Long and Variable Lags in Argentina's Monetary Policy: Evidence from Disaggregated Price Indices

Using monthly data for 2016-2025, I identified unexpected policy innovations as residuals from a Taylor rule and estimated Local-Projection impulse responses for headline CPI, tradables/non-tradables, and nine disaggregated components, treating those innovations as monetary policy shocks. Tightening is no quick fix: a +10 pp policy-rate shock leaves monthly inflation above baseline for roughly two years and cumulates sizable price-level gains (approx. 7-8 pp at one year, remaining positive at two to four years). Movements are faster and larger in tradables; services adjust more slowly, with wide heterogeneity across categories. Inference relies on heteroskedasticity- and autocorrelation-consistent (HAC) standard errors with wild-bootstrap checks. These patterns indicate long and variable lags in this setting; effective disinflation requires persistence and coordination with complementary instruments. Evidence suggests the monetary policy rate was either ineffective or not the right instrument to achieve price stability, perhaps favoring the choice of monetary-aggregate targeting for Argentina. I find early 90% significance for most series and zero-hit times clustering near two years.