Triggered by the recent financial crisis, the regulation of banks has gained new traction among academics, regulators, and politicians. One of the key challenges in effective regulation is time inconsistency of regulation. While a regulator would like to commit not to bail out banks in order to set the right ex-ante incentives, this threat is generally not credible since the government does not follow through in the event of a crisis. Banks therefore have an incentive to expose themselves to risk that is partially insured by the government.
To mitigate this problem, regulators attempt to reduce the likelihood of banking crises by regulating both banks’ asset side and liability side. While there has been a recent push to focus on the liability side by mandating higher equity capital requirements, the very nature of a deposit-taking institution implies that leverage is an integral part of the business model of banks, unlike for other firms. In this paper, we therefore focus on the regulation of banks’ asset holdings. The starting point of our paper is the natural assumption that a regulator cannot directly observe the riskiness of assets, but needs to rely on an external (private) assessment of risk. Since the introduction of the Basel I framework, credit ratings have played an important role in bank regulation as “objective” measures of credit risk. This role has been confirmed in the Basel III (2011) guidelines, which still rely on credit ratings as measures of creditworthiness.
In our paper, Bank Regulation with Private-Party Risk Assessments, which was recently made publicly available on SSRN, we aim to solve for the optimum reliance on credit ratings when regulating banks. In our model, banks are socially valuable, since they possess a comparative advantage at enforcing debt claims relative to retail investors. While this advantage allows banks to create value on the one hand, it also implies that the government has an incentive to bail out banks ex-post, in the event of a banking crisis.
We find that the optimum regulatory reliance on credit ratings is reduced when banks have a smaller equity cushion, since rating agencies have a stronger incentive to inflate ratings when banks are more likely to default. The mechanism for this result is intuitive: asset substitution towards risky assets is particularly valuable for profit-maximizing banks when leverage is high. The rating agency can facilitate asset substitution through rating inflation and hence undermine regulation of banks’ risk-taking behavior. Only in the extreme case where bank equity is so plentiful that no existing investment strategy can lead to bank default, rating-contingent regulation becomes superfluous, since banks bear the full downside losses of any investment, implying that their private NPV of any investment equals the social NPV. Yet, apart from this extreme case, equity requirements and rating-contingent regulation complement each other, that is, larger equity cushions also allow regulators to rely more heavily on the risk assessments of credit ratings without triggering undesirable rating inflation.
Our paper characterizes an important dimension to the regulator’s decision problem that appears to have been overlooked in the Dodd Frank Act and the Basel III regulatory framework. When a regulator decides on the optimal reliance on any given risk measure (e.g. credit rating) in a given security class (e.g. government bonds, corporate bonds, structured products, etc.) not only the risk measure’s precision and security class riskiness (e.g. maximum loss) matter. An additional key economic variable is banks’ comparative advantage (measured by social value added) in being the investor in a given security class. A typical comparative advantage may for example stem from banks’ superior ability to monitor underlying projects, or from enforcing contractual debt payments, relative to next-best investors outside the banking system. If this comparative advantage is small, a regulator optimally severely restricts banks’ holdings in the given asset class. In fact, if banks do not have a comparative advantage from a social perspective in a given security class, the trivial optimum regulation is to disallow banks to hold any of these securities, independent of their risk profile and/or the quality of available risk measures. Note that it might be socially desirable to force a bank to hold securitized assets even if it has no superior ability to collect cash flows, provided that holding onto these assets creates an incentive to exert screening effort before origination. Our model thus suggests differential rules depending on the whether the bank originated the assets or purchased them on the secondary market.
Within our framework, the government sets optimum regulation which takes into account the incentives of banks, unregulated investors, issuers, and a profit-maximizing rating agency. On the one hand, restricting banks to hold extremely safe assets (cash) may cause under-investment in risky, but positive-NPV projects. On the other hand, lax restrictions allow banks to overexpose themselves to risky securities, facilitated by rating agencies’ practice of rating inflation. If regulators have to set the regulatory regime before the accuracy of ratings is known, the regulator has to trade off these costs depending on the relative likelihood of various states of the world. In particular, the regulator may optimally choose to incur rating inflation in some states of the world in order to increase valuable investments in risky securities in other states. As in Opp, Opp, and Harris (2012), regulation affects the equilibrium precision of ratings and may induce rating inflation. Further, the general equilibrium setting of our model allows us to show that reliance on market prices instead of credit ratings has a generic weakness, since equilibrium prices in markets in which banks are marginal reflect government bailouts, and thus tend to reveal little information about actual risk exposures.
The economy in our model features good and bad types of securities. Bad types are not only negative NPV projects but also exposed to aggregate risk, that is, an aggregate shock induces fluctuations in default rates of bad types. The government has an incentive to allow banks to hold risky securities since banks are better at recovering contractual debt payments from issuers (relative to retail investors). On the other hand, banks have a private incentive to over-expose themselves to bad securities since the government cannot commit not to bail out banks in case of default. Exposure to bad securities and thereby to aggregate risk is necessary for the bank to profit from the government’s put. The government cannot directly observe the types of securities held by banks and therefore may wish to use ratings as measures of creditworthiness when regulating banks’ asset mix. The accuracy of a credit rating, however, will depend on how complex the rated security is. Uncertainty about the complexity of securities, therefore, implies that the optimal ex ante sensitivity of capital requirements to ratings may result in rating inflation in the state of the world when the aggregate amount of complex securities is high.
The full paper is available for download here.