View article

We provide an alternative econometrics methodology to estimate a standard heterogeneous income profiles (HIP) model. Our alternative setup allows for the HIP coefficients to be fixed in the sense that they can be arbitrarily correlated with the explanatory variables of the HIP equation and can be treated as parameters to be estimated. As an empirical application, we analyse the extent to which different sources of labour income shocks account for the persistence and variation of composite income shocks based on the HIP model. Our estimation results from the Panel Study of Income Dynamics (PSID) suggest that the random effect assumption of no correlation can lead to biases. We also find that job displacements account for the persistence of income shocks the most, especially for high school educated individuals.