Decomposition of systematic and total risk variations in emerging markets

• Author: Meng‐Horng Lee,Chee‐Wooi Hooy,Robert Brooks
• DOI: http://doi.org/10.1111/infi.12127
• Published date: 01 June 2018
• Date: 01 June 2018

DOI: 10.1111/infi.12127

ORIGINAL ARTICLE

Decomposition of systematic and total risk

variations in emerging markets

Meng-Horng Lee

1

|

Chee-Wooi Hooy

2

|

Robert Brooks

3

1

Research Department, Hong Leong

Investment Bank, Kuala Lumpur,

Malaysia

2

School of Management, Universiti Sains

Malaysia, Penang, Malaysia

3

Department of Econometrics and

Business Statistics, Monash University,

Melbourne, Victoria, Australia

Correspondence

Chee Wooi, Hooy, School of

Management, Universiti Sains Malaysia,

11800 USM, Penang, Malaysia.

Email: cwhooy@usm.my

Funding information

The Ministry of Higher Education,

Grant number: 203/PMGT/6711253

Abstract

This paper examines the country-versus-industry debate in

international diversification in the context of systematic

risk. We employ the popular decomposition approach

proposed by Heston and Rouwenhorst and cover 2,045

individual firms from 36 emerging countries across 39

industries. Generally, we find a domination of country

factors in the sample period of 1990–2012, particularly after

the 1997 Asian Financial Crisis, which implies that country-

based diversification is still superior to industry-based

diversification in emerging markets. In addition, we

document high variability of the common factor coupled

with a diminishing trend of country and industry factors,

which limits the desired risk reduction from diversification

in emerging markets. A convergence between systematic

and total risk is found beginning in 2007, which coincides

with the timing of the Global Financial Crisis.

1

|

INTRODUCTION

International diversification is grounded in the notion that country-specific factors such as fiscal and

monetary policies or the characteristics of financial systems affect the company's financial

performance and stock returns. However, with the intensification of world financial market

integration over the past few decades, corporate performance has been increasingly exposed to

worldwide dynamics and has become less sensitive to domestic economic shocks. As a result, the

potential of country-based diversification has declined, and industry diversification has become more

important. However, empirical studies on this topic are inconclusive, particularly in the context of

emerging markets.

158

|

© 2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/infi International Finance. 2018;21:158–174.

The extent to which emerging stock markets are driven by country factors remains an

important question that must be addressed in order to gauge their financial market integration

with the world. Heston and Rouwenhorst (1994) suggested a factor decomposition approach (the

HR approach henceforth) to examine the relative importance of country versus industry factors

in explaining stock return variation. The HR decomposition approach has become a popular

research method in this area; it has been applied in studies such as Rouwenhorst (1999), Serra

(2000), Brooks and Del Negro (2004), Ferreira and Ferreira (2006), Campa and Fernandes

(2006), Bai and Green (2010, 2012), Lee and Hooy (2013), and Chou, Zhao, and Suardi (2014).

The HR approach has also been extended in studies such as Baele, Ferrando, Hördahl, Krylova,

and Monnet (2004) and Pieterse-Bloem, Qian, Verschoor, and Zwinkels (2016) to decompose

the country and industry factors in European corporate bonds.

So far, almost all studies have utilized the HR approach to decompose cross-sectional stock returns

and bond yields, with the exception of Bai and Green (2010) and Anginer and Demirgüç-Kunt (2014).

Bai and Green (2010) employed the HR method to decompose the total risk of 1,537 firms from 13

emerging countries with 11 industry identities, while Anginer and Demirgüç-Kunt (2014) used the HR

method to decompose the default risk of 1,866 banks over 65 countries. As argued in Bai and Green

(2010), investors who pursue a diversification strategy based exclusively on the return decomposition

may simply be shifting towards securities with greater total risk, without necessarily achieving the

desired overall risk reduction from diversification. This provides a strong basis for scholars to

investigate diversification strategy from the risk perspective instead of focusing only on stock return.

Focusing on total risk decomposition, however, may not be effective. This is because non-systematic

risk is negligible in a well-diversified portfolio, and the systematic variation of returns remains the only

priced risk to which rational investors should be paying attention. Ryan (1997) noted that systematic

risk is considered more analytically tractable than total risk because the systematic risk of a portfolio is

represented by a value-weighted average, whereas the total risk of a portfolio depends on the variances

and covariances of the portfolio components. In addition, Eun and Lee (2010) argued that the risk-

return characteristics of stock markets tend to diverge internationally when markets are volatile,

especially in the case of emerging markets. Thus, investors and portfolio managers should be

concerned with the systematic risk of their particular portfolio and demand a sufficient risk premium to

compensate for bearing the systematic risk.

This paper extends the country- versus-industry debate in th e context of systematic risk. We

employ the HR decomposition ap proach with a large coverage of 36 emerging markets across 39

industries and monthly data ov er a long period, 1990–2012. Emerging markets provide a go od scope

for the study as they are not fully in tegrated in world capital markets (Bai & Green, 2010). With th e

increase in financial libera lization, integration, and e conomic turbulence over the pas t few decades,

it would be interesting to exa mine how the dynamics of the deco mposed factors affect firms '

systematic risk in emergin g markets. By knowing the sour ces of systematic risk, int ernational

investors and portfolio man agers can then align their str ategy to assume more or less risk according

to their risk-aversion level. W e attribute the systematic risk of e merging markets to three main

orthogonal factors: worl d, country, and industry fact ors. By definition, these factors are constructed

in such a way that country factors take in to account international diff erences in industry

compositions, and industry facto rs take into account inter-industr y differences in country

compositions compared with a ba se (world) factor.

Our paper is novel in several ways. First, moving from return decomposition to risk

decomposition, we provide the first empirical results regarding the use of HR decomposition on

systematic risk instead of stock returns or total risk, and we also provide a comparison with total

risk decomposition, as done in Bai and Green (2010). Second, we employ comprehensive data on

LEE ET AL.

|

159