Causal effects of reserve requirements on the business cycle in Costa Rica: a narrative approach

We estimate the effect of reserve requirements (RR) on production and employment in Costa Rica using local projections with an instrumental variable. Our instrumental variable for RR is constructed using a narrative approach, which involves classifying changes in reserve requirements as endogenous or exogenous. Additionally, we propose two robustness exercises. In the first exercise, we randomly reclassify a percentage of changes originally classified as endogenous as exogenous, and vice versa. We repeat this process 1000 times using Monte Carlo simulations. In the second exercise, we estimate the changes in RR that can be predicted using a linear probability model. The changes found to be predictable are reclassified as endogenous, and the model was re-estimated. We find that a 1 percentage point increase in reserve requirements reduces production by 1.56% after one year, while reducing employment by 1.34%.