Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This article develops and tests a simple dynamic model of statistical discrimination. The model improves on earlier static models both by allowing ex ante uncertainty about worker productivity to be resolved as on-the-job performance is observed and by generating several testable empirical implications. These predictions are tested using a sample of young men from the National Longitudinal Survey of Youth, producing mixed evidence for the model. The main empirical result is that no black-white wage gap exists at labor-force entry but that one develops as experience accumulates, mainly because blacks reap smaller gains from job mobility. Copyright 1996 by University of Chicago Press.