The Impact of Supervision on Bank Performance

• Author: ANNA KOVNER,BEVERLY HIRTLE,MATTHEW PLOSSER
• DOI: http://doi.org/10.1111/jofi.12964
• Date: 01 October 2020
• Published date: 01 October 2020

THE JOURNAL OF FINANCE •VOL. LXXV, NO. 5 •OCTOBER 2020

The Impact of Supervision on Bank Performance

BEVERLY HIRTLE, ANNA KOVNER, and MATTHEW PLOSSER∗

ABSTRACT

We explore the impact of supervision on the riskiness, profitability, and growth of

U.S.banks. Using data on supervisors’ time use, we demonstrate that the top-ranked

banks by size within a supervisory district receive more attention from supervisors,

even after controlling for size, complexity, risk, and other characteristics. Using a

matched sample approach, we find that these top-ranked banks that receive more

supervisory attention hold less risky loan portfolios, are less volatile, and are less

sensitive to industry downturns, but do not have lower growth or profitability. Our

results underscore the distinct role of supervision in mitigating banking sector risk.

SUPERVISION AND REGULATION ARE critical tools for the promotion of sta-

bility and soundness in the financial sector. Despite supervision’s key role, it

is rarely examined separately from regulation and relatively little is known

about the distinct impact of supervisory efforts. In this paper, we exploit new

supervisory data and develop a novel identification strategy to estimate the

impact of greater supervision on bank risk-taking and performance. We find

that more supervision adds value over and above the effects of regulation. In

particular, banks that receive more supervisory attention have less risky loan

portfolios, are less volatile, and are less sensitive to industry downturns, but

do not exhibit lower growth or profitability.

While prior work examines specific supervisory actions such as examina-

tions or enforcement actions, little is known about the effectiveness of super-

visors’ more general efforts to promote sound risk management. For example,

supervisors meet frequently with bank management, discussing both specific

∗Beverly Hirtle, Anna Kovner, and Matthew Plosser are with the Federal Reserve Bank of New

York. The authors thank Maya Bidanda, Angela Deng, Brandon Zborowski, and Samantha Zeller

for excellent research assistance. The authors thank Mark Carey, Stefan Lewellen, Mark Levo-

nian, Antoinette Schoar,Philip Strahan, Vish Viswanathan, two anonymous referees, and seminar

participants at the NY Fed, AFA Annual Meetings,Bank of France, NBER Summer Institute, and

FDIC/JFSR Bank Research Conference for helpful comments. The views in this paper are those of

the authors and do not necessarily reflect the views of the Federal Reserve Bank of New Yorkor of

the Federal Reserve System. The paper was subject to review when it was issued as a Staff Report

by the Federal Reserve Bank of New York.A more detailed disclosure statement is available in the

supplementary information section of the article.

Correspondence: Matthew Plosser, FederalReserve Bank o f New York, (212) 720-7486; e-mail:

matthew.plosser@ny.frb.org.

DOI: 10.1111/jofi.12964

© 2020 the American Finance Association

2765

2766 The Journal of Finance®

issues related to activities at the bank and more general perspectives on the

industry’s environment and outlook. These conversations, along with analy-

sis of internal firm reports and external data, allow supervisors to understand

a bank’s risks and procedures and to put these in context relative to the in-

dustry. By focusing on a broad concept of supervisory attention, our analysis

captures the breadth of supervisory efforts without restriction to a single su-

pervisory program.

Reflecting the potential externalities of bank failures, supervisors seek to

reduce failure risk relative to what banks themselves might otherwise choose.

Our analysis sheds light on whether supervisors achieve this objective and at

what cost to financial intermediation. Because actual bank failures are infre-

quent, especially among the largest banks that receive the most supervisory

attention, we rely on a variety of risk measures, including measures of lending

riskiness (e.g., loan losses and reserve practices), measures of earnings risk

(e.g., earnings volatility), as well as measures of market riskiness (e.g., return

volatility).

