Cross-validating administrative and survey datasets through microsimulation and the assessment of a tax reform in Luxembourg

Using EUROMOD, we cross-validate two types of micro-data presently available in the Grand-Duchy of Luxembourg, administrative data on one hand and survey data on the other hand. While administrative data, extracted from the recently implemented Social Security Data Warehouse, contain information of the whole population of Luxembourg (449,000 observations) in 2003, survey data, extracted from the Luxembourg household panel PSELL3/EU-SILC for 2004 (incomes from 2003), is a representative sample of around 3,600 private households (9,800 individuals) living in Luxembourg with detailed information on incomes, household structure and other socio-economic dimensions. As a concrete application of this cross-validation, we analyze the 2001-2002 tax reform in Luxembourg. The main aspects of this reform are the reduction of the number of the tax brackets and the fall of the maximal marginal tax rate (from 46% in 2000 to 42% in 2001 and to 38% in 2002). The distributional effects of the tax reform are measured in terms of losers and winners, change in inequalities and poverty rates. The results issued from different types of input data are compared for cross-validation and allow us to emphasize methodological difficulties as well as to underline the advantages and limitations of each dataset.