test for DCC of Ethereum-Stock pair……… …98

Table 4.6: Parameter estimates for the mean and variance equations: Full period_Dummy: DCOV……… ……… ……….….101

Table 4.7: Parameter estimates for the mean and variance equations: Covid-19 period_Dummy: DPI……… ……… ……… …103

Table 4.8: Parameter estimates for the mean and variance equations: Covid-19 period_Dummy: DMHI……… ……… ……… 105

Table 4.9: Parameter estimates for the mean and variance equations: Covid-19 period_Dummy: DSI……… ……….106

Table 4.10: Summary of the mean contagion effect……… …… 107

Table 4.11: Wald test of restrictions on spillover parameter: Covid-19 period and high panic period……… ……… ……….110

Table 4.11(const): Wald test of restrictions on spillover parameter: High media hype period and negative sentiment period……… ……… 111

Table 4.12: T-test for the contagion channel between Bitcoin market and Stock market……… ……… ……… ……… 114

Table 4.13: T-test for the contagion channel between Ethereum market and Stock market……… ……… ……… …117

LIST OF FIGURES

Figure 1.1: Global Crypto Adoption Index………4Figure 1.2: Emerging Asian stock price return over the period 2016-2021.……… 9Figure 1.3: Bitcoin and Ethereum return over the period 2016-2021 ……… … 10Figure 2.1: The research framework.……… ……… …… ……… …… …… 63

Figure 4.1: DCC - GARCH estimates for Bitcoin and stock

indices……….85

Figure 4.2: DCC - GARCH estimates for Ethereum and stock indices ………

… 86

SOCIAL REPUBLIC OF VIETNAM Independence – Freedom – Hanppiness

ABSTRACT OF THE THESIS

Thesis title: THE FINANCIAL CONTAGION EFFECTS OF THE GLOBAL COVID-19 PANDEMIC: EVIDENCE FROM FINTECH AND TRADITIONAL FINANCIAL MARKETS

Major: Finance

Ph.D Student: Pham Thi Ngoc Dung

Keywords: Covid-19, Cryptocurrency, Fintech, Contagion, Safe Haven, Spillover,Volatility, Wealth effect

Abstract:

In this thesis, we investigate the contagion effect of the global Covid-19pandemic in terms of the shift in mean spillover, volatility spillover, and time-varying correlation between Asian emerging stock and Bitcoin, as well as betweenstock and altcoins The trivariate GARCH-BEKK models are estimated, whichinclude Covid-19 related dummies corresponding to the Covid-19 arrival date, thepanic index, the media hype index, and the sentiment index The time-varyingcorrelation obtained through the DCC-GARCH model between two markets isunder investigation to examine the transmission mechanism of the contagion effectand the safe-haven properties of Bitcoin and altcoins during various contexts of theCovid-19 pandemic Our results indicate that both Bitcoin and altcoin cannot serve

as a safe haven against Asian emerging stock markets in most contexts of thisrapidly escalating pandemic, as we find evidence of the presence of a contagioneffect, both in terms of a shift in mean spillover, volatility spillover, and dynamiccorrelation, between Asian emerging stock markets and cryptocurrency markets

since the appearance of the Covid-19 pandemic Especially, this contagion effectbecomes more obvious in the high panic period, the high media hype period, andthe negative market sentiment period Additionally, we also provide evidence insupport of the investor-induced contagion hypothesis during various turmoil context

of the Covid-19 pandemic We find that the source of the contagion effect betweenthe stock market and the cryptocurrency market is the wealth effect Furthermore,there is heterogeneity among the bitcoin and altcoin markets in how the globalCovid-19 pandemic affected their respective relationship with the Asian emergingstock market These findings are important for investors, risk managers, orpolicymakers making decisions during high uncertainty periods to understand when

to act and how much

Ph.D Student

Pham Thi Ngoc Dung

CHAPTER 1: INTRODUCTION

1.1 Research Background

In recent years, trade barriers and restrictions on international capital flow havebeen gradually removed In addition, the scope and scale of information technologyhave expanded dramatically, thus encouraging the globalization of the traditionalfinancial market, such as stocks, bonds, and foreign exchanges This phenomenon hascontributed to greater global financial integration and provided investors,corporations, and economies with a wide range of opportunities and strategic benefits.However, extraordinary financial integration is usually associated with an increasingvolatility spillover and a financial contagion effect, thus increasing the vulnerability ofthe investor to systematic risk arising from domestic or global factors (Ahmed,2021a), especially during crisis or uncertainty periods

During times of financial crisis, it is a stylized fact of global financial marketsthat asset volatility rises and asset classes become more correlated (Bekaert et al.,2009; Hartmann et al., 2004; Lee et al., 2011) The sign and magnitude of thesecomovements may lead economists, policy markers, investors, and financialinstitutions to raise the question of whether the global propagation mechanism hasdiscontinuities? Is there an existance of transmission mechanisms that only operateduring times of high uncertainty? Whether there is a crisis in one market followed by

an increase in asset price comovements across financial markets and national borders?and whether the international transmission of financial shocks should be interpreted ashaving different regimes during times of peace and during times of crisis?

Moreover, there is little consensus on how to create an appropriate system toregulate international capital movements Fischer (1998) indicated two importantreasons for the need to reform the 'international financial architecture' The first reason

is the high volatility of international capital flows and asset prices in emerging

markets and the limited ability of these markets to deal with this high volatility,increasing the propability to deal with excessive frequent, large, and disruptive shocks

in recipient markets The second reason is that international capital markets are highlysusceptible to contagion To effectively reduce volatility and the effect of contagion,proposals for a revised international financial architecture may benefit from a betterunderstanding of the causes and effects of international contagion

As several periods of global crisis occurred in the recent decade, such as theglobal financial crisis (2007-2009) and the European sovereign debt crisis (2010-2012), and most recently the global Covid-19 pandemic This may be an importantgap in the empirical literature on financial linkage and contagion These global crisisraises the question of whether the cross-boader diversification effects still remainduring market turmoil conditions, when the benefits of portfolio diversification aremost needed; whether the contagion effect between asset classes exist, and if so, whatare the transmission mechanisms of these contagion effect Furthermore, increasedcross-market linkage during turbulent periods is likely to increase portfolio risk andreduce the benefits of diversification, motivating investors and portfolio managers tolook for alternative investment assets that can act as a safe haven

