Over‐investment or risk mitigation? Corporate social responsibility in Asia‐Pacific, Europe, Japan, and the United States

• Author: Sebastian Utz
• DOI: http://doi.org/10.1016/j.rfe.2017.10.001
• Published date: 01 April 2018
• Date: 01 April 2018

ORIGINAL ARTICLE

Over-investment or risk mitigation? Corporate social

responsibility in Asia-Pacific, Europe, Japan, and the United

States

Sebastian Utz

School of Finance, University of St.

Gallen, 9000 St. Gallen, Switzerland

Correspondence

Sebastian Utz, School of Finance,

University of St.Gallen, 9000 St. Gallen,

Switzerland.

Email: sebastian.utz@unisg.ch

Abstract

We study the relationship of corporate social responsibility (CSR) and the distri-

bution of stock returns for an international sample. Firms with a high level of

CSR generally exhibit superior stock price synchronicity in the markets of Eur-

ope, Japan, and the United States. In particular, we identify optimal levels of

CSR to minimize idiosyncratic risk for each region. Moreover, CSR has a mitigat-

ing effect on crash risk in Europe and the United States. In contrast, firms from

the Asia-Pacific region display CSR over-investment followed by a higher crash

risk. This appears to be a consequence of globalization, which forces firms from

Asia-Pacific to overinvest in CSR to adapt western standards.

JEL CLASSIFICATION

G19, M14, G12

KEYWORDS

Corporate social responsibility, Crash risk, Stock synchronicity

1

|

INTRODUCTION

Financial analysts and economists have attached great importance to stock market movements and stock price changes con-

nected to new market-wide, industry-wide, and firm-specific information. Stock price synchronicity (Roll, 1988) and mea-

sures for crash risk (see Chen, Hong, & Stein, 2001) are applied in order to review the relationship between the appearance

of firm-specific information and stock price reactions. In perfectly diversified portfolios, idiosyncratic risk is unimportant.

Nevertheless, equity portfolios of private investors are under-diversified with an average of seven stocks (Goetzm ann &

Kumar, 2008) and therefore idiosyncratic risk needs to be considered. Brown and Kapadia (2007), and Fu (2009) indicate

additional reasons for the importance of managing idiosyncratic risk for investors. In their empirical studies, Kim, Li, and

Li (2014) and Lee and Faff (2009) show the mitigating effect of corporate social responsibility (CSR) on idiosyncratic risk

for U.S. firms. In this paper, we consider the link between CSR and the distribution of stock returns within an international

cross-section. We calculate measures for idiosyncratic risk and crash risk and find that a firm’s commitment to high CSR

standards is a significant predictor for these quantities. To the best of our knowledge, this is the first study to test for and

document the CSR-effect on the distribution of daily stock returns for the four major developed markets.

There is no valid universal definition of the concept of CSR (see Griffin, 2000; van Beurden & G€

ossling, 2008). CSR is

not only a concept that considers the non-financial aspects of a firm’s business strategy, but also a way of interacting with

stakeholders (Oikonomou, Brooks, & Pavelin, 2014). We follow the common approach to measure CSR as being the aver-

age score of the two dimensions of environment and social (El Ghoul, Guedhami, & Kim, 2017; El Ghoul, Guedhami,

First published online by Elsevier on behalf of The University of New Orleans, 8 October, 2017, https://doi.org/10.1016/j.rfe.2017.10.001

Received: 27 April 2017

|

Revised: 3 August 2017

|

Accepted: 2 October 2017

DOI: 10.1016/j.rfe.2017.10.001

Rev Financ Econ. 2018;36:167–193. wileyonlinelibrary.com/journal/rfe ©2017 The University of New Orleans

|

167

Wang, & Kwok, 2016). Our endeavor is to analyze the CSR-effect on the distribution of stock returns for cross-sections of

publicly traded firms from the four major developed markets: Asia-Pacific, Europe, Japan, and the United States (U.S.).

Therefore, we require international CSR data with an appropriate level of variability to attain statistical inference. Since

MSCI-KLD provides binary data for U.S. firms, we use data from Asset4, which offers continuous scores updated on an

annual basis for a global rating universe.

A high level of CSR is a possible solution to pleasing investors’claims for solving agency problems. Accordingly,

socially responsible investments have increased significantly over time and many firms have contributed to this trend

through professional CSR management and reporting. Approximately $6.57 trillion AUM were managed in a sustainable

way in 2014 in the U.S. (US, 2015).

