Bank Procyclicality, Credit Crunches, and Asymmetric Monetary Policy Effects: A Unifying Model

Article excerpt

Much concern has recently been expressed that both large, procyclical changes in bank assets and "credit crunches " caused by banks' reluctance to expand loans during recessions contribute to economic instability. These effects are difficult to explain using the standard textbook model of deposit expansion, in which deposits are constrained only by reserve requirements. However, these effects follow easily if the model is expanded to include a second, capital constraint. [E51, E32, G21]

Much concern has been expressed recently about the perceived excessive procyclicality of banks, which may exacerbate the cyclical behavior of the macroeconomy and, in particular, hamper recoveries from recessions. As economists at the Bank for International Settlements have recently noted "[f]inancial developments have reinforced the momentum of underlying economic cycles, and in some cases have led to extreme swings in economic activity. . . . These experiences have led to concerns that the financial system is excessively procyclical, unnecessarily amplifying swings in the real economy" (Borio, Furfine, and Lowe, 2001).

Although most industries experience cyclical movements in output and profitability in sympathy with the cyclical swings in the economy as a whole, such cyclicality in bank assets, loans, and capital tends to exceed that in the macroeconomy as well as in many other individual sectors of the economy, expanding faster in upturns and contracting faster in downturns. This pattern is perceived to be more important for banks than for most other sectors of the economy because banks provide demand deposits, the largest part of the money supply (M1 and M2), and are the major provider of credit to many sectors of the economy. Furthermore, the Federal Reserve uses banks as its primary channel for transmitting monetary policy. Thus, fluctuations in bank deposits and credit have significant, indeed critical, effects on macroeconomic activity and may amplify swings in the macroeconomy.

In addition, the ability of the economy to recover from recessions may be restricted, because banks are unwilling or unable to increase their loans or total credit to satisfy the increasing demand for such loans or credit. This can occur even under an expansive Federal Reserve monetary policy, which increases bank reserves. The result may be a "credit crunch," characterized by sharp increases in effective bank loan rates and widespread reports of unmet credit needs during periods of perceived expansive monetary policy. Although the empirical evidence of the existence of credit crunches is inconclusive, primarily because of the inability to clearly differentiate between demand and supply forces, such crunches, if they do exist, may partially or totally frustrate the intended impact of the expansive monetary policy.

However, perceived credit crunches, excessive procyclicality in bank behavior, and limited effectiveness of expansionary Fed monetary policy in economic recessions are not easily predicted by the usual simple textbook bank deposit or bank credit expansion model as outcomes of an expansive monetary policy. But, with a relatively simple modification to this model that introduces a market or regulatory capital constraint in addition to the traditional reserve constraint, we demonstrate in this article that all three effects may be predictable outcomes.1 This contribution provides new insights that may permit policy-makers to offset the negative impact of these effects or avoid them altogether.

I. The Textbook Model: A Single Constraint

Typical textbook models (e.g., Mishkin, 2001 and Kaufman, 1995) constrain aggregate bank deposit and earning asset expansions on the supply side only by reserve requirements, usually expressed as a percent of deposits. Reserves are held by banks both voluntarily against the possible liquidity demands of depositors wishing to withdraw funds and by statute to satisfy requirements imposed by central banks. …