How Sensitive Are Bank Market Values to Regulatory Adjustments of Capital?

Abstract

Listen

Translate article

Synopsis The research problem We measured the sensitivity of bank market values to capital and regulatory adjustments (RAs) applied to bank capital using a novel approach of measuring value relevance. Motivation or theoretical reasoning Regulators require banks to apply adjustments to book equity to calculate regulatory capital, where book equity is the starting point of the calculation of capital ratios. The regulators emphasize the benefits of these adjustments. Making banks hold more capital, the adjustments should contribute to the reduction of procyclical amplification of financial shocks throughout the banking system, financial markets, and the broader economy. However, RAs are underresearched: empirical research focuses on a few of them, such as the prudential filter on AFS securities, and their effects in specific circumstances. This prompts questions about the overall relevance of RAs. This paper then examines the sensitivity, expressed in elasticities, of market values to capital and RAs. Are adjustments, on their own or in aggregate, value relevant? Are market values more or less sensitive to different measures of bank capital, such as book equity, Tier 1, or total capital? Does the sensitivity evolve over time? And if so, why? The test hypotheses H1: The elasticity of bank book equity with respect to market value is 1. H2: The elasticities of RAs with respect to market values are 0. Target population US bank holding companies over the years 2001Q1–2022Q3. Adopted methodology We used panel data analyses to estimate log–log models. These models transform all variables such as market values and accounting values into logs. Analyses We used log–log models as an alternative to traditional additive-linear models because the coefficients of the latter are sensitive to sample choice, choice of time period, and outlier treatment. The response coefficients in log–log models are scale-free elasticities that measure, with some precision, the proportional change in the dependent variable associated with a proportional change in the independent variables. Findings Our results show that the sensitivity of bank book equity converges to 1 when market uncertainty is low and when banks’ Tier 1 ratios reach 12% of risk-weighted assets (RWAs). Market values are more sensitive to changes in capital of highly geared banks when market uncertainty is high, with shareholders responding positively in particular to increases in Tier 1 and total capital. This is consistent with risk-shifting by shareholders. Market values are generally less sensitive to prudential filters, such as those on unrealized gains and losses on AFS securities.