The labor supply response to (mismeasured but) predictable wage changes

Problems with measurement error have led many researchers to criticize panel data studies of intertemporal labor supply. In this study I address the measurement error problems explicitly. I estimate the properties of measurement error in the Panel Study of Income Dynamics Validation Study. I then use this information about measurement error to purge measurement error from the Panel Study of Income Dynamics. I show there exists a large transitory wage changes predictably disappear. When estimating the labor supply response to these predictable wage changes, I account for serially correlated measurement error and for measurement error that is correlated with true hours and wages. I find almost no correlation between hours worked and predictable wage changes. Therefore, failure to control for measurement error cannot explain the low estimated labor supply elasticities in other papers.