ACTIVITY STRATEGIES, AGENCY PROBLEMS, AND BANK RISK

• Author: Dung V. Tran,Pascal Louvet,Isabelle Girerd‐Potin,M. Kabir Hassan
• Published date: 01 August 2020
• DOI: http://doi.org/10.1111/jfir.12216
• Date: 01 August 2020

The Journal of Financial Research Vol. XLIII, No. 3 Pages 575–613 Fall 2020

DOI: 10.1111/jfir.12216

ACTIVITY STRATEGIES, AGENCY PROBLEMS, AND BANK RISK

Dung V. Tran

Saigon International School of Business, Banking University Ho Chi Minh City

M. Kabir Hassan

University of New Orleans

Isabelle Girerd‐Potin

Univ. Grenoble Alpes, Grenoble INP, CERAG

Pascal Louvet

Univ. Grenoble Alpes, Grenoble INP, CERAG

Abstract

We investigate whether diversification affects bank risk taking in the U.S. banking

industry, and whether this relation is partially explained by agency theory. Our

results show that U.S. banks with a relatively high share of noninterest income

become riskier when moving toward non‐interest‐income‐generating activities,

especially activities from investment banking, proprietary trading, and so on.

Diversification not only affects conditional average risk, but also the dispersion of

risk. Moreover, diversified banks that received assistance from the Troubled Asset

Relief Program (TARP) become riskier than diversified nonrecipients after TARP

capital injections. Our main findings are robust to a battery of robustness tests. The

results are partially explained under agency frameworks related to poor corporate

governance.

JEL Classification: G21, G28, G34, G38

I. Introduction

The banking industry is a particularly important sector in our economy, serving as a

channel through which disruptions in its smooth functioning could translate into

adverse fluctuations in the real economy. A sound banking system is a primary

objective of regulators and policy makers. In response to the recent financial crisis

along with the adjustments of capital adequacy, liquidity requirements, or mandatory

stress testing of systemically important financial institutions, financial regulators are

considering structural bank regulation measures, which aim to review and eventually

limit the scope of activities that banks can offer (Gambacorta and Van Rixtel 2013).

Historically, banks face strict restrictions on business lines and suffer high

competition pressures from nonbanks on both sides of the balance sheet, leading to

return and risk problems (Saunders and Cornett 2008, chap. 21). This heavy‐handed

regulation assumes the return to stability of the banking system, but at the cost of

575

© 2020 The Southern Finance Association and the Southwestern Finance Association

inefficiency and poor innovation (Dewatripont, Rochet, and Tirole 2010). Since the

1970s, these restrictions have been attenuated along with the wide deregulation of

financial markets. Banks are increasingly allowed to expand to activities previously

prohibited and now provide one‐stop shopping for a number of consumer services

(DeYoung and Rice 2004). They diversify their income stream into new activities such

as investment banking, venture capital, trading securities, and other activities that

generate noninterest income, and they traditionally earn profit from lending activities

in the form of interest income. Beyond deregulation, technological progress as well as

financial innovation spur banks to further diversify. This integration of new

nontraditional business lines into existing business activities consequently changes

the allocation of bank capital and risk–return preferences. The financial crisis of

2007–2009, with the failure of many banks and the ensuing economic recessions,

unveiled the dark side of functional diversification. Consequently, various initiatives

such as the Volcker Rule in the United States, Vickers in the United Kingdom,

Liikanen in the European Union,

1

and the recent call for the 21st‐century

Glass–Steagall Act propose narrow banking policies that aim to limit some of the

permissible activities of banks. This assumes that these activities contribute to higher

riskiness in banks. In line with this view, banks should concentrate on their main and

traditional activities.

Meanwhile, whether from a theoretical perspective or empirical studies, how

diversification affects bank risk taking has been a contentious debate among financial

economists over the past two decades (see Sanya and Wolfe 2011; DeYoung and

Torna 2013). However, these studies ignore the agency problem that can derive the

relation between diversification and bank risk taking. Therefore, the main objective of

this study is twofold: (1) how diversification affects bank risk taking and (2) how the

agency problem affects the relation between diversification and bank risk taking.

