Dec 31, 2006
The authors provide both a theoretical and empirical analysis of the relation between administrative and survey data. By distinguishing between different sources of deviations between survey and administrative data they are able to reproduce several stylized facts in the literature. In doing so, they deviate from the almost universal assumption that the administrative data represent the truth. They illustrate the implications of different error sources for estimation in (simple) econometric models and find potentially very substantial biases, both when using survey data and when using administrative data. The analysis is applied to Swedish data that have been collected for a validation study as part of a larger European health and retirement study (SHARE: Survey of Health, Ageing, and Retirement in Europe). Thus this paper makes two contributions: (1) it adds to the limited number of empirical validation studies of earnings measurement in surveys and (2) it shows the sensitivity of some findings in the literature for the assumption that administrative data represent the truth. They find in particular that the common finding of substantial mean reversion in survey data largely goes away once we allow for a richer error structure.