The determinants of reputational risk in the banking sector

Abstract

What determines reputational loss following operational losses in banking? The purpose of this paper is to empirically address this question. We estimate the reputational risk for a large sample of banks in Europe and the US between 2003 and 2008. We have two main results. First, we provide evidence that there is the probability that reputational damage increases as profits and size increase. Second, we show that a higher level of capital invested and intangible assets reduce the probability of reputational damage.

Introduction

Reputational risk is the “risk arising from negative perception on the part of customers, counterparties, shareholders, investors, debt-holders, market analysts, other relevant parties or regulators that can adversely affect a bank’s ability to maintain existing, or establish new, business relationships and continued access to sources of funding” (Basel Committee on Banking Supervision, 2009, p. 19). Reputation assumes special importance in banking because asymmetric information, the qualitative-asset-transformation made by banks, and the supply of payment and risk management services create a systemic risk (Allen and Santomero, 1997, Allen and Santomero, 2001).

Interest in reputational risk in the financial sector has grown over the past two decades after the occurrence of some prominent examples of operating losses due to internal frauds (Dyck et al., 2010) (e.g. fraudulent trading in the Allied Irish Bank, Barings and Daiwa Bank Ltd generating operational losses of $691 million, $1 billion, and $1.4 billion, respectively), external fraud (e.g. the $611 million operational losses at the Republic New York Corp. for fraud committed by custodial clients), damage to physical assets (e.g. Bank of New York suffered $140 million operational losses to damage to facilities related to September 11, 2001), business disruption and system failures (e.g. Salomon Brothers suffered $303 million operational losses because of a change in computer technology that resulted in unreconciled balances). The credit turmoil from 2007 onwards has definitively increased the attention of academics, regulators and practitioners on bank reputation.

Despite its importance, few studies (De Fontnouvelle and Perry, 2005, Cummins et al., 2006, Gillet et al., 2010, Lin and Paravisini, 2011, Fiordelisi et al., accepted for publication) provide empirical evidence about reputational risk in financial services. Most of these studies focus on estimating the extent of reputational losses as market reaction to the operational loss announcement¹ running an event study: overall, operational loss announcements are usually found to generate statistically significant reputational damage. As far as we are aware, there is only one paper (Gillet et al., 2010) going beyond the measurement of reputational risk assessing the role of various determinants. Generally, the understanding that drives reputational losses in the banking industry is unknown and the need for empirical studies is noticeable.

What determines reputational damage in banking after operational losses? The purpose of this paper is to empirically address these questions. We show that the probability of reputational damage is influenced by bank risks, profits, size, capital invested, the level of intangibles and the business area experiencing the operational loss. In summary, we have two main results: first, the probability of reputational damage increases as there is an increase in bank profits and size. Second, we show that a higher level of capital invested and intangibles reduce the probability of reputational damage.

Our paper is one of the few estimating reputational risk using a large sample of operational losses. Specifically, we focus on listed banks in Europe and the US between 2003 and 2008. First, we run an event study to estimate the reputational damage following Gillet et al. (2010). Second, we estimate a multivariate model to assess the determinants of operational losses.

The remainder of the article is organized as follows: Section 2 presents the literature review and Section 3 outlines the research hypotheses. In Section 4, we provide an overview on the empirical research design, including a description of data, econometric approach and robustness tests. Our results are discussed in Section 5. Lastly, Section 6 offers concluding remarks.