Choques hidrometeorológicos extremos e inflación: evidencia para Costa Rica

Abstract

In this paper, we analyze the effect of extreme hydrometeorological events on inflation in Costa Rica (2010–2025). We use precipitation data from “Climate hazards Center InfraRed Precipitation with Stations"(CHIRPS) to define extreme wet and dry shocks based on the historical 95th and 5th percentiles, respectively. We use the local projections method applied to an augmented Phillips curve to estimate impulse response functions from these shocks over a 12-month horizon. The results reveal significant asymmetric effects: wet events cause an immediate but transitory increase in prices, particularly in agricultural sub-indices, followed by a downward correction, whereas dry events generate a gradual increase that becomes statistically significant with a longer lag. Furthermore, we identify spatial heterogeneity by clustering municipalities according to their climate anomaly patterns. Finally, we find that the occurrence of consecutive extreme events amplifies the shock’s effect and its persistence, and quantile regressions confirm pronounced tail inflationary risks at the 90th percentile of the conditional distribution. These findings highlight the need to employ disaggregated climate data and explicitly incorporate their effects into macroeconomic forecasting models.