CAMELS (capital adequacy, assets, management capacity, earnings, liquidity, and sensitivity) ratings can be used to assess a bank’s condition. The authors explore whether there has been a material change over time in the standards that bank examiners use when assigning CAMELS ratings to commercial banks.
What’s Inside?
What looks to be a rather slow economic recovery from the Great Recession in the United States has bankers and clients questioning whether unduly stringent supervision of commercial banks has limited credit growth and hampered a return to economic well-being. A robust examination and testing of models that use time-varying parameters and macroeconomic variables suggests the contrary—namely, that supervisory standards used to assign capital adequacy, assets, management capacity, earnings, liquidity, and sensitivity (CAMELS) ratings in the aftermath of the crisis have been consistent with historical precedent. Heightened supervisory rigor has noticeably curbed lending activity in subsequent quarters.
How Is This Research Useful to Practitioners?
Commercial banks play a critical role in the financial system through their control of credit provision. Lenders and borrowers alike have asserted that what they deem to be excessive regulatory scrutiny of their sector has contributed to its lackluster performance, with ripple effects throughout the US economy, particularly in the wake of the recent financial crisis. In response to these concerns, regulators have evaluated their own performance and standards to ensure that their efforts do not unnecessarily constrain the availability of credit to worthy borrowers. Moreover, this situation is occurring against the backdrop of new regulation under the Dodd–Frank Act and Basel III that is designed to enhance the macroprudential supervision of financial institutions and provide countercyclical capital buffers.
In this heightened regulatory environment, the authors consider the variation in supervisory stringency or rigor that may be evident in composite CAMELS ratings over a 22-year period that encompasses credit downturns in the early 1990s and early 2000 and the 2007–09 recession. They expand on the limited earlier literature on the subject by considering a larger sample, using improved measures of macroeconomic conditions, and comparing a number of econometric specifications to avoid deficiencies of any particular evaluation methodology.
The authors conclude that supervisory standards used to assign these ratings are mostly consistent and stable over the examination period. Additionally, the degree of stringency estimated from the CAMELS ratings depends to some extent on the models used. Using macroeconomic variables instead of time-period fixed effects to control for macroeconomic conditions seems to imply decreased stringency on several occasions over the time period considered. Finally, meaningful changes in lending standards and changes in lending itself are linked to the level of supervisory rigor that the models identify.
Academics and regulators will find the authors’ conclusions interesting and informative. Portfolio managers and analysts covering the banking sector may want to consider the salient points of this research in their ongoing appraisal of financial institutions.
How Did the Authors Conduct This Research?
The existing body of literature on the effects of banking supervision and economic activity is fairly robust, but the research on bank supervisors’ changes to standards and policy over economic cycles and the effects of these changes on the economy are less so. The authors observe deficiencies in previous researchers’ use of time-period fixed effects, which may pick up extraneous regulatory responses to broader market conditions at the expense of regional or state control variables and fail to allow for differential changes in standards across different CAMELS rating criteria.
From here, the authors devise an improved process to incorporate macroeconomic variables and time-varying standards in the CAMELS ratings. The data include supervisory data on CAMELS ratings along with bank-specific financial statements, capital adequacy, asset and management quality, earnings, liquidity, market risk sensitivity, and macroeconomic and financial sector variables. Econometric parameters also consider the bank’s structure—for example, whether it is a standalone entity or part of a bank holding company with different regulatory remit and more robust oversight. The period under examination is from 1991 to 2013.
The heterogeneity of banking institutions is likely evident in the CAMELS ratings, notwithstanding extensive use of control variables in the models. The infrequency of and irregularly timed supervisory exams preclude a simple model analysis. Therefore, the authors use different panel data models—all of which exhibit various deficiencies.
Results across the three regression models indicate that the same variables are statistically significant, which suggests that the outcomes are robust to these models’ specification errors. How they indicate degree of supervisory stringency points to a limitation of the methodology that measures it at a given point in time relative to the average over the sample period rather than an absolute level over time. Exclusion of macroeconomic variables materially affects the outcome in terms of degree of supervisory stringency. Their inclusion, by contrast, mitigates the cyclicality of CAMELS ratings. The authors’ conclusions withstand robustness checks for such things as selection bias and exclusion of less than full-scope examinations. Market dislocations can result in changes in supervisory stringency that may influence economic activity, but overall, supervisory standards for CAMELS ratings have adhered to historical norms.
Abstractor’s Viewpoint
The authors undertake a rigorous evaluation of how changes in standards that banks supervisors use to assign CAMELS ratings affect supervisory stringency over commercial banks and, by extension, economic activity. They use several different types of models to simulate CAMELS ratings and compare their output with the actual ratings. Discrepancies observed relate to lending cycle fluctuations. For the most part, CAMELS rating standards have been consistent over time. The results hold up to changes in specifications and robustness tests. Possible additional research could consider items other than the use of CAMELS ratings to influence bank risk management—for example, loan classification supervisory standards. Alternative proxies for management quality and access to qualitative measures could widen the scope of further research.