Skills and earnings revisited

Regression and neural network models of wage determination are constructed where the explanatory variables include detailed information about skills. People skills, strategic skills, and IT skills all carry strong and significant wage premia; problem-solving skills (surprisingly) and physical skills (less surprisingly) do not. In contrast to the impact of school curriculum on subsequent earnings, the neural network modelling procedure does not pick up any significant nonlinearities in the relationship between skills and earnings.