Gender Wage Gap in Rural Bangladesh: Assessing the Sticky Floor or Glass Ceiling Phenomenon and Its Determinants

This paper empirically examines the determinants of the gender wage gap across the wage distribution in rural Bangladesh, applying the Oaxaca–Blinder decomposition to the unconditional quantile regression approach. Using panel data from three rounds (2011–12, 2015, and 2019) of the Bangladesh Integrated Household Survey, this study controls for both household-level unobserved heterogeneity and selection bias in rural wage employment. Results reveal that the gender wage gap is most pronounced at the lower end of the distribution, providing strong evidence of the sticky floor phenomenon. We find that larger return to education and coefficient of full-time employment for females significantly contributed to a reduced wage gap at the upper end of the distribution, while larger return to education and coefficient of non-agricultural employment for females contributed to narrowing the wage gap at the lower end. Notably, as education is associated with a higher probability of wage employment for females, the contribution of education becomes stronger after controlling for selectivity bias. The findings suggest that promoting female education and expanding access to non-agricultural employment are key to reducing the gender wage gap in rural Bangladesh, although unobservable factors continue to perpetuate the sticky floor phenomenon.