The purpose of this paper is to investigate self-competence—the ability to act responsibly on one's own—and likely nonlinear wage returns across different levels of self-competence as part of training curricula.
The authors identify the teaching of self-competence at the occupational level by applying machine-learning methods to the texts of occupational training curricula. Defining three levels of self-competence (high, medium, and low) and using individual labor market data, the authors examine nonlinearities in wage returns to different levels of self-competence.
The authors find nonlinear returns to teaching self-competence: a medium level of self-competence taught in an occupation has the largest wage returns compared to low or high levels. However, in occupations with a high cognitive requirement profile, a high level of self-competence generates positive wage returns.
This paper first adds to research on the importance of teaching noncognitive skills for economic outcomes, which recently—in addition to personality traits research—has primarily focused on social skills by introducing self-competence as another largely unexplored but important noncognitive skill. Second, the paper studies not only average but also nonlinear wage returns, showing that the right level of self-competence is crucial, i.e. neither teaching too little nor too much self-competence provides favorable returns because of trade-offs with other skills (e.g. technical or professional skills). Third, the paper also examines complementarities between cognitive skills and noncognitive skills, again pointing toward nonlinear returns, i.e. only in occupations with a high cognitive requirement profile, high levels of self-competence generate positive wage returns.