The financial technology (Fintech) and its integration with traditional financialmarket have attracted the interest of policy makers, researchers, financial institutions,and investors around the world According to Chen et al (2019), cybersecurity,mobile transactions, data analytics, blockchain, peer-to-peer (P2P), robo-advising andthe internet of things (IoT) are the seven subcategories that can be used to categorizefintech The dramatic surge of interest in FinTech over the past few years hashighlighted the need for a better understanding of the implications of FinTech invarious aspects of the financial system Our study contributes to the knowledge ofblockchain-related aspects of fintech and contributes to the body of knowledge onasset diversification by taking into account the investment capabilities ofcryptocurrencies, including Bitcoin and altcoins, and their connections to traditionalmarkets

Since Nakamoto introduced the concept of Bitcoin in 2008, the popularity ofdigital currencies, commonly known as cryptocurrencies, has increased rapidly.Therefore, global investors may consider including cryptocurrencies such as Bitcoinand altcoins in their current equity portfolio Hence, they might require information onthe connection between cryptocurrency and other financial markets both duringtranquil periods and turmoil periods as well as the potential hedge and safe havenproperties of these cryptocurrencies Cryptocurrency and its role in portfoliodiversification have attracted the interest of researchers There is a great deal ofresearch investigating the financial links between the equity market and thecryptocurrency market (Bouri, Gupta et al., 2017; Corbet et al., 2018; Dyhrberg,2016a, 2016b; Guesmi et al., 2019; Ji et al., 2018), as well as the potential hedgingand diversification benefits of these digital currencies (Bouri, Jalkh et al., 2017; Giudici

& Abu-Hashish, 2019; Urquhart & Zhang, 2019) However, these studies have focusedprimarily on the implication of the return correlation of cryptocurrencies and thespread of volatility in portfolio management and have generally ignored the possibilityof'shift contagion' during periods of high volatility and turmoil In addition, the exactcauses of this shift in international transmission mechanism between cryptocurrencyand equity markets are not yet well known This research has been conducted toaddress this gap

1.2 Research context

1.2.1 The adoption of cryptocurrency in Asia emerging countries

The Chainalysis (2021) crypto adoption index gives a general summary ofwhich nations have the highest rates of mainstream acceptance of cryptocurrencies.This index is made up of three metrics including the cryptocurrency value received onthe chain, the volume of peer-to-peer (P2P) exchange trade and the retail valuereceived on the chain The methodology for the Crypto Adoption index, whichconsists of the three components mentioned above, is shown in Figure 3.3 by

summing the index scores of all 154 countries for each quarter from Q2 2019 to Q22021.

Figure 1.1: Global Crypto Adoption IndexSource: The 2021 Geography of Cryptocurrency Report (Chainalysis, 2021)

Figure 3.3 illustrates how people in more and more countries are adoptingcryptocurrencies or are seeing their current use increase The Global Crypto AdoptionIndex scores show that after a period of modest increase, the total global adoption was2.5 at the end of Q2 2020 At the end of Q2 2021, that total score is 24, showing thatglobal use has grown by more than 2300% since Q3 2019 and by more than 881% inthe past 12 months The reasons for this increased adoption vary by region Manypeople in emerging economies use cryptocurrencies to send and receive remittances,perform business transactions, and protect their funds from currency depreciation;however, institutional investment has pushed adoption in North America, WesternEurope, and Eastern Asia during the past year

Asian emerging nations make up five of the top 15 countries with the highestcrypto adoption index (Table 1.1), since adoption, transaction values, and activity ofcryptocurrencies have all increased dramatically in Asia in recent years

Table 1.1: The 2021 Global Crypto Adoption Index - Top 15 Countries

Country Index score

Overall index ranking

Ranking for individual weighted metrics feeding into the Global Crypto Adoption

Index

On-chain value received

On-chain retail value received

P2P exchange trade volume

Source: The 2021 Geography of Cryptocurrency Report (Chainalysis, 2021)

When adjusted for PPP per capita and internet-using population, some Asianemerging nations, including India, Pakistan, Thailand, China, and the Philippines,rank well on the crypto adoption index This is due to the enormous volumes oftransactions on peer-to-peer (P2P) platforms In these nations, many people use P2Pcryptocurrency exchanges as their main entry point into cryptocurrencies, typicallybecause they lack access to centralized exchanges, according to Chainalysis (2021)expert interviews

The increasing adoption of cryptocurrency in Asian emerging countries hasmade investors question whether or not to include Bitcoin and altcoin in their currentAsian emerging equity portfolio More especially, when the international financialmarket suffered quite a lot of crisis in recent decades, investors may also wonderwhether the inclusion of cryptocurrency can reduce risk and improve their portfolioperformance during turmoil periods Therefore, this research examines the safe havenproperties of Bitcoin and altcoin, as well as the potential for contagion and thetransmission mechanism of contagion between the Bitcoin/altcoin markets and theAsian emerging stock market Therefore, our findings will provide investors whocurrently have Asian emerging stocks in their portfolio with useful information onwhether or not to include Bitcoin and altcoins in their current equity portfolio andwhether the inclusion of cryptocurrencies in their portfolio can impact the risk andperformance of their current portfolio during high uncertainty periods.

1.2.2 The impact of the global Covid-19 pandemic on the financial market

The Covid-19 pandemic has caused an unprecedented shock in economiesaround the world and has a strong influence on the global financial market The globalhealth crisis not only impacts country-specific risk, but also has a strong influence onsystemic risk, which is why its impact on the linkage between financial assets hasattracted the attention of investors, policy makers, and economists The coronavirusdisease outbreak had been confirmed in almost all countries as of 5 October 2022 Thevirus had infected 624 million people worldwide, with a total of 6.55 million deaths(Worldometer, 2022)

The global Covid-19 coronavirus pandemic has had a large detrimental impact

on the world economy, despite the fact that it is impossible to quantify Lockdownsforced many industries to close for months and almost all economic activity wasrestricted In 2020, the global gross domestic product (GDP) decreased by 3.4% Toput this figure in context, a 3.4% decline in economic growth leads to a loss of nearlytwo trillion dollars in economic production (Centre for Economic Policy Research,2021) In response to the economic downturn, several governments implemented

stimulus packages to support national economies and provide relief to theunemployed Indeed, after its initial decrease in 2020, global GDP has grown,demonstrating the beneficial effects of stimulus packages and easing of limitations(IMF, 2021) The global economy recovered quickly from the initial shock andreturned to positive growth in 2021 It reached 92.3 trillion US dollars that year and isexpected to grow in the coming years, though Russia's war in Ukraine since February

