The impact of quality certifications on firms' performance : a random forest diff-in-diff approach

Abstract

This study evaluates the impact of internationally recognized quality certification (QC) adoption on firms’ performance. Using a panel dataset of 4,668 firms from Latin America and the Caribbean we follow a two-stage identification strategy: (i) We estimate firms’ probability of QC adoption using Random Forest algorithm, and (ii) we use the estimated probabilities for applying a weighted Diff-in-Diff approach. Our findings show that acquiring a QC has a positive effect on firm export behavior. Interestingly, we find that this effect is driven by an increase in both the intensive and extensive margin of indirect exports. QC also helps to ease constraints in access to finance but no effect was found on Local Sales and various measures of firm productivity. However, our analysis reveals that positive impacts of QC on firms’ performance are contingent on firm size: it only benefits small enterprises. This evidence may be consistent with the idea that QC adoption in developing countries contributes mainly to the growth of small enterprises in supply chains by reducing information asymmetries.