Unlike in many other transition countries, where the gender pay gap has remained stable while female employment rates have reduced, in the case of Belarus women’ activity rate has been practically unchanged despite an increase in the gender pay gap. This paper investigates why this is the case by looking at the determinants of female labour force participation in 1996 and 2001 (data from the Belarusian Household Survey). The selectivity corrected wage equation is estimated to compute an expected wage offer for women. The latter is included, in the second step, as a regressor in the structural female labour supply equation, estimated by probit. Several measures for the care of children and elderly people, proxies for the opportunity cost of working, affect female participation, but do not generate sample selection mechanisms. The estimated elasticity of female participation to wages is low, at about 0.45 in 1996 and 0.41 in 2001. Moreover the data allows detecting poverty trap mechanisms, whereas women in low-income households have much lower than average participation rates. At the same time the elasticity of female labour supply with respect to the own wage appears to be much higher for the low-paid groups of women.