Working Paper: NBER ID: w15181
Authors: Bruce D. Meyer; Wallace K.C. Mok; James X. Sullivan
Abstract: 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.
Keywords: Underreporting; Transfers; Household Surveys; Economic Wellbeing
JEL Codes: C42; H53; H55
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| underreporting (C24) | overstatement of income distribution dispersion (D31) |
| underreporting (C24) | bias in estimates of program take-up rates (C83) |
| underreporting (C24) | incorrect conclusions about impact on poverty and income distribution (F61) |
| imputation procedures (C26) | inflation of reported numbers (E31) |
| underreporting (C24) | understatement of effects of income transfer programs (H53) |