2022 and its impact on the global economy may stymie economic growth

Although the pandemic had an impact on the world economy, some nations andregions were more severely affected than others For example, China's GDP climbed

by almost 5% in the third quarter of 2020, compared to the UK's GDP decline of about8% However, it had increased by about 7% in the same quarter the following year.Asia's GDP changed from minus 7.7% in South Asia to minus 0.2 percent in EastAsia Although there are many complicated reasons for this, East Asian nations'relatively swift response to the pandemic in its early stages allowed restrictions to beloosened as numbers decreased (Statista, 2021)

The global expansion of the Covid-19 epidemic has caused financial marketsaround the world to increase risk and alter cross-market relationships According toStatista (2021), all major stock market indices saw a significant decline in March 2020

as a result of the global Covid-19 outbreak, although the size of the decline and thenature of the rebound that followed differed greatly The global stock marketsexperienced dramatic drops following the coronavirus outbreak, but recoveredquickly The Dow Jones posted its biggest single-day loss ever on March 16, 2020,losing nearly 3,000 points, shattering its previous record of 2,300 points set just fourdays earlier The major European stock markets and conventional US stocks had lostaround 40% of their value as of March 15, 2020 Meanwhile, the NASDAQComposite Index and Asian markets have only lost 20 to 25% of their value Acomparable story can be found in the post-coronavirus recovery The NASDAQcomposite index was approximately 65 percent higher on November 14, 2021 than itwas in January 2020, while most other markets were only 20-40 percent higher

Furthermore, share prices have rebounded much more slowly on two main stockindices, the FTSE 100 index and the Hang Seng index, than on other major exchanges.However, in both instances, the root cause of the longer recovery is probablyconnected to political developments unrelated to the coronavirus, specifically Brexit,and broader political uncertainty After falling to 18,917.01 points in March 2020 due

to the global coronavirus (COVID-19) pandemic, The Nikkei Stock Average (Nikkei225) reached 26,000 points for the first time since 1991 in November 2020, thenclosed at 28,791.71 points in December 2021

Silver and gold were the most profitable investment metals in 2020 Despite thefact that gold was the most popular asset, silver had a higher average return that year.Silver returned approximately 47%, while gold returned nearly 25% Looking forward

to 2021, silver and gold produced negative investment returns Gold had a negative3.6% return that year, while silver had a negative 11-7% return (Statista, 2021)

According to Topcu and Gulal (2020), the impact of the Covid-19 outbreak isgreatest in Asian emerging markets compared to their counterparts in South America,the Middle East and Europe The Shanghai Stock Exchange Composite Indexmeasures the performance of all stocks traded on the Shanghai Stock Exchange Thisindex fell from 3050.12 points in December 2019 to 2750.3 points in March 2020,then gradually recovered to close at 3,024.39 points in September 2022 The KoreaComposite Stock Price Index (KOSPI) and the Korean Securities Dealers AutomatedQuotations (KOSDAQ) indexes stood at 3,001.8 and 1006.04, respectively, onDecember 9, 2021 The Korean government announced a plan to help financialmarkets recover after fears of the coronavirus (COVID-19) caused the KOSPI to dropbelow 1,600 points for the first time in ten years The coronavirus had a negativeimpact on the South Korean economy, which recovered quickly as early as 2021 TheS&P BSE Sensex index, one of India's two main stock indices, lost nearly one-quarter

of its value between the end of February and the end of March 2020 as a result of theeconomic impact of the economic impact of the global coronavirus (COVID-19)pandemic It has since recovered and is expected to surpass its precorona level in

November 2020 Figure 3.1 shows the stock price returns of eight emerging Asianmarkets from 2016 to 2021.

Figure 1.2: Emerging Asian stock price return over the period 2016-2021

Source: Own Processing

The global Covid-19 pandemic also has impacted the cryptocurrency market.Figure 3.2 represents the return of the two most capitalized cryptocurrecies, namelyBitcoin and Ethereum, over the period from 2016 to 2021 Bitcoin (BTC) prices began

at 378.17 in January 2016, rose to 13062.17 in December 2017, and then felldramatically to 3441.03 in January 2019 After that, Bitcoin's price peaked during theCovid-19 era in 2021, when it reached an all-time high of over $65,000 USD inNovember, at 9545.08 in January 2020 The price spike was related to the debut of aBitcoin ETF in the US, whereas others in 2021 were brought about by eventsinvolving Tesla and Coinbase, respectively However, due to speculative governmentregulation, the most popular cryptocurrency in the world saw a major correction in

April 2021 According to a 2020 study based on IP addresses of so-called hashers whoused specific Bitcoin mining pools, China accounted for more than half of all Bitcoinmining Experts also highlighted a power outage in China's Xinjiang area as acontributing reason The quantity of Bitcoins mined decreased as a result of thisunexpected occurrence, which may have scared investors into selling their holdings

The price of Ethereum (ETH) in US dollars hit new highs in November 2021,rising to more than $4,800 USD Similarly to Bitcoin (BTC), the price of ETH rose in

2021, but for quite different reasons The rally on Ethereum was supported bytechnological advancements that aroused traders' curiosity, as opposed to Bitcoin,whose price growth was powered by the IPO of the largest crypto trader in the UnitedStates First, the so-called "Berlin upgrade" for the Ethereum network was released inApril 2021 with the goal of lowering ETH gas prices, or transaction fees Second, thelaunch of the smart contract protocol Uniswap V3 in May 2021 is anticipated togreatly enhance Ethereum trading

