Marginal College Wage Premiums Under Selection Into Employment

Abstract We identify female long-term wage returns to college education using the educational expansion between 1960–90 in West Germany as exogenous variation for college enrolment. We estimate marginal treatment effects and propose a simple partial identification technique accounting for women selecting into employment due to having a college education. College-educated women are, on average, more than 18 percentage points more likely to be employed due to having a college education than those without college education. Taking this into account, we bound wage returns to 5.7%–13.9% per year of education completed (average treatment effects on the treated).