Determinantes del empleo informal en Bolivia: Un análisis conjunto de técnicas econométricas tradicionales y métodos de machine learning

Abstract

This study examines the determinants of informal employment in Bolivia by combining traditional econometric techniques, machine learning methods, and hybrid approaches. Using data from the 2022 and 2023 Household Surveys, we identify individual and household-level factors influencing the likelihood of being in informal employment. The results show that variables such as age, education level, household income, and gender are key determinants. Random Forest highlights the central role of labor income, often excluded due to endogeneity concerns. Adaptive Lasso helps identify nonlinear relationships and complex interactions, such as those associated with gender, indigenous group membership, and the presence of young children in the household. We conclude that informal employment is a multidimensional phenomenon requiring integrated analytical approaches for the design of more effective and targeted public policies.