The bank safety net inherently introduces moral hazard. For example, deposit insurance effectively serves as a put option whose value increases as risk-taking increases, potentially encouraging a bank to increase its value by taking on more risk. Also, due to banks’ systemic importance, they typically receive state support during periods of crisis and even outside crisis periods. As a result, bank shareholders generally do not bear the full extent of the costs associated with risk-taking despite receiving all of the benefits. Under these circumstances, bank shareholders have incentives to take more risks than is socially optimal, either by investing in riskier assets or by taking on more leverage.
Moral hazard can be reduced through a variety of factors. For example, a vast array of laws and regulatory agencies are already in place; or, managers could be concerned about their reputation in the labor market. Other mitigating factors may include the composition of liabilities, bank charter value, and managerial incentives. Despite the possibility of mitigation, moral hazard is considered the main driver of bank behavior in the banking literature and dominates the way banking is taught to students.
Given the tension between the incentives for moral hazard and the factors that limit them, it is an empirical question of whether banks ultimately display behavior driven by moral hazard considerations. Our study investigates how important moral hazard incentives are in actual bank behavior. We focus on the subset of banks where the moral hazard incentives should be most substantial. These are the banks that have relatively less to lose by being risky—banks that are already in financial distress. If moral hazard incentives are more important than the mitigating factors for these banks, we should see these banks increase leverage and risk.
We find little evidence for behavior driven by moral hazard, on average. Our results show that distressed banks reduce both their leverage and their risk, which is the opposite of what we would expect if moral hazard incentives dominate those banks. We conclude that the mitigating factors that limit the role of moral hazard incentives for banks are quite effective.
Empirical Setting and Data
Our study focuses on two periods: 1985–1994 and 2005–2014. Each period contains a banking crisis: the Savings and Loan (S&L) crisis and the Global Financial Crisis (GFC), respectively. Thus, we can examine how the importance of moral hazard incentives differs across regulatory regimes and during crises. In periods of economic crisis, reducing the size of a bank’s balance sheet is likely to be more costly because many other banks are trying to do the same. Consequently, banks may have a harder time deleveraging and reducing risk during crisis periods. The banking crises during our two sample periods affected different types of banks. The earlier crisis disproportionately affected S&L associations, whereas the second crisis impacted the entire banking system. Regulation also differed during the two periods. In 1985–1994, banks did not have formal capital requirements of the type now in use. (The Basel Accord was concluded in 1989 and implemented in the 1990s in the US.) Further, the US tightened regulations substantially in the wake of the S&L crisis. The FDIC Improvement Act (FDICIA), adopted in 1991, introduced prompt corrective action designed to resolve banks before they could engage in activities that would be detrimental to the deposit insurance fund.
Measuring Financial Distress
We define banks as distressed if they are in the top decile of leverage and the bottom decile of the bank Z-score. The Z-score is a measure of bank distress risk that proxies for a bank’s distance-to-default. We show that our approach selects banks that have a much higher probability of failure in subsequent quarters. A bank in the lowest decile of the distribution of the equity capital ratio and the lowest decile of the Z-score is about 17.8% more likely to fail within three years during the earlier period and 19.3% to fail in the later period (relative to an unconditional base rate of 2.0% and 1.7%, respectively).
Distressed Banks Deleverage, On Average
Moral hazard incentives should strongly influence distressed banks’ leverage decisions. All else being equal, a decrease in the leverage of a distressed bank would reduce the value of the deposit insurance for that bank and of the safety net in general. Further, because a decrease in leverage increases the value of the distressed firm’s debt at the expense of equity, some theories predict that highly leveraged firms will not decrease their debt leverage.
We begin our analysis by assessing whether distress predicts an increase in leverage, that is, a decrease in the equity-to-assets ratio. We find that it does not. On the contrary, the leverage of distressed banks decreases the year after they are classified as distressed. Looking a year out following a quarter when a bank is financially distressed, we document that distressed banks increase their equity capital ratio (deleverage) by about 0.80 percentage points outside a crisis – an economically significant increase in equity capital, equating to 54% and 30% of the standard deviation of annual equity capital changes in the respective periods. Distressed banks deleverage less during a crisis.
