The low beta anomaly: A corporate bond investor's perspective

• Published date: 01 October 2018
• DOI: http://doi.org/10.1002/rfe.1022
• Date: 01 October 2018
• Author: Demir Bektić

ORIGINAL ARTICLE

The low beta anomaly: A corporate bond investor’s perspective

Demir Bekti

c

1,2

1

Department of Law and Economics,

Darmstadt University of Technology,

Darmstadt, Germany

2

Deka Investment GmbH and IQ-KAP,

Frankfurt am Main, Germany

Correspondence

Demir Bekti

c, Department of Law and

Economics, Darmstadt University of

Technology, Darmstadt, Germany.

Email: demir.bektic@gmail.com

Abstract

The low beta anomaly is well documented for equity markets. However, the exis-

tence of such a factor in corporate bond markets is less explored. I find that Euro-

pean corporate bonds of firms with a low equity beta have higher risk-adjusted

returns, on average, than European corporate bonds of firms with a high equity

beta. The results are economically and statistically significant as low beta credit

portfolios improve the Sharpe ratio up to 30%. Moreover, even after accounting

for transaction costs and by considering long-only portfolios, the risk-adjusted

return remains substantial indicating practical implementability of the strategy for

corporate bond investors.

JEL CLASSIFICATION

G11, G12, G14, G15

KEYWORDS

anomalies, corporate bonds, factor investing, low beta, risk premium

1

|

INTRODUCTION

Although the capital asset pricing model (CAPM) assumes a positive risk/return relation, there are numerous empirical stud-

ies which report that the relation between risk and return is flatter than implied by CAPM or even negative. One of the first

and best documented anomalies in equity markets is the low beta anomaly.

1

For instance, Haugen and Heins (1972) and

Black, Jensen, and Scholes (1972) show that a portfolio that is short high-beta stocks and long low-beta stock s generates

significant positive abnormal returns. Recent research confirms these puzzling findings and tries to give an explanation for

the existence of this kind of anomaly. Frazzini and Pedersen (2014) provide a model which suggests that investment con-

straints (e.g., leverage) and human behavior are the main drivers behind the low beta anomaly.

However, while empirical research on anomalies is predominantly focused on equit y markets, similar studies for corpo-

rate bonds are rather new. For example, Correia, Richardson, and Tuna (2012) document that value investing is useful for

corporate bond investors. Jostova, Nikolova, Philipov, and Stahel (2013) show that credit momentum is profitable in the

U.S. high yield (HY) corporate bond market. Choi and Kim (2016) suggest that asset growth and investment anomalies

exist in credit markets and Chordia, Goyal, Nozawa, Subrahmanyam, and Tong (2017) provide evidence that size, prof-

itability and past equity returns can predict corporate bond returns. In addition, Bekti

c, Wenzler, Wegener, Schiereck, and

Spielmann (2016) state that the four Fama and French (2015) equity factors (size, value, profitability and investment) exhi-

bit significant excess returns in the U.S. HY market but the results are mixed for U.S. and European investment grade (IG)

bonds. Finally, Lin, Wu, and Zhou (2017) document that momentum is the most pronounced cross-sectional corporate bond

market anomaly using trend signals.

Analyzing the relation between firm’s equity beta and the corresponding corporate bond excess return is promising for

corporate bond investors for at least the following two reasons. On one hand, according to structural credit risk models like

Merton (1974), equities and bonds should be related since they represent claims against the assets of the same firm, and

therefore, this approach should translate to the returns of corporate bonds as well. According to the structural credit risk

Received: 2 January 2018

|

Revised: 19 February 2018

|

Accepted: 8 March 2018

DOI: 10.1002/rfe.1022

300

|

©2018 The University of New Orleans wileyonlinelibrary.com/journal/rfe Rev Financ Econ. 2018;36:300–306.