Theoretically, there are conflicting predictions about whether the potential

benefits from the diversification of activities outweigh the costs. On the one hand, it is

generally believed that a combination of activities reduces the total risk of diversified

banks (e.g., Brewer 1989). The conventional wisdom is that non‐interest‐generating

activities are considered noncorrelated, or at least weakly correlated, with interest‐

generating activities, resulting in a coinsurance effect, diversification gains, a more

stable revenue stream (DeYoung and Roland 2001), and a reduced bankruptcy risk

(Saunders and Cornett 2008). Additionally, assuming an absence of agency conflicts

between banks and borrowers, Diamond (1984) argues that diversified banks can

enhance their credibility in their loan‐making decisions and their borrowers’

monitoring by overcoming information asymmetry between depositors and borrowers.

Diversified banks can retrieve clients’information during the loan decision‐making

process and profitably reuse it for non‐interest‐income‐generating activities such as

securities underwriting or insurance (e.g., Yasuda 2005; Bharath et al. 2007). In turn,

information from non‐interest‐generating activities can facilitate loan‐making decisions

and make them more efficient, thus improving credit risk management. Diamond

1

Please see Casu et al. (2016) for the detailed differences between these initiatives.

576 The Journal of Financial Research

(1984) also suggests that diversified banks have a more stable credit supply under

aggregate shocks, which may in turn lead to lower volatility of cash flow from loan

portfolios. We call this channel the “diversification‐stability channel.”Evidence from

Litan (1985) and Brewer (1989), among others, lends support to this channel by using

U.S. bank data. DeYoung and Torna (2013) suggest that U.S. banks with a higher share

of noninterest income experience a lower probability of failure. Using international

data, Elsas, Hackethal, and Holzhäuser (2010) show that diversification positively

affects bank profitability and market valuations, whereas Sanya and Wolfe (2011)

document that diversification decreases insolvency risk and improves bank profitability

in emerging countries.

On the other hand, little research is concerned that certain nonbank activities

may be riskier than traditional banking activities when viewed on a stand‐alone basis

(Saunders and Walter 1994). Securities underwriting is an example. In a firm

commitment security offering, the underwriter’s profit (i.e., the spread between the

underwriter’s buy price and the public offer price) is capped, whereas the downside

risk could be much higher. The extent of the diversification gains depends on the

comovement of income streams generated from combined activities (Demirgüç‐Kunt

and Huizinga 2010). If the sought‐after activity is inherently riskier than the banking

business and these activities are highly correlated,

2

the cost of diversification could

outweigh the benefit, leading to a higher risk for banks (Boyd, Graham, and

Hewitt 1993). Increased diversification of activities does not translate into risk

reduction if there is a lack of expertise in the newly adopted business (Jiménez and

Saurina 2004). In their theoretical model, Shleifer and Vishny (2010) show that with

a shift toward nontraditional activities, banks are more likely to enlarge their balance

sheets and transmit security market fluctuations into the real economy; the volatility

of sentiment turns into the volatility of real activity through the banking sector.

They suggest that this type of profit‐maximizing behavior may induce the instability

of the banking sector. Diversification raises the concern of intensified agency

problems because functional diversification can increase a bank’ssizeaswellasits

opaqueness, leading to discretionary decisions to undertake value‐decreasing

investments (Berger and Ofek 1995). We call this channel the “diversification‐

fragility channel.”Demsetz and Strahan (1997) examine the stock returns of banks

during 1980–1993 and indicate that better diversification does not translate into risk

reduction. Similarly, DeYoung and Roland (2001) document a greater volatility of

bank earnings with a move toward fee‐based activities during 1988–1995. Stiroh

(2004) and Stiroh and Rumble (2006) find diversification benefits from moving

toward non‐interest‐income‐generating activities, but these gains are offset by

increased exposure to noninterest activities. Avraham, Selvaggi, and Vickery (2012)

describe the typical structure of a large bank holding company (BHC) and

complexity of non‐interest‐generating activities and reasons for diversification. De

Jonghe (2010) finds that diversification increases the tail beta of European banks,

2

Bhattacharyya and Purnanandam (2011) document a substantial increase in systematic risk and decreased

idiosyncratic risk in banking during 2000–2006.

577Activity Strategies, Agency Problems