DP11801 Large and state-dependent effects of quasi-random monetary experiments
|Author(s):||Òscar Jordà, Moritz Schularick, Alan M. Taylor|
|Publication Date:||January 2017|
|Keyword(s):||fixed exchange rates, instru- mental variables, interest rates, local average treatment effect, local projections, monetary experiments, trilemma|
|JEL(s):||E01, E30, E32, E44, E47, E51, F33, F42, F44|
|Programme Areas:||Economic History, International Macroeconomics and Finance, Monetary Economics and Fluctuations|
|Link to this Page:||www.cepr.org/active/publications/discussion_papers/dp.php?dpno=11801|
Fixing the exchange rate constrains monetary policy. Along with unfettered cross-border capital flows, the trilemma implies that arbitrage, not the central bank, determines how interest rates fluctuate. The annals of international finance thus provide quasi-natural experiments with which to measure how macroeconomic outcomes respond to policy rates. Based on historical data since 1870, we estimate the local average treatment effect (LATE) of monetary policy interventions and examine its implications for the population ATE with a trilemma instrument. Using a novel control function approach we evaluate the robustness of our findings to possible spillovers via alternative channels. Our results prove to be robust. We find that the effects of monetary policy are much larger than previously estimated, and that these effects are state-dependent.