Determinan Pendapatan Tenaga Kerja di Sumatera Barat: Pendekatan Model Heckman Two-Step

Abstract

This study analyzes the determinants of labor income in West Sumatra while accounting for sample selection bias, because income is observed only for individuals who work and report earnings. The analysis uses microdata from the August 2024 National Labor Force Survey (Sakernas) with 23,875 observations (14,260 workers with positive income), and applies the Heckman two-step model. The first stage estimates the probability of working with a probit model, and the second stage estimates a Mincer-type log income equation augmented with the Inverse Mills Ratio. Education, experience, marital status, sector, gender, and residence area are significantly associated with income, while dependents are not. One additional year of schooling is associated with 3.80 percent higher income, and workers in the tertiary sector earn 44.4 percent more than those in the primary sector. In contrast, women earn 30.6 percent less than men and rural workers 16.2 percent less than urban workers. Experience has a concave relationship with income, with a turning point at about 30.7 years. The Inverse Mills Ratio is significant at the 5 percent level, so the selection correction is relevant. Policies that raise human capital and narrow gender and regional gaps are needed to improve labor income.