Making Banks Transparent

In the paper, Making Banks Transparent, forthcoming in the Vanderbilt Law Review, I propose a mandatory disclosure regime designed to make the credit risk residing on the balance sheets of financial institutions more transparent to the marketplace than is presently the case. As the Financial Crisis and the more recent European sovereign debt crisis each illustrated, U.S. financial institutions represent uniquely opaque organizations for investors in capital markets. Although bank regulatory policy has long sought to promote market discipline of banks through enhanced public disclosure, bank regulatory disclosures are notoriously lacking in granular, position-level information concerning their credit investments due largely to conflicting concerns about protecting the confidentiality of a bank’s proprietary investment strategies and customer information. When particular market sectors experience distress, investors are thus forced to speculate as to which institutions might be exposed, potentially causing significant disruptions in credit markets and contributing to systemic risk. Together with the failure of bank regulators to monitor bank risk-taking prior to the Financial Crisis, these concerns have prompted renewed calls for making financial institutions more transparent.

Motivated by the demand for greater bank transparency, the Article proposes a mandatory disclosure regime designed to enhance accurate pricing of a bank’s exposure to credit risk while respecting banks’ proprietary interests. In particular, by turning to the insights of credit risk modeling, I argue that redesigning bank disclosures to facilitate credit modeling by market participants has the potential to meaningfully increase market discipline while minimizing the disclosure of sensitive bank data. Notwithstanding the complex ways a bank can be exposed to credit risk, the practical need for institutions to manage it has nevertheless facilitated a rich literature on credit risk modeling that provides an understanding of the type of disclosures that can enable more effective market discipline. By analyzing credit risk in a bank’s investment portfolio in terms of a limited, standard set of quantifiable metrics, credit risk models provide an architecture for analyzing a bank’s overall exposure to credit risk that is both well understood within the financial sector and parsimonious in the information required to be processed. For the same reasons, disclosure of these standard metrics provides a potentially simple but powerful method for a financial institution to communicate useful information concerning its exposure to credit risk without the need to disclose proprietary position information. Yet, to date, standard bank disclosures generally omit mention of these parameter estimates, thus missing an important opportunity to make banks more transparent by using the very analytical tools banks themselves developed to make credit risk less opaque.

To explore the ways in which credit risk models could better inform bank disclosure policy, I turn to a pair of case studies examining two of the most severe banking crisis in U.S. history: the collapse of the Continental Illinois National Bank and Trust Company in 1984 and the near collapse of Citigroup in 2008. In each instance, the bank’s distress prompted either a subsequent government investigation or private litigation that provided sufficient details concerning the composition of each bank’s credit portfolio to estimate the core set of parameters needed for a simple Monte Carlo-based credit risk model of the bank’s credit portfolio. As such, each crisis provides a unique opportunity to explore how a “model sensitive” disclosure regime might better enable market participants to detect a bank’s insolvency risk and assess its overall capital adequacy.

Although the problems afflicting each bank’s credit portfolio differed markedly, the Article illustrates how standard approaches to credit portfolio modeling might have used disclosure of these parameter estimates to detect each bank’s insolvency risk well in advance of its distress. At the same time, in neither case would the public disclosures have required the firms to reveal position-level data, suggesting the potential for greater market discipline without the need to reveal proprietary trading information. Of course, the fact that such parameter estimates are publicly available for so few failing firms—and not at all available for non-failing firms—makes it impossible to assess the error rate of such an approach. Nor is it possible to know with confidence the precise credit models market participants would use were such disclosures routinely provided. Yet by providing a detailed thought experiment of how market participants might use such disclosures with even basic, textbook credit models, the two case studies suggest that the same credit modeling techniques long valued by bank managers to assess their portfolio’s credit risk might also be used by investors in the capital markets to understand better a bank’s overall capital position and insolvency risk. For similar reasons, designing pilot disclosure programs that facilitate credit modeling among market participants can provide more general data concerning the conditions under which such modeling would be conducted and the error rates associated with it. Such programs could prove useful as bank regulators around the world consider how to revise the disclosure obligations for systemically important financial institutions.

The full paper is available for download here.