We develop a dynamic model of learning and wage determination.
Education may convey initial information about ability, but subsequent
performance observations also are informative. Although the role of
schooling in the labor market's inference process declines as performance
observations accumulate, the estimated effect of schooling on the level of
wages is predicted to be independent of labor-market experience. The model
also predicts that time-invariant variables correlated with ability but
unobserved by employers should be increasingly correlated with wages as
experience increases and that wage residuals should be a martingale.
We present evidence from the National Longitudinal Survey of Youth that
is generally consistent with the model's predictions, but a chi-squared
goodness-of-fit test does reject the martingale prediction for wage residuals
even after accounting for classical measurement error. We investigate
alternative specifications and find that a modification of the learning model
that allows for worker ability to evolve as an AR1 process fits the data
quite well.