Qualifications and earnings in Britain : how reliable are conventional OLS estimates of the returns to education?

The paper estimates the returns to education for a cohort of individuals born in Britain in March 1958 who have been followed since birth until the age of 33. The data used has a wealth of information on family background including parental education, social class and interest shown in the child's education as well as measures of ability. The nature of our data allows us to directly assess the relative importance of omitted ability and family background bias as well as biases arising from measurement error in education qualification variables which have been found to be important in other studies. The paper also looks at possible biases arising from compositional differences between individuals in work and those out of work. This 'Composition bias' arising from self-selection into employment is generally ignored in the returns to schooling literature and is why most studies focus only on men (for whom it is assumed this is much less of a problem). The paper also examines whether there is evidence of heterogeneity in the returns to education as well as the impact of education on gender wage differentials. The paper finds that conventional OLS estimates, which assume that education is exogenous, are reasonable estimates of the true causal impact of education on wages. In the UK it would appear that the effects of measurement error bias and composition bias directly offset the countervailing effect of unobserved ability and family background bias for most qualifications. The results from the paper suggest that conventional OLS estimates of the returns to education can generally be relied upon for policy decisions. The paper also finds evidence of heterogeneity in the returns to education in Britain. The results from the paper suggest that individuals undertaking schooling involving some sort of formal qualification have significantly larger rates of return than individuals who complete the same number of years of schooling but who obtain no formal qualifications. There is also some evidence that individuals with lower tastes for education, have significantly higher marginal returns to certain education qualifications. We also find that post-school qualifications, particularly degree qualifications, play an important role in reducing gender wage differentials. The data used has a wealth of information on family background including parental education, social class and interest shown in the child's education as well as measures of ability. The nature of our data allows us to directly assess the relative importance of omitted ability and family background bias as well as biases arising from measurement error in education qualification variables which have been found to be important in other studies. The paper also looks at possible biases arising from compositional differences between individuals in work and those out of work. This 'Composition bias' arising from self-selection into employment is generally ignored in the returns to schooling literature and is why most studies focus only on men (for whom it is assumed this is much less of a problem). The paper also examines whether there is evidence of heterogeneity in the returns to education as well as the impact of education on gender wage differentials. The paper finds that conventional OLS estimates, which assume that education is exogenous, are reasonable estimates of the true causal impact of education on wages. In the UK it would appear that the effects of measurement error bias and composition bias directly offset the countervailing effect of unobserved ability and family background bias for most qualifications. The results from the paper suggest that conventional OLS estimates of the returns to education can generally be relied upon for policy decisions. The paper also finds evidence of heterogeneity in the returns to education in Britain. The results from the paper suggest that individuals undertaking schooling involving some sort of formal qualification have significantly larger rates of return than individuals who complete the same number of years of schooling but who obtain no formal qualifications. There is also some evidence that individuals with lower tastes for education, have significantly higher marginal returns to certain education qualifications. We also find that post-school qualifications, particularly degree qualifications, play an important role in reducing gender wage differentials.