Abstract

Understanding households’ economic well-being requires looking beyond income alone. Consumption and wealth provide complementary perspectives on households’ living standards, their ability to smooth shocks, and their longer-term economic resilience. Yet, these dimensions are rarely observed jointly in a single data source with sufficient quality and detail, limiting the scope for comprehensive distributional analysis.
This presentation will describe the Italian experience in addressing this information gap through an ongoing collaboration between Istat and the Bank of Italy. The project explores the use of statistical matching to integrate information from three major household surveys: the EU Statistics on Income and Living Conditions (EU-SILC), the Household Budget Survey (HBS), and the Survey on Household Income and Wealth (SHIW). The objective is to bring together complementary information on income, consumption and wealth to provide a more comprehensive picture of households’ economic conditions.
The presentation will focus on the statistical matching process and on the methodological and practical choices involved in combining information from different survey sources. Alternative matching strategies, using different surveys as reference datasets, provide a useful framework for assessing the robustness of the resulting joint distributions. Particular attention will be given to the selection of matching variables, differences in concepts and populations across surveys, the preservation of marginal and conditional distributions, sensitivity to alternative matching specifications, and the uncertainty associated with reconstructing relationships between variables that are not jointly observed.
The discussion will highlight both the potential and the limitations of this approach. Statistical matching can exploit complementary high-quality data sources, enrich the information available at the household level, and enable analyses that would otherwise not be possible without collecting additional data. At the same time, it relies on assumptions that cannot always be directly tested, and the resulting estimates may be sensitive to the choice of matching strategy, common variables, and reference dataset.
Drawing on the experience gained so far, the presentation will discuss how the matched data can be validated, the main trade-offs encountered in the integration process, and the lessons learned from the collaboration. It will also illustrate the potential uses of the resulting integrated microdata for analysing household saving, economic vulnerability and multidimensional poverty, as well as for broader distributional analyses of household economic well-being.