The key empirical hurdle to identifying the impact of supervision is that

larger and riskier firms receive more supervisory attention. To address this

challenge, we exploit the structure of supervisory responsibilities within the

Federal Reserve System, under which bank holding companies (BHCs or

“banks”) are geographically assigned to 1 of 12 Federal Reserve districts. We

hypothesize that within each district, the largest institutions receive more su-

pervisory attention, ceteris paribus, than institutions that are not among the

largest. We validate this hypothesis using proprietary Federal Reserve data

by showing that examiners spend more time at the largest firms in a district,

even when controlling for firm characteristics such as size, market share, com-

plexity, and supervisory rating.

We match top-ranked BHCs in each Federal Reserve district by size, deposit

market share, organizational complexity, types of banking subsidiaries, diver-

sity of lending activities, and other characteristics to non-top-ranked banks in

other districts. Doing so allows us to construct a sample of banks that have

varying ranks in their Federal Reserve districts but that are otherwise simi-

lar. The matching procedure is designed to control for differences across banks

that might be correlated with rank and performance, but excludes outcome

variables that might be directly influenced by supervision. Thus, by comparing

outcomes of banks that are among the largest in a district to otherwise similar

banks that are not among the largest in other districts, we are able to inter-

pret differences in outcomes as reflecting the impact of greater supervisory

attention.

Our findings suggest that enhanced supervisory attention is associated with

lower risk. Accounting earnings and market returns are less volatile for the

largest banks in a district than for otherwise similar BHCs. Moreover,top-size-

ranked banks on average have 15% to 25% lower nonperforming loan (NPL)

ratios than peers not among the largest in their district, and they appear to

engage in more conservative loan loss reserve practices. The reduction in risk

is greater during downturns, precisely when the externalities associated with

The Impact of Supervision on Bank Performance 2767

failure are costlier. With respect to volatility, our findings suggest that a 20%

increase in supervisory hours is associated with a 9% to 10% decrease in risk

as measured by the volatility of earnings, a reduction in distance to default

equivalent to 150 to 200 bps of additional capital.

Importantly, while top-ranked BHCs appear to be less risky, they do not ex-

hibit lower profitability, nor do they exhibit significantly slower loan or asset

growth. Specifically,we find no meaningful differences in market or accounting

measures of profitability, including return on assets (ROAs) and excess stock

price returns, between top-ranked BHCs and their matches. Further, the mar-

ket Sharpe ratio of top-ranked BHCs is similar to that of BHCs not among

the top-size-ranked firms. These findings are consistent with additional su-

pervisory attention resulting in lower risk exposure but having a positive-to-

neutral impact on the risk-adjusted performance. Our findings therefore sug-

gest that supervision meaningfully improves social welfare by decreasing the

risk of failure among large banks without significantly reducing banks’ inter-

mediation activities.

Our identifying assumption is that being among the largest firms in a Fed-

eral Reserve district is not associated with other unobserved factors that also

affect bank performance. We take a number of steps to account for potentially

confounding factors. For example, differences in risk across districts might

drive our results. To account for unobserved district-level differences, we use a

larger matched sample and include district-quarter fixed effects. Differences in

competitiveness and franchise value may also play a role. To address this, we

consider additional matching criteria, including recent firm performance and

proxies for franchise value such as market-to-book. Our conclusions are robust

to using these alternative specifications.

Overall, our results suggest that increased supervisory attention results in

lower risk without significantly reduced performance. However, our empirical

approach does not shed light on the specific mechanisms by which supervi-

sion achieves these outcomes. One possibility is that supervision helps mitigate

principal-agent problems within the firm. In particular, enhanced supervision

may result in more weight being given to concerns of risk managers, resulting

in more disciplined risk-taking. Supervisors may thus improve a bank’s risk

culture, by increasing attention to risk as a balance against a focus on short-

term profitability. Supervisory requests for risk and activity information may

also lead banks to invest in data and technology systems that allow them to

manage their business more efficiently over the long run. Finally, because su-

pervisors oversee many banks, they may transmit knowledge of best practices

in the industry when they provide feedback to banks about their risk manage-

ment practices.

Previous work on the supervision and regulation of banks focuses largely

on the impact of regulation, though the distinction between supervision and

regulation is not always clearly recognized or articulated.1Fewer papers focus

1For instance, a substantial body of work examines the impact of regulatory capital require-

ments (for a recent example, see Bridges et al. (2014)) and of legislative changes that enabled