Figure 1.3: Bitcoin and Ethereum return over the period 2016-2021

Source: Own Processing

It can be seen from these figures that financial instability is essentially driven

by the Covid-19 health crisis (Ji et al., 2020) Data also indicate that this global healthcrisis can have a powerful impact on the stock market, as well as the bitcoin and

altcoin markets of very different sizes and structures throughout the world Zhang et

al (2020) indicate that global financial market risks have increased significantly inresponse to the global Covid-19 pandemic when the high uncertainty of the epidemic,

as well as its associated economic losses, have contributed to the high volatility andunpredictable of the financial market Therefore, the global impact of the Covid-19pandemic can drive investors’ decision on portfolio management to avoid potentiallosses in a period of high uncertainty There is a need to consider on the question ofwhether the high rate comovement between stock and crypto market during variousstates of the Covid-19 pandemic constitute contagion? Or are these markets sointerdependent that they have a similar high rate of co-movement in all states of themarkets? Whether cryptocurrencies can provide diversifier, hedge, or safe havenproperties against stock during various contexts of this global Covid-19 pandemic.This research is conducted to answer this question with the following research objectsand questions

1.3 Research objectives and scope

1.3.1 Research objectives

The analysis in this study is divided into four parts First, this research explores thesafe haven properties of bitcoin and altcoins for the Asian emerging stock market byanalyzing the change in the dynamic correlation between stocks and these digitalcurrencies during the high uncertainty periods of Covid-19 These analyzes willprovide evidence on the cross-border diversification benefits of crypto assets,particularly during turbulent periods, where hedge, safe haven, and diversifiedbenefits of cryptocurrencies are required most

Second, we tested the existence of a contagion effect between bitcoin and stock, aswell as between altcoin and stock during various turmoil contexts of the global Covid-

19 pandemic To be more specific, we are interested in assessing the contagion effectdefined as the shift in the mean spillover, volatility spillover, and time-varying

correlation between Asian emerging stock market and bitcoin / altcoins market, duringturmoil periods related to the Covid-19 pandemic

Third, we investigate the transmission mechanism that occurs during the Covid-19pandemic and causes contagion by testing the investor-induced contagion hypothesis

on the relationship between the stock market and the cryptocurrency market We focus

on determining whether the wealth effect mechanism or the portfolio rebalancingmechanism can be considered as the source of investor-induced contagion between theAsian emerging stock market and the cryptocurrency market during the period from

2016 to 2021 Especially, we conducted empirical research in the various distincttumoil contexts of the Covid-19 pandemic period including the Covid period, the highpanic period, the high media hype period, and the negative sentiment period toprovide more thorough information

Four, we also consider the probability that the alcoin will differ from Bitcoin withrespect to being a safe haven during various turmoil periods of the global Covid-19pandemic We consider that altcoins will co-move with Aian emerging stock marketsdifferently from Bitcoin because Bitcoin and altcoin differ fundermentally from eachother

1.3.2 Research scope

In this thesis, we choose the two largest market capitalizations ofcryptocurrencies, namely Bitcoin and Ethereum, which represent the altcoin market,and 8 Asian emerging stock indices: China (SSEC), Thailand (SETI), Philippines(PSEI), Indonesia (JKSE), India (S&P BSE100), Taiwan (TSE 50), Malaysia (KLSE),and Parkistan (KSE 100) The sample period spans from 4 January 2016 to 13 August

2021 Therefore, our sample periods can cover both the pre-Covid-19 period and theCovid-19 period

1.4 The research questions

Explicitly, the main issues under analysis in this study are as follows

i Do cryptocurrencies provide safe haven properties against Asian emerging

stock markets during various turmoil contexts of the Covid-19 pandemic?

ii Does the contagion effect between cryptocurrency markets and Asian

emerging stock markets exist during various turmoil contexts of the

Covid-19 pandemic?

iii Does the contagion effect between the Asian emerging stock market and the

cryptocurrency market source from the wealth effect mechanism or theportfolio rebalancing mechanism during various turmoil contexts of theCovid-19 pandemic?

iv Is there heterogeneity between the bitcoin market and the altcoins market in

how the global Covid-19 pandemic impacted their respective relationshipwith the Asian emerging stock market?

1.5 Methodology

In our analysis, the two cryptocurrencies, including Bitcoin and Ethereum, areconsidered First, following the DCC-GARCH estimation, the time-varyingcorrelations of the Bitcoin-stock and Ethereum-stock pairs are extracted into separatetime series and regressed on interactive dummy variables representing Covid-19turmoil periods in order to test these two digital assets as safe haven assets against theequity market The dummy variables are constructed based on the arrival date ofCovid-19 and several factors related to Covid-19 news such as the panic index (PI),the media hype index (MHI), and the sentiment index (SI) to identify turbulentepisodes

Second, we employ two methods to test for the contagion effect betweenbitcoin/altcoins and Asian equity stock markets The first test is the test for test for thechanges in the dynamic correlations between stock and Bitcoin; and stock and altcoinduring various contexts of the Covid-19 period by employing the DCC-GARCH

process in order to identify the contagion effect between stock and bitcoin/altcoin interms of a shift in dynamic correlation.

In addition, the second test for the contagion effect is the test for the shift in terms

of mean spillover and volatility spillover between bitcoin/altcoins and Asian equitystock markets We employ the standard VAR-GARCH framework with BEKKrepresentation with dummy variables that represent for turmoil periods related to theglobal Covid-19 pandemic, which allows for shifts in the parameters capturingspillover between Asian emerging equity market and Bitcoin/ altcoin market Thisapproach accommodates multiple shifts at different stages of the market The question

of whether there was a contagion effect in terms of volatility spillover from Asianemerging stock markets to cryptocurrency market exits during the high-uncertaintyperiods of the Covid-19 pandemic is addressed by Wald tests, which examinehypotheses on volatility spillover from stock market to cryptocurrency markets.Specially, three following possibilities are under consideration in this research: nomean or volatility spillover whatovers from the stock market to the digital currencymarket; no shift contagion effect, which means, no change in the transmission of mean

or volatility during the turmoil period of Covid-19; and no mean or volatility spilloverduring the tranquil period, which is a special case of shift contagion if spillover effectsare only presented during turbulent periods

Finally, we also employ the time-varying correlation between bitcoin/altcoins andAsian equity stock markets that was extracted from the DCC-GARCH model to testfor the transmission mechanism of the contagion effect between these two markets.This method enables us to derive useful information on the transmission mechanismthat causes contagion between stocks and cryptocurrencies to exist during variousCovid-19 pandemic-related turbulence contexts The hypothesis under consideration isthat the investor-induced contagion between stock and cryptocurrencies through thewealth effect

