Combining household income and expenditure data in policy simulations

The analysis of the distributional impact of fiscal policy proposals often requires information on household expenditures and incomes. It is unusual to have one data source with information on both and this problem is generally overcome with statistical matching of independent data sources. In this paper Grade Correspondence Analysis (GCA) is investigated as a tool to improve the matching process. GCA draws out the relationships between the common variables to enable the sample to be partitioned into more homogeneous groups, prior to matching. An evaluation is conducted using the UK Family Expenditure Survey, which is unusual in containing both income and expenditure at a detailed level of disaggregation. Imputed expenditures are compared with actual expenditures through the use of indirect tax simulations. The most successful methods are then employed to enhance data from the Family Resources Survey and the synthetic dataset is used as a microsimulation model database.