The availability of microdata on both income and expenditure is highly recommended if one wants to assess the distributional consequences of policy changes. In Italy, the main sources used for estimating household income and expenditure are the Bank of Italy's Survey on Household Income and Wealth and the Italian National Institute of Statistics Household Budget Survey.
However, there is no single data source containing information on both expenditure and income. The problem is generally overcome with statistical matching procedures based on the conditional independence (CIA) assumption.
The aim of this paper is to present a method to combine information coming from different databases relaxing the CIA assumption. In particular we propose a method to combine household income and expenditure data under logical constraints regarding the average propensity to consume. We also propose an estimate of a plausible joint distribution function for household income and expenditure.
Forthcoming in: Statistical Methods & Applications