The identification of aggregate human capital externalities is still not fully understood. The existing (Mincerian) approach confuses positive externalities with wage changes due to a downward sloping demand curve for human capital. As a result, it yields positive externalities even when wages equal marginal social products.

We propose an approach that identifies human capital externalities whether or not aggregate demand for human capital slopes downward. Another advantage of our approach is that it does not require estimates of the individual return to human capital. Applications to US cities and states between 1970 and 1990 yield no evidence of significant average-schooling externalities.