Who are the vectors of contagion? Evidence from emerging markets

https://doi.org/10.1016/j.irfa.2023.102599Get rights and content

Highlights

  • We propose a dynamic measure of Contagion based on the coincidence of large negative returns, unexplained by fundamentals.

  • We investigate the interaction of the Contagion measure and the flows of local institutions and individuals, and foreigners.

  • The structural VAR models show that foreigners are the main vectors of contagion, as implied bytheoretical models.

  • The models also show that foreigners actively sell and impact local prices in response to large negative US returns.

  • In the days following Contagion peaks, foreign investors keep selling while local institutional investors become net buyers.

Abstract

We test whether foreign investors are the vectors of contagion to emerging markets, as various theoretical models imply. We also explore the role of local institutions and individuals during and after contagion days. To do this, we propose a novel measure of contagion and estimate its dynamic relationship with the net purchases of each of the three groups of investors, from 2007 to 2016, in seven emerging markets. We find that foreign investors bring contagion by actively selling and impacting local prices on days of large declines in the US stock market and the days following. Local institutions are also net sellers on the day of contagion, while individuals act as the main liquidity providers, but institutions become net buyers soon after.

Introduction

Contagion is one of the most important issues in emerging market finance, due to its effects on capital flows, risk-sharing, and financial integration processes. However, little is known about its transmission channels. In this article, we propose a novel empirical approach to identify what types of investors are most actively involved in transmitting contagion from developed to emerging markets.

Interest in contagion to emerging markets has revived from time to time, particularly after worldwide financial upheavals, such as the recent global financial crisis (GFC) in 2008, and the sovereign debt crisis in the euro zone in 2009–2011. Unlike the crises in the 1990s, these two originated in developed countries. The GFC is considered a truly global event that affected most financial markets and industry sectors (Bekaert, Ehrmann, Fratzscher, & Mehl, 2014). The Eurozone debt crisis had major effects on some of the largest developing stock markets (Ahmad, Sehgal, & Bhanumurthy, 2013). Indeed, the first two decades of the new millennium have witnessed an increase in political turmoil and geopolitical uncertainty, coming mainly from developed countries (Hedström, Zelander, Junttila, & Uddin, 2020). Thus, this period offers a valuable opportunity to understand the transmission of contagion from advanced to emerging markets.1

Although the benefits of financial integration have been well documented in the literature, a larger exposure to contagion is an important negative side effect that should be included in a trade-off analysis (e.g. Ranciere, Tornell, & Westermann, 2006). Beyond its immediate effects, contagion may have lasting effects. Boubakri, Couharde, & Raymond (2016) argue that contagion contributes to increased risk premia, and may reverse financial integration in emerging markets, at least in the medium term. In addition, contagion is an obvious concern from a portfolio diversification standpoint. Since correlations tends to increase in bearish episodes, the construction of optimal portfolios must be adjusted to a large extent to account for contagion effects (Aït-Sahalia & Hurd, 2015; Buccioli, Kokholm, & Nicolosi, 2019).

Despite its importance, little is known about the channels through which contagion is transmitted to emerging markets. This prevents a better understanding of how stock market crashes spread abroad and what policies can be implemented to mitigate their effects. We contribute to the literature on international finance investigating the role of three types of investors in the transmission of contagion to emerging markets, using a new empirical approach. In particular, we propose a novel measure of contagion based on the concept of coexceedances (Bae, Karolyi, & Stulz, 2003). We study the dynamic interaction of this measure with each of the net buys of foreign, local institutional, and individual investors from 2007 to 2016 in seven emerging stock markets with available data.

In this way, we can test the assumption that foreign flows are the vectors of contagion to emerging markets. Foreign investors are traditionally seen as the usual suspects in the spread of contagion to emerging markets (Ocampo, 2018; Miyajima & Shim, 2014). For example, studies such as Korinek (2018), Reinhart and Reinhart (2009), and IMF (2012) discuss the economic vulnerability and other negative externalities created in emerging market economies by unrestricted capital flows. Moreover, at least four theoretical studies have modeled the propagation of contagion as a result of the rebalancing of international investor's portfolios due to different factors: shared macroeconomic risks (Kodres & Pritsker, 2002), rational herding with asymmetric information (Calvo, 2004), changes on risk aversion (Broner, Gelos, & Reinhart, 2006), and portfolio constraints (Pavlova & Rigobon, 2008). Admittedly, there is some circumstantial evidence on foreign investors transmitting negative returns to Emerging Markets as summarized in the Section 2. However, to our knowledge, no previous study offers direct evidence of a link between foreign flows and contagion. Despite this lack of evidence, policy makers and analysts uphold this assumption as part of the arguments to erect barriers to flows of foreign capital, especially during and after the GFC (e.g. Chamon & Garcia, 2016; Korinek, 2011; Gallagher, 2012).

