The Formal/Informal Employment Earnings Gap: Evidence from Turkey

Abstract This study investigates the formal/informal employment earnings gap in Turkey. We focus on the earnings differentials that can be explained by observable characteristics and unobservable time-invariant individual heterogeneity. We first, estimate the standard Mincer earnings equations using ordinary least squares (OLS), controlling for individual, household, and job characteristics. Next we use, panel data and the quantile regression (QR) techniques in order to account for unobserved factors which might affect the earnings and the intrinsic heterogeneity within formal and informal sectors. OLS results confirm the existence of an informal sector penalty almost half of which is explained by observable variables. We find that formal-salaried workers are paid significantly higher than their informal counterparts and of the self-employed confirming the heterogeneity within the informal employment. QR results show that pay differentials are not uniform along the earnings distribution. In contrast to the mainstream literature which views informal self-employment as the upper-tier and wage-employment as the lower-tier, we find that self-employment corresponds to the lower-tier in the Turkish labor market. Finally, fixed effects estimation indicates that unobserved individual characteristics combined with controls for observable characteristics explain the pay differentials between formal and informal employment entirely in the total and the female sample. However, informal sector penalty persists in the male sample.