Determinants of nonfarm earnings of farm-based husbands and wives in Northern Ghana

The nonfarm work participation decisions of married men and women in rural Northern Ghana were jointly and separately estimated for married couples through a bivariate probit, using recent survey data. Selectivity bias was corrected for in estimating wage offer and labor supply equations, using Heckman's procedure. Education, experience, infrastructure, distance to the capital, and population density, as well as interactions between education and infrastructure and between education and distance to the city, were found to be significantly related to the probability of nonfarm labor market participation, wages, and the amount of nonfarm labor performed, with significant differences by gender.