The Role Of Bank Capital In Bank Holding Companies’ Decisions

Adolfo Barajas

International Monetary Fund (IMF) – Western Hemisphere Department

Thomas F. Cosimano

University of Notre Dame – Department of Finance

Dalia Hakura

International Monetary Fund (IMF)

Sebastian Roelands

Bowling Green State University

March 2015

IMF Working Paper No. 15/57


This paper examines the role of bank capital in decision-making by bank holding companies (BHCs) in the United States. Following Chami and Cosimano’s (2001) call option approach to bank capital, BHCs optimally choose the amount of capital to insure the bank against becoming capital constrained in the future. We provide empirical support for this model, and find that a higher optimal level of capital leads to higher loan rates. Furthermore, higher loan rates result in lower amounts of lending. Thus, an increase in capital requirements is likely to lead to higher loan rates and a significant reduction in lending.

The Role Of Bank Capital In Bank Holding Companies’ Decisions – Introduction

Given the prevalent regulatory focus on setting minimum capital ratios, especially in the aftermath of the subprime crisis, a key question to ask is: What, if any, is the role of capital in banks’ decision-making? If policy is to have an influence on bank behavior, presumably to align it more closely to a social objective of reduced risk-taking, then it is crucial to understand the channels through which such a change in behavior might take place.

A considerable portion of the macro-theoretical literature, particularly pre-crisis, does not contemplate a meaningful role for bank capital. In most of these models, banks operate in a competitive environment in which the cost of raising equity is minimal. If financial frictions are present, they are represented by means of the financial accelerator introduced by Bernanke, Gertler and Gilchrist (1996, 1999), where a time-varying external finance premium follows from endogenous changes in the agency costs of lending. Furthermore, banking is characterized as being perfectly competitive (see also Kumhof et al., 2010), and therefore banks’ decisions are never constrained by bank equity, since in such an environment additional equity can always be raised costlessly when needed.

In financial economics, a set of assumptions{such as the absence of frictions from taxation{lead to the well-known MM theorem that states that the capital structure of banks is irrelevant.1 Thus, the MM theorem also implies that the financing of banks operations would not be constrained by bank equity. Moreover, as argued by Admati et al. (2013), an increase in bank capital, by lowering the bank’s probability of default, might even lower the marginal cost of capital relative to other financing sources such as deposits or other debt. Furthermore, Admati and Hellwig (2014) argue that capital requirements should be raised, considering this low cost of capital.

On the other hand, the corporate finance literature has developed multiple environments in which the Modigliani-Miller Theorem does not hold. A recent study by Aiyar, Calomiris and Wieladek (2012) examines the effectiveness of capital regulation, which relies on bank equity being costly. It provides both a summary of conditions under which this is the case, as well as empirical evidence drawn from the United Kingdom experience since the adoption of Basel I.

The conditions under which equity is relatively costly include: insufficient information about the bank’s loan portfolio, favorable tax treatment of dividends, `Too Big To Fail’ (TBTF)2, and deposit subsidies through deposit insurance. By distinguishing between `regulated’ and `unregulated’ banks in the U.K., the study argues that the empirical results show that exogenous increases in capital requirements are associated with declines in lending among the regulated banks only, which implies that equity is costly.

A key empirical challenge in identifying a link between relative capital scarcity and loan supply is to isolate supply shocks from demand shocks. A common criticism of empirical work based on economy-wide data is the lack of convincing evidence that a change in supply has indeed been completely isolated from the change in demand for loans, and whether or not other financial institutions can easily replace bank financing.3 The empirical solution then has often been to focus on events in which a `natural’ experiment leads to a decline in bank equity.

Two papers which do include costly equity with an imperfectly competitive banking sector in a DSGE setting are Gerali et al. (2010) and Roger and Vlcek (2011). Both find that negative shocks to bank capital have significant negative effects on investment and other real variables. In addition, Meh and Moran (2010) and Dib (2010) develop DSGE models in which the cost of raising equity by banks is determined endogenously. As a result, highly leveraged banks experience a relatively high cost of equity, and this financial friction has the potential to amplify the effects of economic shocks on the real economy.

Bank Capital

Bank Capital

See full PDF below.