1.6 The research contributions

1.6.1 Theory contribution

Our analysis departs from previous research in several aspects First, there is agreat deal of literature that examines the hedge and safe haven properties of gold(Baur & Lucey, 2010; Baur & McDermott, 2010; Goodell & Goutte, 2021; Klein etal., 2018), bitcoin, altcoin, or cryptocurrencies in general (Dyhrberg, 2016b, 2016a;Goodell & Goutte, 2021; Klein et al., 2018; Smales, 2019) The question of whethercryptocurrencies might share some of gold's hedging or safe-haven qualities orwhether they are linked to conventional financial asset classes is still up for debate.The disparities in the findings regarding the impact of the Covid-19 pandemic on thesafe haven properties of cryptocurrencies can be attributed to differences in samplecomposition, inference procedures, and the asset classes against whichcryptocurrencies were compared, as well as the turmoil periods under consideration.Therefore, this issue needs to be reconsidered in light of the Covid-19 pandemic

Our study contributes to the ongoing research on the hedge and safe haven properties

of Bitcoin and altcoins by determining the extent to which Bitcoin and altcoins can act

as a hedge, safe haven, and/or diversifier against stock price movements in Asianemerging markets under different conditions of the global Covid-19 pandemic andconditional on relevant uncertainty indicators Furthermore, it should be highlightedthat very little work has attempted to investigate the impact of diverse uncertaintycontexts on the safe haven features of bitcoin and altcoins.Almost all of theseresearches are concerned with particular uncertainty indicators such as VIX index(Bouri, Gupta et al., 2017), EPU index (Jiang et al., 2021; Mokni et al., 2020; Wang etal., 2019; Wu et al., 2019), financial stress index (Bouri, Gupta et al., 2018; Zhang &Wang, 2021) , extreme negative market condition(Ahmed, 2021a) or Covid-19declaration date (Conlon & McGee, 2020; Corbet, Larkin, et al., 2020a; Jiang et al., 2021)

In this analysis, we account for four uncertainty indicators, including the Covid-19arrival date, the panic index, the media hype index, and the sentiment index, to capturethe core impacts of uncertainty on Bitcoin/stock and altcoin/stock relationships Theuse of Covid-new related indicators such as the panic index, the media hype index,

and the sentiment index as indicators to construct turmoil periods related to the globalCovid-19 pandemic provides a deeper measure of the time varying correlationbetween Bitcoin/altcoin and the Asian emerging stock markets, as well as the safehaven properties of Bitcoin and altcoins conditional on different states of this globalhealth crisis This method allows us to detect the key elements of the dependencebetween Bitcoin/altcoin returns and Asian emerging stock returns while controllingfor several uncertainty proxies that earlier studies overlooked Evidence on the safehaven features of Bitcoin and altcoins is crucial for investors who want to hedge theirrisk exposure to Asian emerging stock price swings as well as their downside riskduring the Covid-19 outbreak Such precise information would enable those investors

to e ectively hedge against an uncertain exposure, construct more effective portfolios,ffand clearly comprehend the similarities and differences between Bitcoin and altcoinsunder various circumstances

Second, we test for the existence of contagion effect between stock market andcryptocurrency market by applying the concept of "shift contagion" to analyze themean spillover, volatility spillover and dynamic correlation between the Asianemerging stock market and the two groups of cryptocurrency markets, namely Bitcoinand altcoins, during the turmoil episodes of the global Covid-19 pandemic, which is

an ideal period to study this effect According to Forbes & Rigobon (2002), a'shiftcontagion' can be defined as a significant shift in linkage between markets duringturmoil periods The contagion effect should be distinguished from regularinterdependencies and spillovers across regions or asset markets, as contagion refers

to an unexpected transmission of shocks (Beirne et al., 2013) The behavior offinancial assets in general during high uncertainty periods and, more specifically, thebehavior of cryptocurrencies during turbulent periods continue to be of interest(Ahmed, 2021a; Bouri, Gupta et al., 2017; Bouri, Molnár et al., 2017; Colon et al., 2021;Handika et al., 2019; Klein et al., 2018; Matkovskyy & Jalan, 2019; Mokni et al., 2020;Naeem et al., 2021; Selmi et al., 2018; Wang et al., 2019; Zhang & Wang, 2021) There isconsiderable interest in how the Covid-19 crisis might affect the relationship betweencryptocurrencies and conventional financial markets, both now and in the future

(Goodell & Goutte, 2021) Although earlier studies address the question of whether ornot the link between cryptocurrency and traditional financial assets changes over time

in light of the Covid-19 pandemic, they mainly measure the contagion effect as achange in correlation (Diniz-Maganini et al., 2021; Dutta et al., 2020; Goodell & Goutte,2021), coskewness (Matkovskyy & Jalan, 2019) or an increase in the coincidence ofextreme return (Handika et al., 2019) Additionally, Zhang & Wang (2021) argued thatthe inclusion of volatility spillover analysis will better reflect the real situation andprevent crucial information from missing rather than using the return to examine therelationship between financial factors, which may result in the loss of some detail andreduce the precision of the investigation Therefore, the objective of this study is toaddress this gap by identifying the change in mean spillover, volatility spillover, anddynamic correlation between the Asian emerging stock market and the cryptocurrencymarket during the turmoil period of the Covid-19 pandemic To our knowledge, this isthe first research that specifically examines the effect of contagion from the Asianemerging stock market to the cryptocurrency market as a shift in return spillover,volatility spillover, and time-varying correlation in light of the global Covid-19pandemic

Third, it should be noted at this point that there is very little previous studies on thecontagion effect between cryptocurrencies and traditional financial markets that hasattempted to examine the transmission mechnism of contagion effect betweenbitcoin/altcoin and stock market Almost all of these researches are concerned withevidence on the existance of contagion effect during the period of Covid-19 pandemic(Diniz-Maganini et al., 2021; Dutta et al., 2020; Goodell & Goutte, 2021, Nguyen et al.,2022; Huang et al., 2021; Jiang et al., 2021) and generally ignore the transmissionmechnsism that cause this contagion effect Therefore, the hypothesis on investor-induced contagion through the wealth effect versus the portfolio rebalancing as atransmission mechanism that causes the contagion effect on the relationship betweenthe stock market and the cryptocurrency market to exist during various turmoilcontexts of the global Covid-19 pandemic is being formally tested for the first time inthis study, to the best of our knowledge As a result, our research differs from earlier

research that considers contagion as a by-product that follows the arrival of a highuncertainty or negative events.

