1. Introduction
Banks facing external-financing frictions, such as the Myers and Majluf (1984) adverse-selection problem, cannot
immediately restore equity capital reductions caused by economic downturns. The capital crunch theory predicts that
capital adequacy regulation combined with market imperfections leads to pro-cyclical bank lending. Specifically, banks
reduce lending more to avoid potential future violations of regulatory capital minimums during recessions relative to
expansions. In a May 7, 2009 speech entitled ‘‘Lessons of the Financial Crisis for Banking Supervision,’’ Ben Bernanke,
Federal Reserve Bank (FED) chairman, argued that ‘‘working to mitigate pro-cyclical features of capital regulation and
other rules and standards’’ is an important element in enhancing the stability of the financial system as a whole.
Regulators and policy makers argue that current loan loss provisioning rules reinforce the pro-cyclical capital effect.
The extent to which the recognition of expected losses is delayed affects the ability of loan loss reserves to cover credit
losses during economic downturns. When loan loss reserves cannot absorb recessionary credit losses, greater provisioning
is required and reduces capital adequacy, potentially accentuating capital pro-cyclicality.2 Regulators argue that current
rules contribute to pro-cyclicality by delaying the recognition of expected losses. Given these macro-economic concerns
with the current incurred-loss provisioning method, which is used throughout most of the world, regulators and policy
makers are considering alternative loan loss methods. he Financial Stability Forum (FSF) Report (2009) discusses difficulties banks experiencing extensive losses face in
replenishing their capital.3 The report states that during economic downturns ‘‘a weakened financial system cannot absorb
further losses without causing amplifying retrenchment.’’ The FSF identifies loan loss provisioning as one of three policy
action priorities addressing the driving forces of positive feedback between the financial and real sectors. Comptroller of
the Currency John C. Dugan’s remarks on March 2, 2009 to the Institute of International Bankers entitled ‘‘Loan Loss
Provisioning and Pro-cyclicality’’ reflect these concerns. He argues that the limitation on judgment in applying the current
‘‘incurred loss model,’’4 was a fundamental constraint leading ‘‘loan loss provisioning [to] become decidedly pro-cyclical,
magnifying the impact of the [current economic] downturn.’’
While regulators argue that delaying the recognition of expected losses exacerbates pro-cyclicality, the empirical evidence is
limited. We exploit cross-sectional differences in the application of the incurred loss model to examine whether delays in
expected loss recognition affect pro-cyclicality of banks’ lending.5 Based on the capital crunch theory, we first hypothesize a
higher association between lending and capital ratios in recessions versus expansions in the presence of numerical capital
adequacy regulation. Our primary analysis focuses on the period after implementation of the 1988 Basel Risk Based Capital
Regulation and the Federal Depository Insurance Corporation Improvement Act of 1991 (FDICIA). The effect of these regulations
likely differed with bank size, thus we also examine whether the capital crunch effect differs for large versus small banks.
Based on regulators’ arguments that higher recessionary provisioning increases banks’ future capital inadequacy
concerns, we hypothesize a greater reduction in recessionary lending for banks with a greater versus smaller delay in
expected loss recognition. Furthermore, we predict that the lending-capital ratio link during recessions is higher for banks
with a greater delay. Finally, to corroborate the effect of provisioning on pro-cyclicality, we also examine how equity
changes unrelated to the provision differ for banks with greater versus smaller delays during recessions versus expansions,
to better understand how provisioning differences affect lending pro-cyclicality. We predict that banks that delay expected
loss recognition less raise more capital or reduce dividends to maintain their regulatory capital ratios during expansions.
In supplemental tests, we examine whether capital constraints arise directly from regulatory capital requirements by
conducting the same analysis using a period prior to explicit capital regulation. Specifically, we examine the twenty-year
period prior to the introduction of the first explicit regulatory requirements in December 1981.
We test our hypotheses using three different measures of delay in expected loss recognition. The first is a loan loss
specific metric based on Nichols et al. (2009). This measure is the incremental explanatory power of future and
contemporaneous nonperforming loans, beyond that of past nonperforming loans, in explaining the current loan loss
provision. It has the advantage of being a direct measure