A review of the measures of the stock of human capital used in empiricalgrowth research - including adult literacy rates, school enrollmentratios, and average years of schooling of the working-age population -reveals that human capital is mostly poorly proxied. The simple use ofthe most common proxy, average years of schooling, misspecifies therelationship between education and the stock of human capital. Based onhuman capital theory, the specification of human capital is extended toallow for decreasing returns to education and for differences in thequality of a year of education. The different specifications give riseto hugely differing measures of the stock of human capital acrosscountries, and development-accounting results show that misspecifiedhuman capital measures can lead to severe underestimation of thedevelopment effect of human capital.
