Subjective Assessment of Managerial Performance and Decisionmaking in Banking

Abstract

We examine subjective supervisory assessments of managerial performance in the banking industry. Results of empirical tests show that better assessments are (i) positively associated with decisions made by examiners to upgrade relatively objective bank performance ratings; (ii) negatively associated with decisions made by examiners to downgrade relatively objective bank performance ratings; and (iii) positively associated with decisions made by bank holding company managers to distribute resources among subsidiary banks. These results are consistent with the finding that soft information generated in the supervisory process is validated by subsequent decisionmaking both internally (by bankers) and externally (by examiners).

Introduction

Subjective assessments of managerial performance have been widely hypothesized to play an important role in contracting within firms and other organizations. Empirical support for this hypothesis, however, is limited by overlaps with objective metrics on which performance also depends. 

We approach this issue from the perspective of information identified by bank examiners during their evaluations of managers of commercial banks. Examiners are well placed to observe the subtleties of managerial behavior (Baker, Gibbons, and Murphy, 1994) and are therefore able to assess contributions to firm value that are not objectively measurable. Their assessments rely on soft information that requires effort to obtain, judgment to assess, and knowledge that becomes less useful when separated from the environment in which it was collected (Eisenbach, Luca, and Townsend, 2016, and Liberti and Petersen, 2019). 

Performance ratings are prominent in this process. These ratings are assigned by examiners following on-site inspections to measure capital (C), asset quality (A), earnings (E), management (M), liquidity (L), and market sensitivity (S). They range numerically from 1, for the best banks, to 5, for the worst. They collectively are rolled up into a composite (CAMELS) rating. All are non-public.

The management rating, the M component, encompasses a broad range of factors that are subjective in nature. They include community service, the nature and degree of working relationships, and the level and quality of oversight (Federal Deposit Insurance Corporation [FDIC], 2022; Board of Governors of the Federal Reserve System [BOG], 2021). They have been said to include those that are so intuitive as to reflect "a particular banker's temperament" (Effinger, 2017).

Critics contend that the management rating is excessively subjective. They advocate an overhaul that emphasizes "clear, cogent and objective measures of financial condition over vague, arbitrary and subjective ones" (Baer and Newell, 2017). Supervisors appear sympathetic: The BOG, the Federal Financial Institutions Examination Council, and the FDIC are exploring how to ensure that the different agencies and different examiners within each agency are applying ratings consistently and uniformly (McWilliams, 2019a,b).

We contribute to this debate by determining whether soft information about a bank—embodied in its management rating—can be linked to decisionmaking. We use a proxy variable for soft information, derived from CAMELS components, that is statistically independent of information that is more quantitative in nature. Separate analyses consider decisions that are made externally, by examiners in establishing subsequent CAMELS component ratings, and internally, by bank holding company managers in allocating resources among affiliated banks. 

With respect to decisions of examiners, results of tests using a sample of 27,484 commercial banks over the 2011 to 2017 period show that subjective assessments indicating better managerial performance are positively related to improvement and negatively related to deterioration in subsequent CAMELS component ratings. Base-case scenarios are confirmed in robustness tests that account for potential endogeneity in how examiners are assigned to specific banks. 

With respect to decisions of bank managers, results of tests based on 21,259 observations on subsidiaries in multi-bank holding companies over the 1999 to 2017 period show that subjective assessments indicating better performance are positively related to the likelihood that a bank subsidiary receives more resources from its parent holding company. Base-case scenarios are confirmed in robustness tests that account for potential endogeneity in how examiners, or managers, are assigned to specific banks. 

Results across the two analyses are consistent with a production of soft information whose usefulness is validated in decisions about banks that are made both internally (by bankers) and externally (by examiners). The former has theoretical roots in prior empirical analyses of qualitative factors in corporate decisionmaking (Jian and Lee, 2011; Duchin and Sosyura, 2013; McNeil and Smythe, 2009; Glaser, Lopez-De-Silanes, and Zautner, 2013; and Gaspar and Massa, 2011), while the latter can be understood in the context of earlier studies showing that CAMELS ratings predict regulatory outcomes (Cole and Gunther, 1998; Hirtle and Lopez, 1999; Berger and Davies, 1998; DeYoung et al., 2001; and Gopalan, 2018). Both are part of a larger theoretical literature on how information is used in contracting within organizations (Rajan and Reichelstein, 2006; Baker, Gibbons, and Murphy 1994; and MacLeod, 2003).