| Sumario: | The OECD’s “Average-Wage” (AW) concept is commonly used as a benchmark for tax-benefit and pension modeling. The purpose of this paper is to examine whether it is possible to use richer sets of earnings data in order to customize these modeling exercises to the situation of different groups of workers, such as high or low-earning men or women. We first take stock of available sources of earnings distribution data and provide a careful assessment of measurement and definitional differences relative to the AW. In a second step, information on the shape of earnings distributions in OECD countries is used to derive synthetic distributions around the AW, distinguishing between the earnings of men and women. We argue that this pragmatic approach yields data that allow extending the scope of tax-benefit and pensions modelling. Moreover, it does so in a way that is consistent with past modeling exercises that relied on the AW. We highlight data quality issues and discuss the potential limitations of the imputed AW-consistent earnings distributions.
|
|---|
