Assessing the heterogeneous impact of economy-wide shocks: a causal machine learning approach applied to Colombian firms

Our paper presents a causal Machine Learning (ML) methodology to study the heterogeneous effects of economy-wide shocks and applies it to the impact of the COVID-19 crisis on exports. This method is applicable in scenarios where, due to the pervasive nature of the shock, it is difficult to identify a control group that is not affected by the shock and to determine ex ante differences in shock intensity across units. In particular, our study investigates the effectiveness of different machine learning techniques in predicting firms' trade and, by building on recent developments in causal ML, these predictions are used to reconstruct the counterfactual distribution of firms' trade under different COVID-19 scenarios and investigate the heterogeneity of treatment effects. Specifically, we focus on the probability of Colombian firms surviving in the export market under two different scenarios: a COVID-19 setting and a non-COVID-19 counterfactual situation. On average, we find that the COVID-19 shock decreased a firm's probability of surviving in the export market by about 20 percentage points in April 2020. We study the treatment effect heterogeneity by employing a classification analysis that compares the characteristics of the firms on the tails of the estimated distribution of the individual treatment effects.