Two studies come close to provide such evidence. Jotikasthira, Lundblad, and Ramadorai (2012), using a monthly database on mutual funds from Emerging Portfolio Fund Research (EPFR), provide evidence consistent with foreign investors transmitting shocks internationally. They identify that liquidity shocks force portfolio rebalancing across markets in the same portfolio. Puy (2016), explores the geography of contagion with the same monthly EPFR database, reporting that times of financial stress and poor macro expectations in advanced countries are associated with worldwide equity outflows.

However, we consider that using high frequency data is essential for a better analysis of the dynamics of trading activity around contagion days. Moreover, we should look beyond the role of foreign flows, who might not be the only or main vectors of contagion. Financial integration usually works both ways, smoothing not only access the entry of foreign investors to emerging markets but also the exit of local capital. In this sense, the increasing sophistication of local institutional investors in emerging economies, along with a growing connectivity with international markets, suggest that they may also play a role in transmitting contagion.2 For example, local funds with some international exposure are likely affected by the same agency costs and liquidity constraints as their counterparts in advanced markets (Barrot, Kaniel, & Sraer, 2016). Furthermore, the cross-market rebalancing effects that explain contagion in the model of Kodres and Pritsker (2002) potentially apply to both local and foreign institutional investors. In summary, although some studies indirectly suggest that foreign investors play an important part in transmitting negative shocks from developed to emerging markets, this evidence neither directly proves their role nor rules out the possible involvement of local investors, specially institutions. We intend to fill this void in the literature.

With that goal, we join two strands of the international finance literature: research on contagion and studies on capital flows in emerging markets. Firstly, inspired by the approach of Bae et al. (2003) and Mierau and Mink (2013), we propose a measure of contagion based on the probability that large daily negative returns coincide between the US and an emerging market. Following Forbes and Rigobon (2002), our measure uses a commonly agreed definition of contagion: excessive comovement between markets above and beyond what can be explained by fundamentals (Bae et al., 2003; Bekaert et al., 2014; Boyson, Stahel, & Stulz, 2010; Markwat, Kole, & van Dijk, 2009). To note, our proposed contagion measure has two characteristics: First, it only focuses on the negative coexceedances, more relevant to understand contagion episodes. Second, by controlling for global (“push” factors) and country-specific variables (“pull” factors), it takes into account well-known drivers of international fund flows (Griffin, Nardari, & Stulz, 2004; Jotikasthira et al., 2012; Puy, 2016; Sarno, Tsiakas, & Ulloa, 2016; Ülkü, 2015).

Secondly, following the flows literature (Froot, O'Connell, & Seasholes, 2001; Richards, 2005; Lee, Li, & Wang, 2010; Ülkü, 2015; Gonçalves & Eid, 2017; Gyntelberg, Loretan, & Subhanij, 2018), we study the joint dynamic of net buys by investor type and our measure of contagion, in a structural vector autoregressive model (SVAR). Besides, using alternative SVAR specifications we also explore the transmission channel of large US negative returns to emerging markets through foreign flows.

In this study we include six Asian emerging markets: Korea Stock Exchange (KSE) and Kosdaq, both from South Korea, and the main stock exchanges of Indonesia, the Philippines, Taiwan and Thailand. Data on the six Asian markets come from Bloomberg at a daily frequency, allowing for a close analysis of the dynamics between flows, returns, and episodes of contagion. Because of their monthly or quarterly frequency, other databases of foreign net flows available for a larger sample of countries are not as suitable for studying contagion dynamics, despite having been used in related studies (e.g Jotikasthira et al., 2012; Li, de Haan, & Scholtens, 2018). We also use a proprietary transactional database from the Colombia Stock Exchange (BVC) at daily frequency, where Colombia serves as an out-of-sample test of previous results focused primarily on Asian emerging markets.3

Although our sample of emerging markets is limited by data availability, it includes markets of different sizes at different stages of development, allowing some generalization of the results. According to WFE (2016), the Korea Exchange, which includes both the KSE and Kosdaq, is the world's thirteenth-largest stock market andbelongs to the the top quartile of emerging markets by market capitalization, along with Taiwan. In turn, Thailand and Indonesia are in the third quartile, whereas Colombia and the Philippines belong to the fourth. Moreover, towards the end of the sampled period, Korea, Taiwan and Thailand are classified as advanced emerging markets, while Colombia, the Philippines, and Indonesia are labeled as secondary emerging markets (FTSE, 2013; MSCI, 2016). Previous studies in the dynamics of foreign and institutional flows in several emerging markets have also used similarly varied but limited samples (Griffin et al., 2004; Múnera & Agudelo, 2022; Richards, 2005; Ülkü & Weber, 2014).

