Evaluating Micro-Survey Estimates of Wealth and Savings
This proposal extends prior work on integrating and comparing the wealth and saving data from three household surveys. It will extend the analysis of the active saving concept of the Panel Study of Income Dynamics (PSID) and develop statistical models based on the panel data of the Health and Retirement Study (HRS) and the PSID in order to correct estimates of saving and wealth change in the larger Survey of Consumer Finances (SCF). Repeated cross-section surveys can introduce a bias into the age profile of wealth, because persons with lower wealth are known to die at a younger age. This is believed to be a significant problem with using cross-sectional data to evaluate wealth decumulation in the postretirement years. Similarly, the active saving portion of the PSID can be used to identify and model the capital-gain component of wealth accumulation. Both of these adjustments should make it possible to construct estimates of saving from successive waves of the HRS and synthetic cohorts for the SCF surveys. The result will be an evaluation of the pattern of saving across socio-economic groups based on three independent but comparable micro surveys.