The Effect of Education on Employment of Individuals Close to Retirement Age

Abstract

In labour economics, numerous studies are looking at the link between education and earnings. Instrumental-variable models are often used because education is endogenous: worker ability is usually unobserved and hypothesized to be correlated with both education and earnings. Angrist and Krueger (1991) used quarter-of-birth as an instrument for education of compulsory school attendance on education and earnings. The instrument quarter-of-birth is referred to which quarter people are born, and it is related to education because Angrist and Krueger showed that people who are born in the 3rd and 4th quarter tend to have a higher education than people who are born in the 1st and 2nd quarter under the compulsory school law. In this paper, I use the same instrument to study the effect of education on the employment probability of retirement-age workers. Such information is not available in most censuses and surveys. I conclude that education increases the probability of employment: explicitly, one more year of education would increase the employment probability of retirement-age workers by 1- 8 percentage points, depending on the model used, people's age and marital status.