The sample period: 2007–2016, includes both the GFC and the Eurozone turmoil, events clearly exogenous to emerging markets and that significantly affected them (Ahmad et al., 2013; Bekaert et al., 2014). In this sense, this period is particularly suitable for studying the transmission of contagion from developed to emerging countries, due to the sizable global financial uncertainty and the large outflows from emerging markets, characteristic of the crises (Fratzscher, 2012).

In general, the results present an overview of the role of different agents in an emerging market on contagion days and shortly after. We find that foreign investors act as the most important transmitters of negative external returns, providing direct empirical evidence to support this widespread assumption. This result is statistically significant and long lasting (at least 20 days) in all seven emerging markets studied. Local institutional investors are also aggressive sellers on the contagion day, but they tend to become opportunistic buyers soon afterward. Retail investors are net buyers on contagion days, providing liquidity, but they do not exhibit a definite buying or selling trend afterwards.

We contribute to the related literature by proposing a novel contagion measure, and using it to provide direct evidence of the role of different types of investors as contagion drivers. As previously stated, this allows us to test the implications of extant theoretical models of international funds as transmitters of shocks between financial markets (Broner et al., 2006; Calvo, 2004; Kodres & Pritsker, 2002; Pavlova & Rigobon, 2008). Moreover, we provide evidence of the process by which foreigners transmit contagion: we show that large US stock market drops lead foreign outflows from emerging markets, and that those outflows move emerging market prices, all this to a greater extent than in normal times. Finally, we provide evidence of the role of local institutional and individual investors in and after contagion days.

The remainder of this article is organized as follows. Section 2 presents an overview of the relevant literature. Section 3 describes the databases used. Section 4 explains the methodology, the proposed contagion measure, and the different SVAR specifications. Section 5 presents and discusses the results of the SVAR models. Finally, Section 6 concludes.

Section snippets

Background

Our primary objective is to test whether foreigners are the main vectors of contagion in emerging countries. Since the 90's crises, foreign portfolio investors have been regarded as the culprits of bringing turmoil to emerging markets (Sachs, Tornell, Velasco, Calvo, & Cooper, 1996; Stiglitz, 1999, Stiglitz, 2000), and some early evidence, focused on the financial crises of the 1990s, points in this direction (Borensztein & Gelos, 2003; Kim & Wei, 2002).

Furthermore, empirical evidence suggests

Data

This study focuses on six emerging countries, represented by their main indexes: Colombia (Colcap), Indonesia (JCI), KSE (Kospi), Kosdaq (Kosdaq), the Philippines (PSEi), Taiwan (TWSE), and Thailand (SET50). Where the data is available, we study the trading by each of the three types of investors in each market: foreign investors, local individual investors, and local institutional investors. We have daily net purchasing data for all three types of investors at the market level in Colombia,

Contagion measure

Because our main objective is to determine the type of investors that drive financial contagion in an emerging market, we first need to decide how to measure contagion. Several methods have been proposed in the literature. Corsetti, Pericoli, and Sbracia (2005) and Forbes and Rigobon (2002) measure contagion with the heteroskedasticity-adjusted correlation between stock market returns. Bae et al. (2003), in turn, use the coincidence between extreme returns across stock markets (coexceedances)

Who drives contagion?

Our main interest is to identify the drivers of contagion among the three groups of investors, that is, whose net sales increase the probability of a contagion episode. To do this, we observe the response of the contagion measure to an innovation in net buys, using an impulse response analysis of the structural 4-SVAR in eq. (3) for each country.8

Conclusion

In this study, we tested the differential role of foreign investors, local institutional investors, and local individual investors in times of market contagion. We report new evidence to confirm that the behavior of foreign investors is consistent with the role usually attributed to them: their sales increase the probability of contagion, and they tend to keep selling even after the contagion has ended. Furthermore, we confirm that foreign investors transmit contagion from advanced countries

CRediT authorship contribution statement

Diego A. Agudelo: Conceptualization, Supervision, Funding acquisition, Resources, Methodology, Formal analysis, Project administration, Writing – original draft, Writing – review & editing. Daimer J. Múnera: Data curation, Methodology, Software, Formal analysis, Writing – original draft, Visualization, Validation, Writing – review & editing.

