Does education improve labour market outcomes?

This paper estimates the causal relationship between years of schooling and unemployment. The results are derived from a high quality dataset containing observations on 13,896 twins. A control for twin pair specific heterogeneity is introduced by exploiting the twin dimension through the fixed effects estimator. Although the OLS estimator reports a strong negative correlation between education and unemployment, the relationship vanishes once a control for unobserved ability is introduced. These results are consistent with previous studies about the effect of education on unemployment in a Swedish context. Altogether, the evidence would suggest that the negative correlation between schooling and unemployment is a result of more able individuals self selecting into higher education