1

Moreover, the signatories of the global initiative Principles of Responsible Invest-

ments, launched by the United Nations, amount to more than $59 trillion AUM worldwide. Signatories commit themselves

to being active owners and to incorporating CSR issues into their ownership policies and practices. An important pillar of

active ownership is shareholder engagement, which is a fast-growing trend of shareholders’monitoring with respect to CSR

aspects.

We hypothesize that a high level of CSR, which limits managers’concealment of firm-specific information, incre ases

the R

2

of a firm’s stock price. Strong CSR is an instrument employed to restrain managers’extraction of private benefits

and to increase firm transparency (cf. El Ghoul, Guedhami, Nash, & Patel, in press). A discussion on whether lower opac-

ity leads to higher idiosyncratic risk (Hutton, Marcus, & Tehranian, 2009) or not (Datta, Iskander-Datta, & Singh, 2014)

prevails. Nevertheless, following our theoretical framework, a higher level of transparency drive n by CSR results in more

predictable cash flows as unexpected firm-specific information is improbable. Therefore, firms with high CSR are less

prone to deviating from the market; in consequence they have a high R

2

. Through implication, firms with low CSR are

exposed to a lax monitoring of stakeholders and therefore tend to be less predictable, yielding a low R

2

. Our results support

the R

2

-increasing effect of high CSR as we find a significantly lower idiosyncratic risk in the U.S., Europe, and Japan.

Along with the rise of CSR campaigns, several derogatory opinions and reluctance regarding the benefits of CSR have

emerged. From an economic point of view, shareholders’expectations in managers primarily concern the maximization of

long-term returns while the distribution of investment dollars to CSR projects is not efficient (Friedman, 1970). In parti cu-

lar, since firms’financial resources are limited, costly CSR programs also compete with other critical marketing instrum ents

such as advertising or research and development. Critics claim that CSR does not maximize the firm’s long-term stock

wealth (over-investment view).

In contrast to this over-investment view, recent literature (Stellner, Klein, & Zwergel, 2015) reports upon a risk mitiga -

tion effect of CSR on credit risk. As a measure for this risk mitigating effect, we analyze the influence of CSR on crash

risk. Jin and Myers (2006) predict that individual firm stock price crashes occur when accum ulated negative firm-specific

information suddenly becomes publicly available. If investors’expectations of cash flows are higher than the actual cash

flows itself, managers conceal the bad news to protect their jobs (Jin & Myers, 2006). When the accumulated negative

information finally crosses a tipping point, managers cease trying to hide the information and all bad news is released at

once, consequently resulting in a stock price crash. In an investor’s portfolio which lacks a certa in extent of diversification,

a crashing stock destroys a large proportion of wealth all at once. Since a sustainable business approach and monitoring by

socially responsible investors attenuate bad-news hoarding, we expect a negative relationship between CSR and stock price

crash risk. By using three alternative measures of firm-specific crash risk, we find that a negative relationship between

CSR and crash risk in the U.S. and the European sample exists. In Japan, virtually no relationship can be found. The Asia-

Pacific sample displays a positive relationship.

In summary, we demonstrate that measures of CSR credibly forecast both stock price synchronicity and crash risk. Our

results for the U.S. sample are similar to those of Kim et al. (2014), which are based on weekly returns and MSCI-KLD

CSR data. A high level of CSR is associated with a decrease in the risk of stock price crashes. In a further analysis, we

find that even both dimensions (i.e. environment and social) of CSR separately have a positive significant impact on stock

price synchronicity. Our findings also support the risk mitigation concept of CSR for Europe and United States. Since CSR

and crash risk are positively associated in Asia-Pacific, the findings support the over-investment hypothesis. This is consis-

tent with the globalization forces-explanation for firms from Asia-Pacific adopting to western CSR standards (Chapple &

Moon, 2005). Moreover, the effect of CSR on crash risk occurs at a higher extent for large firms, i.e. there is strong evi-

dence for the risk mitigation hypothesis in Europe and the United States and for the over-investment hypothesis in Asia-

Pacific.

Our results are significant for several reasons: First, they add insights of the proces s through which information is dis-

closed to the marketplace. Several studies (Bekaert & Wu, 2000; Campbell & Hentschel, 1992; French , Schwert, & Stam-

baugh, 1987) show that stock prices are more prone to large downward movements than to upward ones and that this

168

|

UTZ