Arresting Financial Crises
The Fed versus the Classicals
Nineteenth-century British economists Henry Thornton and Walter Bagehot established the classical rules of behavior for a central bank, acting as lender of last resort, seeking to avert panics and crises: Lend freely (to temporarily illiquid but solvent borrowers only) against the security of sound collateral and at above-market, penalty interest rates. Deny aid to unsound, insolvent borrowers. Preannounce your commitment to lend freely in all future panics. Also lend for short periods only, and have a clear, simple, certain exit strategy. The purpose is to prevent bank runs and money-stock collapses—collapses that, by reducing spending and prices, will, in the face of downward inflexibility of nominal wages, produce falls in output and employment.
In the financial crisis of 2008–09 the Federal Reserve adhered to some of the classical rules—albeit using a credit-easing rather than a money stock–protection rationale—while deviating from others. Consistent with the classicals, the Fed filled the market with liquidity while lending to a wide variety of borrowers on an extended array of assets. But it departed from the classical prescription in charging subsidy rather than penalty rates, in lending against tarnished collateral and/or purchasing assets of questionable value, in bailing out insolvent borrowers, in extending its lending deadlines beyond intervals approved by classicals, and in failing both to precommit to avert all future crises and to articulate an unambiguous exit strategy. Given that classicals demonstrated that satiating panic-induced demands for cash are sufficient to end crises, the Fed might think of abandoning its costly and arguably inessential deviations from the classical model and, instead, return to it.