The contagion versus interdependence controversy between hedge funds and equity markets

• Published date: 01 June 2018
• Date: 01 June 2018
• DOI: http://doi.org/10.1111/eufm.12125
• Author: Hee Soo Lee,Tae Yoon Kim

DOI: 10.1111/eufm.12125

ORIGINAL ARTICLE

The contagion versus interdependence controversy

between hedge funds and equity markets

Tae Yoon Kim

1

|

Hee Soo Lee

2

1

Department of Statistics, Keimyung

University, Daegu, South Korea

Email: tykim@kmu.ac.kr

2

Department of Business Administration,

Sejong University, Seoul, South Korea

Email: heesoo@sejong.ac.kr

Abstract

This study considers the ‘contagion versus interdependence’

controversy between hedge funds and equity markets. We find

that contagion effects break down the established interdepen-

dence between hedge funds and equity markets and

conditional return smoothing could play a key role in the

contagion process by increasing or decreasing the contagion

likelihood during crisis and prosperity. It is noted that the

return smoothing tends to produce a biased pattern of returns

during crisis and a decreased amount of return during

prosperity. These findings are obtained by linking a single

equation error correction model to a factor model and carrying

out quantile regression, Z-test and Wald–Wolfowitzruns tests.

KEYWORDS

conditional return smoothing, contagion, factor model, hedge funds,

interdependence, single equation error correction model

JEL CLASSIFICATION

C22,C58,G01

1

|

INTRODUCTION

In a simple mean-variance framework, because hedge funds are not strongly correlated with equity

markets, investors have used such funds to minimise risk in a diversified portfolio. However, in the

The authors thank the Editor, Professor John Doukas and two anonymous referees for valuable comments and suggestions.

The authors also thank Professor Young Ho Eom and seminar participants at Yonsei University, Seoul National University,

and Finance Symposium. T.Y. Kim’s work is supported by the National Research Foundation of Korea (NRF

2016R1D1A1B03934375). H.S. Lee’s work is supported by the IREC, The Institute of Finance and Banking, Seoul

National University and Sejong University.

Eur Financ Manag. 2018;24:309–330. wileyonlinelibrary.com/journal/eufm © 2017 John Wiley & Sons, Ltd.

|

309

wake of the global financial crisis that substantially affected the markets between 2007 and 2009,

investors, regulators, and the financial press began to express concerns about particular hedge fund

bankruptcies and to criticise the hedge fund industry in general. Subsequently, researchers started to

explore whether hedge funds had experienced contagion from representative markets and within hedge

fund styles during the crisis. Unsurprisingly, these studies generated different results from those

presented by pre-crisis research. For example, whereas Boyson, Stahel, and Stulz (2006) and Li and

Kazemi (2007) provide no evidence of contagion between equity markets and hedge funds (i.e., before

the global financial crisis), Viebig and Poddig (2010) produce empirical evidence that a strong

contagion effect exists between equity markets and several hedge funds during periods of extreme

stress in financial markets. These conflicting results on hedge funds and equity markets have led us to

investigate the so-called ‘contagion versus interdependence’controversy. This controversy is in fact

closely related to the controversy introduced by Forbes and Rigobon (2002) for international equity

markets. Forbes and Rigobon (2002) defined contagion as conditional correlation during crisis, while

they defined interdependence as unconditional correlation formed between international equity

markets. In a similar spirit, regarding contagion between equity markets and hedge funds, Boyson,

Stahel, and Stulz (2010) defined contagion as correlation over and above that expected from economic

fundamentals; they measure it by clustering the bottom 10% of hedge fund returns across all hedge fund

styles. Note that by clustering the bottom 10% of hedge fund returns, Boyson et al. (2010) explicitly

focus conditional correlation on crisis in the context of contagion, while they implicitly refer

interdependence to unconditional correlation in the underlying fundamentals.

Essentially, the contagion versus interdependence controversy for international equity markets

poses two empirically intriguing questions regarding excessive correlations across markets during

crisis periods: does excessive correlation exceed that in the underlying fundamentals across markets (or

is the contagion effect identified by a breakdown of the established interdependence between the two

markets)? And, is this excessive correlation driven by strategic complementarities between markets? In

this study, we consider similar questions regarding contagion between hedge funds and equity markets.

More precisely, we consider two specific questions during periods of both economic crisis and

economic prosperity: is the contagion effect identified by a breakdown of the established

interdependence between hedge funds and equity markets? And, is this breakdown driven by

strategic complementarities in hedge fund markets?

To investigate the two questions above, we link the single equation error correction model

(SEECM) to the latent factor model, and then implement quantile regression and the Z-test together

with the Wald–Wolfowitz runs test. The SEECM is used to handle contagion dynamics as well as

interdependence simultaneously. Instead of intuitively defining interdependence as correlation in the

underlying fundamentals, as done by Boyson et al. (2010), the SEECM defines it econometrically as

the correlation adjusted by the corresponding market volatilities in the sense of Forbes and Rigobon

(2002). Recall that the usual correlation calculation is subject to bias due to increased volatility and is to

be adjusted by the volatility particularly during crisis periods (see Corsetti, Pericoli, & Sbracia, 2005;

Forbes & Rigobon, 2002 for related discussions).

1

Linking the latent factor model to the SEECM also

allows us to measure the effect of conditional return smoothing (the tendency of hedge funds to

underreport losses rather than gains) more accurately. Conditional return smoothing is a well-known

and widespread strategic complementarity among hedge funds (Bollen & Pool, 2008).

The major finding of this paper is that contagion effects break down the established

interdependence between hedge funds and equity markets and that conditional return smoothing

......................................................................................................................................................................................................................................

1

See Remark 1 below.

310

|

KIM AND LEE