Next, we explore how banks attempt to deleverage. They have two main options. They can either deleverage by reducing assets and using the proceeds to pay back debt, or raise new equity or retain more earnings by cutting dividends. We find that distressed banks use both approaches to deleverage. Specifically, banks in financial distress shrink their assets (e.g., reduce the asset base, close branches, lay off employees), reduce their liabilities (e.g., shrink deposits, reduce deposit rates), and increase their equity capital (e.g., add equity capital, cut dividends).
Risk Taking by Distressed Banks
If moral hazard incentives dominate, then distressed banks should increase their risk. We explore this prediction in two ways. First, we examine how bank metrics that proxy for risk evolve after a bank is classified as distressed. Using these metrics, we find that a bank becomes less risky in the year after it is classified as distressed: its Z-score increases, the nonperforming loan (NPL) ratio decreases, earnings volatility decreases, and, in the second period, risk-weighted assets decline. Again, banks’ behavior is consistent across the two periods we examine. Second, we look at whether banks in distress increase lending to its executives. This action could potentially reflect an attempt by managers to use their power to increase their banks’ risk if such loans are extended on favorable terms. We find the opposite: loans to executives decline among banks in distress.
Deleveraging and Regulation
To better understand how banks respond to regulation, we measure whether the deleveraging of distressed banks changes over time. We examine whether distressed banks behave differently after the adoption of FDICIA, which occurred during the latter part of our first study period. We find that they do. Surprisingly, however, firms deleverage less in the second period than immediately after FDICIA was enacted. The Dodd-Frank Act was adopted during the second period. We find no evidence that the deleveraging behavior of banks after its adoption is more intense than before the Global Financial Crisis.
Private versus Public Banks
We next split the sample of distressed banks into a private and a public subsample and compare the deleveraging behavior of the two subsamples. Moral hazard incentives might be more substantial for public banks with diversified shareholders. Public banks differ from private banks in ways that may affect the role of moral hazard incentives. For example, public banks are, on average, much larger than private banks, which could increase their ability to benefit from public support. Also, public banks can raise funds in equity markets, potentially better positioning them to deleverage. Despite these differences between private and public banks, we find little disparity between the deleveraging and risk policies of public and private distressed banks.
Broad Interpretation of the Study’s Results
Overall, our evidence is inconsistent with the view that moral hazard incentives dominate banks’ leverage and risk decisions. We find that distressed banks, on average, deleverage and reduce their risk. During the Global Financial Crisis, distressed banks appear to have deleveraged less and reduced their risk less, primarily because of lower equity issuance. Although distressed banks deleveraged more aggressively after FDICIA was adopted in the first period, we find that this benefit of FDICIA did not carry over to the second period. We also find no evidence that the post-crisis changes in regulations, including the adoption of Dodd-Frank, affected the actions of distressed banks.
We do not claim that moral hazard incentives do not affect the leverage and asset risk decisions that banks make. Our findings only show that the forces that mitigate these incentives—including liabilities that are runnable, banks franchise value, managers’ bank-specific risk exposure, and laws, regulations, and monitoring by prudential authorities—are significant enough that banks do not act as if these incentives are dominant.
Our results apply on average to distressed banks, so it is certainly possible, even likely, that some distressed banks choose to take on more risk rather than deleveraging. However, our evidence shows that this view does not help to understand the behavior of the average distressed bank. Many factors can drive banks to deleverage and reduce their risk. Distressed banks that take actions to increase their leverage even further might find it challenging to attract and keep customers, and counterparties would be reluctant to deal with them. Irrespective of the regulatory regime, they would be under pressure from regulators. Managerial reputations would be endangered. As a result, commercial and market incentives, as well as those of managers, may make it optimal for the typical distressed bank to deleverage rather than keep pushing its leverage up.
Itzhak Ben-David and René M. Stulz are at The Ohio State University and NBER, and Ajay A. Palvia is at the Federal Deposit Insurance Corporation (FDIC). Stulz has a consulting practice where he is at times retained by financial institutions or their attorneys. Ben-David is a co-founder and a partner in an investment advisor that manages clients’ accounts. The views expressed in this paper are those of the authors and do not necessarily reflect those of The Federal Deposit Insurance Corporation, Office of the Comptroller of the Currency, or the US Treasury Department.
This post is adapted from their working paper, “How Important Is Moral Hazard For Distressed Banks?,” available on SSRN.