Acknowledgements

This research was funded by an Universidad EAFIT grant. We acknowledge useful suggestions from Gustavo Canavire-Bacarreza, Carlos Castro, Hernán Herrera-Echeverri, Jesus Otero, Andrés Ramirez-Hassan, Diego Restrepo-Tobón, Hermilson Velasquez and seminar participants in Universidad del Rosario, Universidad EAFIT, the Workshop on Innovative Methods in Finance and Economics, Bilgi University, the III ICASQ Conference and the IFABS 2019. We are also grateful to Adriana Cárdenas, for providing data

References (78)

  • J.J. Choi et al.

    Are individual or institutional investors the agents of bubbles?

    Journal of International Money and Finance

    (2015)
  • G. Corsetti et al.

    “Some contagion, some interdependence”: More pitfalls in tests of financial contagion

    Journal of International Money and Finance

    (2005)
  • H.J. Edison et al.

    Cross-border listings, capital controls, and equity flows to emerging markets

    Journal of International Money and Finance

    (2008)
  • M. Fratzscher

    Capital flows, push versus pull factors and the global financial crisis

    Journal of International Economics

    (2012)
  • K.A. Froot et al.

    The portfolio flows of international investors

    Journal of Financial Economics

    (2001)
  • W. Gonçalves et al.

    Sophistication and price impact of foreign investors in the Brazilian stock market

    Emerging Markets Review

    (2017)
  • M. Grinblatt et al.

    The investment behavior and performance of various investor types: A study of Finland’s unique data set

    Journal of Financial Economics

    (2000)
  • Y. Guo et al.

    Tail risk contagion between international financial markets during COVID-19 pandemic

    International Review of Financial Analysis

    (2021)
  • J. Gyntelberg et al.

    Private information, capital flows, and exchange rates

    Journal of International Money and Finance

    (2018)
  • J. Hincapié-Salazar et al.

    Is the disposition effect in bonds as strong as in stocks? Evidence from an emerging market

    Global Finance Journal

    (2020)
  • G. Kaminsky et al.

    Managers, investors, and crises: Mutual fund strategies in emerging markets

    Journal of International Economics

    (2004)
  • G.L. Kaminsky et al.

    On crises, contagion, and confusion

    Journal of International Economics

    (2000)
  • W. Kim et al.

    Foreign portfolio investors before and during a crisis

    Journal of International Economics

    (2002)
  • A. Korinek

    Regulating capital flows to emerging markets: An externality view

    Journal of International Economics

    (2018)
  • B.S. Lee et al.

    The dynamics of individual and institutional trading on the Shanghai stock exchange

    Pacific-Basin Finance Journal

    (2010)
  • S. Li et al.

    Surges of international fund flows

    Journal of International Money and Finance

    (2018)
  • P.X. Liew et al.

    Foreign equity flows: Boon or bane to the liquidity of Malaysian stock market?

    North American Journal of Economics and Finance

    (2018)
  • T. Markwat et al.

    Contagion as a domino effect in global stock markets

    Journal of Banking & Finance

    (2009)
  • J.O. Mierau et al.

    Are stock market crises contagious? The role of crisis definitions

    Journal of Banking & Finance

    (2013)
  • D. Puy

    Mutual funds flows and the geography of contagion

    Journal of International Money and Finance

    (2016)
  • R. Ranciere et al.

    Decomposing the effects of financial liberalization: Crises vs. growth

    Journal of Banking & Finance

    (2006)
  • J.C. Rodriguez

    Measuring financial contagion: A copula approach

    Journal of Empirical Finance

    (2007)
  • L.P. Samarakoon

    Stock market interdependence, contagion, and the U.S. financial crisis: The case of emerging and frontier markets

    Journal of International Financial Markets, Institutions and Money

    (2011)
  • L. Sarno et al.

    What drives international portfolio flows?

    Journal of International Money and Finance

    (2016)
  • J.E. Stiglitz

    Capital market liberalization, economic growth, and instability

    World Development

    (2000)
  • G.S. Uddin et al.

    Stock market contagion of COVID-19 in emerging economies

    International Review of Economics & Finance

    (2022)
  • N. Ülkü

    The interaction between foreigners' trading and stock market returns in emerging Europe

    Journal of Empirical Finance

    (2015)
  • H.Y. Yu et al.

    The effect of attention on buying behavior during a financial crisis: Evidence from the Taiwan stock exchange

    International Review of Financial Analysis

    (2010)
  • Y. Aït-Sahalia et al.

    Portfolio choice in markets with contagion

    Journal of Financial Econometrics

    (2015)
  • View full text