Four, previous research on the linkage between cryptocurrency market and otherconventional markets as well as the hedge, safe haven, and diversifier ofcryptocurrencies mainly choose Bitcoin to represent the cryptocurrency market In thisthesis, we consider that the safe haven properties of Bitcoin and altcoins against thestock market and their correlation with the stock market may be different in variouscontexts of the Covid-19 pandemic may be different In other words, there will beheterogeneity among the bitcoin market and the altcoins market in how the globalCovid-19 pandemic impacted their respective relationship with the stock market, asthese two groups of the cryptocurrency market might have fundamental differencesamong them

Finally, a global portfolio index or developed market index is typically used inearlier research as a standard proxy for the global equity market, or it may concentrate

on a relatively small number of countries in one or two regions; therefore, it isdifficult to capture the different responses in different regions of the market AsShaikh (2020) and Wang et al (2019) point out, the heging and diversificationproperties of cryptocurrencies vary from country to country; therefore, we cover alarge sample of 8 Asian emerging countries during the period from 2016 to 2021,which may provide a rich basis for comparison of the contagion effect acrosscountries and cover a longer time horizon of the Covid pandemic than earlier studies.Furthermore, we also examine the effect of contagion in different contexts of theglobal health crisis and conditional on relevant Covid new-related indicators,including the panic index, the media hype index, and the sentiment index Therefore,our study attempts to draw a holistic picture of the potential mean spillover, volatilityspillover, and dynamic correlation from Asian emerging equity market tocryptocurrency market not only under different market circumstances (tranquill andturmoil period) but also in different regional contexts

1.6.2 Practice implications

The diversification implication of the cryptocurrencies derived from this study isalso important for investors and portfolio managers, who seek an optimal portfolio bycombining cryptocurrencies with traditional financial assets such as stock, bonds,foreign exchange, etc , especially during high uncertainty market conditions, as well

as regulatory interest seeking to maintain financial stability and investor protection.For an investor, studying the mean spillover and volatility spillover mechanism andthe cross-correlation between cryptocurrencies and other asset classes and identifyingthe probability of increasing in systematic risk during turmoil market conditions are ofgreat significance for forming portfolio which is composed of cryptocurrencies.Furthermore, empirical results on the influence of different contexts of Covid-19 onthe link between the equity market and the cryptocurrency market can help investorsdetermine the appropriate portfolio strategies during a period of global financialuncertainty Moreover, by identifying whether investor induce contagion betweenstock market and cryptocurrency market through the wealth effect or portfoliorebalancing hypothesis, our study provides important implications in terms of howmoneytary policy makers should respond in maintaining liquidity levels of financialsystems including both the traditional markets like equity markets and fintech marketslike Bitcoin and altcoins

1.7 Dissertation structure

The thesis is divided into five following chapters The research background, thereasons for choosing the research topics, the research objective and questions, and thescope of the study are all presented in the first chapter The theoretical foundations ofsafe haven properties and the contagion effect are the main topic of Chapter 2 Thedissertation also reviews earlier research on the relationship between the equitymarket and the cryptocurrency market The dissertation highlights the research gapbased on the theoretical and earlier studies in order to develop the research questions,objectives, and hypotheses The methodology section of Chapter 3 introduces the dataset and models for testing the hypotheses The research results and discussion of all

the findings are presented in Chapter 4 In addition, the conclusion, the theoretical andpractical contributions, the research limitations, and a suggestion for further study areall presented in Chapter 5.

CHAPTER 2: LITERATURE REVIEW

In an attempt to provide an overview of contagion, this chapter first addressesdifferent definitions of contagion adopted by the literature and the relevant terms toidentify the specific features of financial contagion, followed by a review of thetheoretical literature on the international propagation of shocks in order to identify thereasons why the international transmission mechanism changes during financial crises.Next, we present a literature review on the connectedness between Bitcoin, altcoin,and traditional financial assets We discuss the following topics to develop theframework that underpins our research questions: the interdependence betweenBitcoin, altcoin, and traditional financial assets; and the contagion effect betweenBitcoin, altcoin, and financial assets during various turmoil periods, especially duringthe global Covid-19 pandemic

2.1 Definitions

2.1.1 Definitions of hedge and safe haven properties

Investors and financial institutions may have a propensity to look for "safehaven" instruments to lower risk, prevent potential losses, and safeguard the value oftheir portfolios during times of high uncertainty Although these phrases arefrequently used interchangeably in the public financial press, there are differencesbetween the definitions of terms such as "hedge," "diversifier," and "safe haven" thathave been established in previous studies Baur & Lucey (2010) defined a hedge as anasset that is negative correlated or uncorrelated with other asset or portfolio onaverage, while a safe haven as an asset that is uncorrelated or negative correlated withother asset or portfolio in turmoil or uncertainty period, and a divesifier as an assetthat is positive correlated with other asset or portfolio on average, except for the case

of perfectly positive correlated

In this research, we employed the hedge and safe haven definitions that weredeveloped by Baur & McDermott (2010) Baur & McDermott (2010)’s definition of hedge

and safe haven based on these concepts from Baur & Lucey (2010) as a foundation,but expanded on their work on one crucial aspect that is the length of the effect anddistinguished between a strong hedge and a weak hedge, as well as a strong safe havenand a weak safe haven According to Baur & McDermott (2010), 'A strong (weak)hedge is defined as an asset that is negatively correlated (uncorrelated) with anotherasset or portfolio on average”; “A strong (weak) safe haven is defined as an asset that

is negatively correlated (uncorrelated) with another asset or portfolio in certainperiods only, eg, in times of falling stock markets' The key characteristic of a hedge

is that it must hold consistently, but the key characteristic of a safe haven is that itonly needs to hold during specific periods, such as during a crisis or high uncertaintytime Safe haven investments hold their value or even increase in value during marketdownturns, despite the fact that investors must always diversify or hedge theirportfolios It is possible that an asset working as a hedge against stock might have atendency to comove with stock during high uncertainty periods, since marketparticipants might sell different types of asset or all assets simultaneously, which can

be explained by herd behavior or contagion effect (Boyer et al., 2005; K J Forbes &Rigobon, 2002; Mendoza & Calvo, 2000) On the other hand, safe-haven investmentsmight only have a negative correlation with other assets during times of marketturbulence and typically comove with other assets (Baur & McDermott, 2010).Furthermore, Ji et al (2020) argued that the safe haven properties of an asset are notonly subject to the particular asset class or market studied, but also have a highprobability to change over time and depend on the financial characteristics ofuncertainty periods

