Large and State-Dependent Effects of Quasi-Random Monetary Experiments

Abstract

The trilemma of international finance explains why interest rates in countries that fix their exchange rates and allow unfettered cross-border capital flows are largely outside the monetary authority’s control. Using historical panel-data since 1870 and using the trilemma mechanism to construct an external instrument for exogenous monetary policy fluctuations, we show that monetary interventions have very different causal impacts, and hence implied inflation-output trade-offs, according to whether: (1) the economy is operating above or below potential; (2) inflation is low, thereby bringing nominal rates closer to the zero lower bound; and (3) there is a credit boom in mortgage markets. We use several adjustments to account for potential spillover effects including a novel control function approach. The results have important implications for monetary policy.