Equity flows, stock returns and exchange rates

• Author: Angelos Kanas,Sotirios Karkalakos
• DOI: http://doi.org/10.1002/ijfe.1574
• Published date: 01 April 2017
• Date: 01 April 2017

RESEARCH ARTICLE

Equity flows, stock returns and exchange rates

Angelos Kanas | Sotirios Karkalakos

Economics, University of Piraeus, Piraeus, Greece

Correspondence

Angelos Kanas, Economics, University of Piraeus,

Piraeus, Greece.

Email: akanas@unipi.gr

JEL Classification: F30; G11; G12

Abstract

We explore the effects of equity flows between U.S. and U.K. investors upon equity

and exchange rate returns within a unified empirical framework on the basis of a

trivariate vector autoregressive system that incorporates mean and volatility spill-

overs and allows for dynamic conditional correlations. Our findings are as follows:

First, we reveal strong evidence of volatility spillovers across equity returns,

exchange rate returns, and equity flows. Second, we find strong evidence that

U.K. investors rebalance their portfolios by engaging in a positive feedback trading

known in the literature as “trend chasing.”Third, we document strong dynamic

effects from net flows to equity returns, illustrating a trading rule that portfolios

are dynamically adjusted over a short‐run horizon influencing changes in stock

returns. Last, correlation uncertainty appears to be reduced from the start of the

1990s onwards.

KEYWORDS

dynamic correlations, equity returns, exchange rates, net flows, trading, volatility

1|INTRODUCTION

Cross border capital flows have attracted attention in the liter-

ature due to their significant increase over the recent period

(Caporale, Ali, & Spagnolo, 2015). The present paper deals

with equity flow movements between the United States and

the United Kingdom, using a unified framework for the joint

determination of equity flows, exchange rates, and equity

returns. A vector autoregressive (VAR) system is employed,

with explicit multivariate generalized autoregressive condi-

tional heteroskedasticity (GARCH) modeling of conditional

variances, volatility spillovers, and allowance for dynamic

conditional correlations (DCCs) among the variables in hand.

The paper’s contribution is twofold. First, we explicitly

model dynamic volatility spillovers between these variables.

Incorporating conditional variance interactions is important

for two reasons. According to Hau and Rey (2006), the typi-

cal foreign equity investor holds currency risk and equity risk

as a bundle (page 276). As markets are not complete (in

which investors could swap and eliminate risk), exposure to

risk implies that the international investor cares about both

the volatilities of exchange rate and equity returns and for

the correlation structure of exchange rates and equity returns.

Another reason is that explicit modeling of volatility linkages

helps us measure more accurately correlations, which

constitute an important decision variable for international

investors. Importantly, using a trivariate vector

autoregressive‐generalized autoregressive conditional

heteroskedasticity (VAR‐GARCH) framework, we simulta-

neously explore mean spillovers between net equity flows,

exchange rate returns, and equity returns and thus, extend

previous results from regression and VAR approaches

(Ahearne, Griever, & Warnock, 2004; Brennan & Cao,

1997; Froot, O’Connell, & Seasholes, 2001; Hau & Rey,

2006). To our knowledge, this is the first study to look at

both mean and volatility spillovers.

Second, in our trivariate VAR‐GARCH model, we allow

for DCCs. According to Hau and Rey (2004), conditional

correlations between exchange rate returns, equity returns,

and equity flows are important and carry implications regard-

ing portfolio rebalancing. Importantly, the dynamic time‐vary-

ing conditional correlations obtained are free of any mean and

volatility spillovers as the latter have been explicitly modeled.

Our approach extends Hau and Rey’s (2006) time‐invariant

unconditional correlations, by showing that the average corre-

lation coefficient for the net flows—exchange rate changes

and for the returns differential—exchange rate changes are

lower than those found by Hau and Rey (2006). Dynamic

Received: 2 March 2015 Revised: 1 March 2016 Accepted: 2 January 2017

DOI: 10.1002/ijfe.1574

Int J Fin Econ. 2017;22:159–168. Copyright © 2017 John Wiley & Sons, Ltd.wileyonlinelibrary.com/journal/ijfe 159