Rethinking Earnings Determinants in the Urban Areas of Bolivia

This paper analyzes the factors that explain earnings in levels and inequality in the urban areas of Bolivia, considering not only the usual individual characteristics (education, experience, gender, and ethnicity) but also firm characteristics. Given the information available at the firm level in the household surveys, two simple models were developed: one for independent workers (for which there is relatively detailed firm-level data), and the other for dependent workers (where firm variables were approximated by sector, size, and by the legal condition of the workers). The main econometric results show that: i) earnings regressions that include only individual variables present highly biased (overestimated) coefficients; ii) firm characteristics are fundamental factors for explaining earnings for independent workers, almost doubling R2 and explaining 45.5% of observed earnings inequality; and iii) firm proxies for dependent workers are also relevant; however, they explain wages at a lower percentage (11.8%), which may be due to non-detailed firm data and to the relatively higher importance of education for these workers. These new findings represent a contribution to the empirical literature on earnings determinants for urban Bolivia as well as to the vision of labor income and poverty problems.