2.1.2 Contagion Definitions

In-depth theoretical and empirical research has recently been done to betterunderstand interdependence and the possibility of contagion Previous studies usuallyuse the following terms, such as "connectedness", "linkage", "relation",

"interdependence", and "spillover," together with "financial contagion." Todifferentiate the term 'contagion' from its relative terms, we will review some of the

most popular definitions of connectedness, interdependence, and contagion, though,indicating the key characteristics of financial contagion

Scott (2016) argued that "connectedness" arises from the direct overexposure of afinancial institution to one another, culminating in a domino effect of collapse if one

of them suffers a shock From an economic perspective, Diebold & Yilmaz (2015)consider connectedness measurement an important aspect of any comprehensivemarket risk assessment because it separates portfolio risk from fundamental risk.Furthermore, comprehending market operations and efficiently managing corefinancial market activities such as asset pricing, risk management, and portfolioallocation are all dependent on financial market connectivity.Additionally, marketinterdependence contains asymmetries, according to one stylized feature of financialmarkets, that large negative returns are more connected than large positive returns(Ang & Chen, 2002)

While interdependence exists in both good and bad times, it simply refers to a

"continuous" and "normal" state or a "transquil-period" linkage between countries(Jung & Maderitsch, 2014) More specifically, "interdependence" may refer to asituation in which a group of countries shares a common market structure and hasdirect economic and financial links With these conditions, even if there is shocktransmission between countries, it should not be regarded as contagion (Forbes &Rigobon, 2001) This is due to the fact that cross-market links appear to persist duringcalm periods in either of the two countries Fundamentals could drive marketconnections and commonly seen phenomena, according to this literature (Baele &Inghelbrecht, 2010), hence, they might be able to fully describe how stock marketrelationships are measured (Jung & Maderitsch, 2014) This means that the theoryignores the consequences of unexpected expectation adjustments and herding behavioramong investors Furthermore, Forbes & Rigobon (2002) and Baele & Inghelbrecht(2010) determine that 'interdependence' is also defined by an excess of co-movementand a limited time change

On the other hand, "contagion" demonstrates a distinct state of "linkage" or

"relationship" rather than "interdependence" The financial contagion label suggests aphenomenon when there is a change in the international transmission mechanismduring financial shocks To put it another way, a sudden crisis that only affects onemarket can spur investors and portfolio managers to act, increasing the possibility of acrisis in other markets (Kannan & Köhler-Geib, 2009) However, by reviewing thecurrent literature, we notice that precious studies on contagion do not alwaysdistinguish between contagion and interdependence and there is no consensus on thedefinition of financial contagion Pericoli & Sbracia (2003) indicated the five mostpopular definitions of contagion that are frequently employed in the literature, asfollows:

Definition 1: “Contagion is a significant increase in the probability of a crisis in

one country, conditional on a crisis occurring in another country”

In empirical studies on exchange rate collapses and their global effects, thisdefinition is frequently utilized (Pericoli & Sbracia, 2003) Given the finding thatexchange rate crises frequently involve a large group of nations and that some of thosenations may not devalue even when they are under attack from a powerful wave ofspeculation, Glick & Rose (1999) supported the hypothesis that the spread of exchangerate collapses may be caused by factors that are above and beyond anymacroeconomic phenomena Since the causes of the initial crisis and its internationalspread are not specified by this definition, it might be consistent with a variety ofperspectives on the international transmission mechanism For example, the spread ofthe crisis may be caused by common shocks, international trade linkage, investorirrational behavior Additionally, a policy game between nations may have resulted inthe cluster of crises as the equilibrium result All of these phenomena could be labeledcontagion Thus, this definition is relatively broad since the term 'contagion' could bebased on fundamentals or not

Definition 2: “Contagion occurs when volatility of asset prices spills over from

the crisis country to other countries”

Volatility is important information in portfolio construction, risk management,option pricing, or volatility trading Strohsal & Weber (2012) discussed the twoperspectives of asset volatility transmission, which could be extended directly tointerpret the phenomenon of volatility spillover In the first perspective, volatility isusually associated with uncertainty or risk (Chowdhury, 1993; R F Engle et al., 1987;Grier & Perry, 2000), so volatility transmission can be considered as the spillover ofmarket participants’ uncertainty The second perspective regarded volatility as ameasure of information flow intensity (Andersen, 1996; Clark, 1973; Ross, 1989);therefore, volatility spillover is usually considered as an outcome of possible(auto)correlated information flow

As the surge of asset price volatility during financial turmoil periods can beseen as a stylized fact in the international financial markets, according to thisdefinition, contagion is the spread of volatility from one market to another Sincevolatility is commonly associated with uncertainty, contagion can be considered as thespread of uncertainty across financial markets This definition ignores the possibilitythat a simultaneous rise in volatility in various financial markets could be the result oftheir regular interdependence or a structural change that could have an impact oncross-market linkages because it only considers the presence of volatility spilloverrather than its causes (Forbes & Rigobon, 2002) It is instead a foundation for thefollowing definition

Definition 3: “Contagion occurs when cross-country co-movements of asset

prices cannot be explained by fundamentals”

The main issue in portfolio construction and risk management is whetherfinancial assets become more interdependent during turmoil periods (Bae et al., 2003).This topic has attracted significantly increasing interest from researchers since the fivemajor crises of the 1990s including the ERM (1992), the Mexican devaluation (1994),the East Asian crisis (1997), the Russian default (1998), and the devaluation of theBrazilian real (1999) Rodriguez (2007) indicated that all of these events had onething in common, that is, the crisis originating in one market and then extended to awide range of markets of very different size and structure in the way that was difficult

to explain on the basis of fundamental change In other words, fundamental factorscannot explain the crisis timing and modality; nonetheless, they may explain whysome countries or markets are vulnerable to crises while others are not In light of themodel framework's support for numerous instantaneous equilibria in the presence ofcoordination problems, this formulation is theoretically accurate (Pericoli & Sbracia,2003).

