The soft part of wages : longitudinal data evidence for Chile

Abstract

Using a longitudinal database that follows 100,899 individuals through nine years of their life from high school into labor market, we are able to identify cognitive and non-cognitive abilities at school age and their effect on wages. For measuring the effect of soft skills on wages we use two proxies: average school attendance and school grades ranking. Since an individual’s productivity is compound of both abilities both types of skills may impact outcomes such as test scores, dropout rates, and wages. While the effect of cognitive skills is widely studied, yet little has been said about the role of soft skills on such outcomes in less developed countries. Our main finding is that non-cognitive abilities are as or more important as cognitive abilities in explaining wages. One standard deviation increase in math SIMCE score raises wages by 2.36% while the same change in grades ranking increases wages by 5.03%. These findings are particularly important in a context of a high inequality country like Chile. The fact that trainable non-cognitive abilities affect wages is another opportunity for social policy to reduce the existent wage gaps in a cost effective way.