The Under-Reporting of Transfers in Household Surveys: Its Nature and Consequences
Working Paper 15181
DOI 10.3386/w15181
Issue Date
High rates of understatement are found for many government transfer programs and in many datasets. This understatement has major implications for our understanding of economic well-being and the effects of transfer programs. We provide estimates of the extent of under-reporting for ten transfer programs in five major nationally representative surveys by comparing reported weighted totals for these programs with totals obtained from government agencies. We also examine imputation procedures and rates. We find increasing under-reporting and imputation over time and sharp differences across programs and surveys. We explore reasons for under-reporting and how under-reporting biases existing studies and suggest corrections.