The size anomaly in European financial stock returns

• Published date: 01 August 2019
• DOI: http://doi.org/10.1111/infi.12330
• Date: 01 August 2019
• Author: Sander Muns

DOI: 10.1111/infi.12330

ORIGINAL ARTICLE

The size anomaly in European financial stock

returns

Sander Muns

Tilburg University, Netspar, Tilburg,

Netherlands

Correspondence

Sander Muns, Tilburg University,

Netspar, Tilburg, Netherlands.

Email: s.muns@uvt.nl

Abstract

Large financial institutions in Europe earn significantly

lower risk-adjusted returns than their smaller counterparts.

This pattern is absent in other industries. We interpret this as

evidence of the latent government guarantees that protect

large European financial institutions from tail events. Panel

regressions suggest that unconditional on distress, the

expected guarantees increase with size, leverage, and the

government debt-to-GDP ratio. These determinants can be

associated with a higher likelihood of being involved in a

systemic crisis and thus receiving more government

guarantees.

1

|

INTRODUCTION

The financial crisis showed that the impact of tail risk on society can be enormous. Governments bailed

out the systemically most important financial institutions to mitigate the impact of the most extreme

outcomes. Regulatory authorities adopted banking regulations (e.g., the Bank Recovery and

Resolution Directive) that enhance banking supervision and prescribed bail-in mechanisms to protect

taxpayers from expensive bailout events. However, the bailouts in the Italian banking system in mid-

2017 indicate that such bailouts are still an attractive resolution mechanism for supervisory authorities.

The impact of government guarantees is not immediately obvious. Moral hazard may induce

protected institutions to opt for riskier investments or higher leverage. On the one hand, the standard

risk-return trade-off in the capital asset pricing model (CAPM) predicts a higher expected return on the

assets of such institutions. On the other hand, investors may anticipate the implicitly embedded put

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction

in any medium, provided the original work is properly cited.

© 2018 The Authors. International Finance Published by John Wiley Sons Ltd

International Finance. 2018;1–16. wileyonlinelibrary.com/journal/infi

|

1

155

International Finance. 2019;22:155–170. wileyonlinelibrary.com/journal/infi

&

option of the protection. Similar to an insurance premium, investors may demand a lower expected

return on an investment in a protected financial institution. As a consequence, a financial institution

facing both a higher volatility and a lower financing cost may enjoy implicit government guarantees.

Several characteristics may suggest the presence of government guarantees. In addition to riskier

assets and a higher leverage, another potential indicator of government guarantees is bank size.

Although not always explicitly formalized, the “too-big-to-fail”(TBTF) concern is generally accepted

(Laeven, Ratnovski, & Tong, 2016; Stern & Feldman, 2009). This suggests that size indicates systemic

importance (Huang, Zhou, & Zhu, 2012) and therefore a higher likelihood of government guarantees in

case of distress. This systemic importance may arise in several ways. First, since large financial

institutions are more involved in interbank activities (Furfine, 2003) and share similarities in terms of

their diversified investment portfolios (Acharya, 2009; Simaan, 2017), a distressed large financial

institution is more likely to be associated with multiple distressed financial institutions. Second, a

default of a large financial institution entails a default on large amounts of debt, which can have a

substantial direct impact on the financial system and a substantial indirect impact through spillover

effects and banking panics (Diamond & Dybvig, 1983). Third, since large financial institutions tend to

invest in sovereign bonds, a feedback loop between public finances and large financial institutions can

emerge (Farhi & Tirole, 2016).

A different regulatory treatment of large financial institutions, ex ante by regulatory scrutiny or ex

post by bailouts, can have important consequences for tail risk and stock pricing (Gabaix, 2012;

Wachter, 2013). If small financial institutions do not receive similar protection, then the higher tail risk

of small financial institutions should translate into an additional size risk premium for small financial

institutions. This risk premium is on top of the size premium in equity returns. The latter size premium

is well documented in the literature (Banz, 1981; Chan & Chen, 1991; Van Dijk, 2011).

We are not the first to study size patterns in stock returns for suggestive evidence on government

guarantees. The findings in Gandhi and Lustig (2015) are consistent with government guarantees for

large US banks. These researchers adjust the returns of bank stocks for the three equity risk (3FF)

factors from Fama and French (1993) and two bond risk factors. Throughout the period 1970–2013, the

largest US banks had significantly lower r-a returns than smaller banks. In contrast, the r-a returns of

the portfolios of non-financial stocks tend to increase with size. Using r-a monthly returns by industry

in Germany, Nitschka (2016) finds that only the banking industry enjoyed TBTF subsidies. In a sample

of 31 countries (10 European countries), Gandhi, Lustig, and Plazzi (2017) find that r-a returns are

significantly lower for large financial stocks than for smaller financial and non-financial institutions.

Institutional settings determine the spread within a country. Bank-specific characteristics are absent in

their analysis.

We extend the literaturein several ways. We show that the size patternin the r-a returns of financial

institutions is present in the stock returns in 16 European countries. This size patternis absent in other

industries andsuggests the presence of government guaranteesfor large financial institutions inEurope.

The more volatilereturns of large financial institutionssupport a narrative of moral hazard. Weconstruct

a European financial risk factor, SMBfin, with a methodology closely related to Zeng, Yong,

Treepongkaruna, and Faff (2014), who find a significant return on their US bank size factor. The

significantreturn of our SMBfin is suggestiveevidence regarding governmentguarantees to the financial

industry in Europe. We add our SMBfin risk factor to the standard Fama and French (1993) model:

RiRF ¼α0þβRM RM RFðÞþβSMBSMB þβHMLHML þβSMBf inSMBfin þεið1Þ

with risk-free rate RF, market return RM, size factor SMB (Small Minus Big), and value factor HML

(High Book-to-Market Minus Low BtoM). Each risk factor is constructed using daily European stock

2

|

MUNS2

|

MUNS

156