Uncertainty and Heterogeneity in Returns to Education: Evidence from Finland

This paper studies the causal effect of education on income uncertainty using a broad measure of income which encompasses unemployment risk. To accomplish this, the variance of residuals from a Mincer-type income regression is decomposed into unobserved heterogeneity (known to the individual when making their educational choices) and uncertainty (unknown to the individual). The estimation is done using Finnish registry data. The marginal effect of having a secondary or a lower tertiary level education decreases income uncertainty. University level education is found to have a small positive marginal effect on income uncertainty. The effect of education on income uncertainty is roughly similar for men in comparison to women, but income uncertainty is larger for men than for women regardless of education. Contrary to some results from the U.S., the role of unobserved heterogeneity is found to be very small.