Older households face health-related risks, including risk of being in need of long-term care and mortality risk. How these risks affect financial portfolio choice of households depends on household preferences for long-term care and bequest. Using linked survey-administrative data on clients of a mutual fund company, this paper finds that the desire to have enough resources for long-term care and bequests are overall strong but also heterogeneous across households. The estimated relationship between actual stock share of households and the strength of these preferences is qualitatively similar but quantitatively much weaker compared to the predictions from the life-cycle model with the estimated preference heterogeneity. Based on the predictions from the model, this paper discusses what financial instruments would better meet the needs of households.
This paper investigates the relationship between stock share and expectations and risk preferences using linked survey responses and administrative records from account holders. The survey allows individual-level, quantitative estimates of risk tolerance and of the perceived mean and variance of stock returns. Estimated risk tolerance, expected return, and perceived risk have economically and statistically significant explanatory power for the distribution of stock shares. Relative to each other, the magnitudes are in proportion with the predictions of benchmark theories, but they are substantially attenuated. MBA graduates have more stable beliefs, more knowledge about their account holdings, and less attenuation.
This paper introduces the Vanguard Research Initiative (VRI), a new panel survey of wealthholders designed to yield high-quality measurements of a large sample of older Americans who arrive at retirement with significant financial assets. The VRI links survey data with a variety of administrative data from Vanguard. The survey features an account-by-account approach to asset measurement and a real-time feedback and correction mechanism that are shown to be highly successful in eliciting accurate measures of wealth. Specifically, the VRI data reflect unbiased and precise estimates of wealth when compared to administrative account data. The VRI sample has characteristics similar to populations meeting analogous wealth and Internet access eligibility conditions in the Health and Retirement Study (HRS) and Survey of Consumer Finances (SCF). To illustrate the value of the VRI, the paper shows that the relationship between wealth and expected retirement date is very different in the VRI than in the HRS and SCF—mainly because those surveys have so few observations where wealth levels are high enough to finance substantial consumption during retirement.
Older Americans, even those who are long retired, have strong willingness to work, especially in jobs with flexible schedules. For many, labor force participation near or after normal retirement age is limited more by a lack of acceptable job opportunities or low expectations about finding them than by unwillingness to work longer. This paper establishes these findings using an approach to identification based on strategic survey questions (SSQs) purpose-designed to complement behavioral data. These findings suggest that demand-side factors are important in explaining late-in-life labor market behavior and may be the most appropriate target for policy aimed at promoting working longer.