The Gender Wage Gap in Uzbekistan, 2022-2023: An Oaxaca-Blinder Decomposition Using the Life in Transition Survey, With a 2013 Historical Comparison

This paper provides a micro-level econometric decomposition of the gender wage gap in Uzbekistan using individual-level data from the 2022-2023 Life in Transition Survey IV (LiTS IV; EBRD/World Bank), the most recent publicly available survey combining individual wages, hours, occupation, sector, and education for Uzbekistan. For the wage-earning subsample (N=283; 132 men, 151 women), the raw monthly wage gap is 24-26% (log-based, unweighted/weighted), while the hourly wage-rate gap, measured robustly via the median and geometric mean to avoid distortion from a small number of very-low-hours outliers, is substantially smaller at approximately 7-8%, indicating that most of the monthly gap reflects men working more hours than women rather than a large difference in the wage rate itself. Using a Mincer-type log-wage regression estimated separately by gender with survey weights and standard errors clustered on primary sampling unit, we decompose the monthly wage gap with the Oaxaca-Blinder method. Controlling for sector of employment (11 categories), 32-38% of the gap is associated with observable characteristics, driven principally by women's underrepresentation in high-paying construction work and overrepresentation in the public sector; the remainder is unexplained. Uniquely, LiTS IV records respondents' actual occupation (ISCO-08 major groups) rather than only industry sector; when occupation replaces sector in the model, the explained share falls to near zero (-4% to +6%, unweighted/weighted), indicating that essentially the entire monthly gap is statistically unexplained by workers' measured occupation, education, age, and location. We interpret this unexplained residual as consistent with, but not proof of, the normative gender-biased wage-setting documented in survey experiments in Central Asia, including Uzbekistan, by Seitz (2023), while cautioning that our data lack measures of actual work experience and bargaining power. To provide historical context, we apply the same methodology to the 2013 Jobs, Skills, and Migration Survey (JSMS; N=2,059): the monthly gap has narrowed from approximately 40-42% in 2013 to 24-26% in 2022-2023, and the wage-rate (hourly) gap has narrowed even more sharply, from approximately 25-28% to 7-8%, a pattern consistent with, though not proof of, the intended effects of Uzbekistan's 2017-2022 reform program, including the 2022 Labor Code's equal-pay provisions. We discuss policy implications and the limitations of both surveys' small samples for future research.