An often-over-looked aspect of regulation is how agencies are organized. Regulatory agencies for many industries, including banking, pharmaceuticals, mining, and agriculture, rely on a mix of centralized decision-making and delegated monitoring. For instance, in the case of banking, federal agencies design regulations in Washington, D.C. but monitor banks at the local level by utilizing semi-autonomous field offices.

A major advantage of this dispersed presence is that it allows local examiners and supervisors to interact with regulated firms more frequently and to collect “soft information” about firms’ performance that is often imperfectly captured through accounting-based reporting measures. The approach may, however, create conflicts of interest between centralized decision makers and local supervisors. For instance, the local supervisors may become subject to regulatory capture, treating regulated firms more leniently than is warranted by the objectives of the central regulatory agency.

In a recent paper “Hub-and-Spoke Regulation and Bank Leverage”, we empirically test this trade-off by examining how geographic proximity to regulators affects banks’ risk taking. We focus on one particular federal banking regulatory agency: the Office of the Comptroller of the Currency (OCC). The OCC is the primary supervisor of nationally chartered commercial banks. While headquartered in Washington, D.C., it operates a system of over 60 field offices spread out in various regions of the country. These offices are primarily responsible for carrying out the day-to-day responsibilities of bank supervision.

Our analysis uses a unique dataset of OCC field office locations between 1984 and 2014 that we hand-collect from archived internal telephone directories. These telephone directories, published annually, list physical addresses of field office locations. Using this data, we are able to track openings and closings of OCC field offices over many years. Our data suggest that the OCC opens new offices in areas that experience an increase in banking activity. However, when banking activity near the large (parent) offices decreases, the OCC consolidates the smaller neighboring offices by merging them with the parent offices. We examine the effect of these office consolidations on risk taking by banks supervised by the small offices.

We find that field office closures lead to an increase in leverage for banks previously supervised by these offices. Our estimates suggest that the tier 1 capital to total assets ratio for affected banks declines by an additional 3.7 percent relative to the unaffected banks located in the same metropolitan statistical area. This finding is robust to controlling for local economic conditions and time varying differences across field offices.

Leverage may increase passively if banks realize more loan losses or make provisions for more loan losses. Instead, we find that banks increase leverage by actively distributing cash back to shareholders, rather than through realizing losses due to poorer performance. Specifically, our estimates suggest that affected banks increase dividends by 10 percent more than unaffected banks.

If banks increase leverage while remaining stable, it may be beneficial to increase payouts or to take on additional risk. However, we find delayed consequences of risk taking in terms of higher failure rates around two to three years following closure. These results suggest that the increase in leverage is perhaps too risky from the regulators’ perspective, whose main objective is bank safety and soundness.

Finally, we document that the proximity of a bank to its supervisors can affect its risk taking. When banks are closer, supervisors can collect information not easily presented in financial statements. It becomes difficult to collect such soft information when the banks are farther away. Further, greater distances may impose higher costs on regulation. Consistent with this explanation, we find that our results are stronger when the closing of a field office increases the driving time between a bank and its regulator.

Our paper provides the first empirical examination of how the organizational structure of regulatory agencies affects firms’ risk taking. Our findings suggest that the benefits of more local supervision outweigh any associated conflicts of interest. These findings could be helpful to policymakers as they continue to reform banking supervision after the Great Recession.

This post comes to us from Yadav Gopalan and Ankit Kalda, who are PhD candidates at Washington University in St. Louis’ John M. Olin Business School, and Asaf Manela, who is a professor at the school. It is based on their recent paper, “Hub-and-Spoke Regulation and Bank Leverage,” available here.