Definition 4: “Contagion is a significant increase in co-movements of prices

and quantities across markets, conditional on a crisis occurring in one market or group

of markets”

Pericoli & Sbracia (2003) argued that this definition emphasizes the quantitativedimension (“a significant increase”), therefore, conveys contagion as “excessivecomovement”, relative to some standards The open question is how to distinguishbetween normal co-movements in prices and quantities caused by simpleinterdependence and excessive co-movements due to the structural break in data Thefollowing definition is based on a similar perception of contagion identification

Definition 5: “(Shift-)contagion occurs when the transmission channel

intensifies or, more generally, changes during a period of turmoil”

Early reaseaches have adpoted the broder and more inclusive definitions ofcontagion that do not distinguish between interdependence and contagion Thebroader definition of contagion defines that contagion is a significant increase in theprobability of a crisis in one market, conditional on a crisis occurring in anothermarket, no matter why that vulnerability occurs or whether there is the existence ofthat linkage at all times (Pericoli & Sbracia, 2003) However, if two countries havestrong links through fundamental channels such as trade or financial linkages, theywill be closely connected to each other during tranquil periods and turbulent periods.Therefore, a shock that is transmitted during turmoil periods is more likely to be acontinuation of a cross-market linkage or interdependence that was present duringstable times (Mit, 2004) Recent studies (Bekaert et al., 2005, 2014; Boyer et al., 2005;Petmezas & Santamaria, 2014) have proposed using the more specific definition of

contagion, which is 'shift contagion' developed by Forbes & Rigobon (2001) Forbes &Rigobon (2001) suggested replacing the term “contagion” with “shift-contagion” todifferentiate their definition from pre-existing defition of contagions When evaluatingthe efficacy of intervention measures and financial aid packages, policymakers andgovernment officials may find it more helpful to distinguish between these twodefinitions.

Shift contagion occurs when the transmission mechanism changes or intensifiesduring a crisis, whereas interdependence implies that there is no significant change incross-market relationships The transmission mechanism between internationalmarkets may strengthen during a crisis, additionally, there is the case that sometransmission channel might be active only during high turbulent periods However, it

is not necessary to limit the definition of contagion to the hypothesis that country linkages become stronger than normal (Pericoli & Sbracia, 2003) Additionally,this concept of shift contagion characterizes contagion with a shift in cross-marketlinkage which can be measured by a number of different methods, such as correlation

cross-in asset return, the transmission of shocks or volatility, or the probability of aspectilative attack; but does not clarify on how this shift occurs

This concept of contagion is somewhat similar to the two previous definitions.However, the concept of shift contagion also includes discontinuities in the behavior

of some economic variables which are produced by learning processes, herdingbehavior, or informational cascades by market participants In definition 3, contagioncan be viewed as a jump between multiple equilibria Although Definition 4's test forstructural breaks in the data-generating process would be more applicable, the shiftcontagion might also be quantified in terms of excessively strong (or weak)comovement of asset price and quantity (Pericoli & Sbracia, 2003) However, someeconomists argue that it is not possible to simply identify contagion based on tests ofchanges in the cross-market linkage As a result, identification of the particular cross-market transmission mechanism is more important because some types of propagationmechanisms, regardless of the size of the propagation, might cause contagion

In this thesis, we focus on the specific definition of shift contagion for severalreasons First, market shocks include either global shock, idiosyncratic shock, orcontagion (Masson, 1998) Since most economic volatility is market-specific, cross-market diversification should lower portfolio risk and increase projected returnsbecause there will be low connection across different markets This is a key concept ofportfolio creation and investing strategy However, if the market correlation increasesafter a crisis in one market, this would reduce the benefits of internationaldiversification Thus, the shift-contagion test would be viewed as a test of theefficiency of a multimarket diversification strategy during turmoil periods (Forbes &Rigobon, 2001) This issue would not be addressed by other, less strict definitions ofcontagion, which place more emphasis on the size of cross-market linkages than onchanges to these relationships Other more stringent notions of contagion, which focus

on how shock transmission between different markets might provide additionalinformation, but this is not necessary to address the issue of the effectiveness ofinternational diversification

Second, to assess the possible efficacy of government involvement and thebailout fund, it may be helpful to measure the shift-contagion effect Policy makersmight worry that a negative shock in one market may have a negative impact onanother market, even if the fundamentals of the second market remain strong and there

is a weak link between two markets According to the definition of shift contagion,this type of shock propagation would constitute contagion; then a financial crisis inthe second market might be avoided by government action and a bailout fund.However, if two markets are fundamentally strongly related, a crisis in one market islikely to have a significant influence on the other market, and this shock transmissionwould not be considered contagion Therefore, shock receiver markets would need toadjust to this shock A bailout under these circumstances would be suboptimal unlessother inefficiencies exist; although it might lessen the first negative impact, it wouldmerely prolong the necessary adjustment (Forbes & Rigobon, 2001) Therefore,evidence of the contagion effect may be useful to support multilateral intervention

when defining contagion as a change in cross-market links Multilateral action would

be less effective and more difficult to defend if there was no evidence on the existence

of a contagion effect This distinction would not be made clear by other, less stringentdefinitions of contagion

Third, some proposed frameworks for measuring the international propagation

of shocks assume that most shocks are transmitted through fundamental mechanismssuch as trade or financial linkage Other approaches focus on altering investorbehavior on the premise that market participants may behave differently following asignificant negative shock However, many of these propagation processes arechallenging to directly test Thus, we follow the definition of shift-contagion to avoidhaving to directly measure and differentiate between various shock transmissionmechanisms

Fourth, it might be helpful to categorize theories into those that assume acontinuation of current transmission mechanisms versus those that assume a change inpropagation mechanisms during the turmoil period when model testing for theexistence of contagion defines the contagion effect as a significant change in thecross-market linkage during crisis (Forbes & Rigobon, 2001)

In conclusion, the definition of shift contagion could be useful in justifying theeffectiveness of international portfolio diversification, assessing the efficacy of bailoutfunding and international involvement, and identifying diverse shock transmissionmechanisms

2.2 Theoretical literature on the international propagation of shocks

According to Claessens et al (2001) and Forbes & Rigobon (2001), contingent theories and noncrisis contingent theories make up the two main categories

crisis-of hypotheses in the theoretical literature on the international propagation crisis-of shocks Ashock in one market causes the transmission mechanism to change, which is explained

by crisis-contingent theories On the contrary, non-crisis-contingent theories assume