Purpose – This paper uses a unique survey of Roma and non-Roma in South Eastern Europe with the aim of evaluating competing explanations for the poor performance of Roma in the labour market. Design/methodology/approach – Following a descriptive analysis, econometric models are employed to identify the determinants of educational achievement, employment and wages for Roma and non-Roma. Limited information maximum likelihood (LIML) methods are employed to control for endogenous schooling and two sources of sample selection bias in the estimates. Non-linear and linear decomposition techniques are applied in order to identify the extent of discrimination. Findings – The key results are that: the employment returns to education are lower for Roma than for non-Roma whilst the wage returns are broadly similar for the two groups; the similar wage gains translate into a smaller absolute wage gain for Roma than for non-Roma given their lower average wages; the marginal absolute gains from education for Roma are only a little over one-third of the marginal absolute gains to education for majority populations; and, there is evidence to support the idea that a substantial part of the differential in labour market outcomes is due to discrimination. Research limitations/implications – The survey data employed do not include information on hours worked. In order to partially control for this, the analysis of wages is limited to employee wages excluding the self-employed. Practical implications – Explanations of why Roma fare so badly tend to fall into one of two camps: the “low education” and the “discrimination” schools. The analysis suggests that both of these explanations have some basis in fact. Moreover, a direct implication of the lower absolute returns to education accruing to Roma is that their lower educational participation is, at least in part, due to rational economic calculus. Consequently, policy needs to address both low educational participation and labour market discrimination contemporaneously. Originality/value – This is the first paper to attempt to econometrically distinguish between discrimination and educational explanations of Roma disadvantage in the labour market in Central and Eastern Europe. The survey data employed are unique and appropriate for the task. Unusually for analyses dealing with returns to education, the LIML econometric approach employed controls for both endogenous schooling and two sources of sample selection bias.