Working Paper: NBER ID: w28828
Authors: Mariacristina De Nardi; Eric French; John Bailey Jones; Rory McGee
Abstract: We estimate a model of savings for retired couples and singles who face longevity and medical expense risks, and in which couples can leave bequests both when the first and last spouse dies. We show that saving motives vary by marital status, permanent income, and age. We find that most households save more for medical expenses than for bequests, but that richer households and couples, who hold most of the wealth, save more for bequests. As a result, bequest motives are a key determinant of aggregate retirement wealth.
Keywords: Retirement Savings; Household Heterogeneity; Bequests; Medical Expenses
JEL Codes: D1; D12; D15; E21
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
Cause | Effect |
---|---|
when the first spouse dies (D14) | net worth drops sharply by an average of $160,000 (G51) |
eliminating medical expenses (H51) | mean wealth drop of 31% (D31) |
removing bequest motives (D15) | mean wealth drop of 148% (D31) |
eliminating both medical expenses and bequest motives (D14) | combined wealth drop of 425% (D31) |
surviving households continue to deplete their wealth at a faster rate than when married (D14) | wealth depletes faster (E21) |
richer households and couples save more for bequests (D14) | aggregate retirement wealth increases (G51) |
44% of aggregate bequests are voluntary (D64) | importance of bequest motives at the aggregate level (D15) |
saving motives for retired couples and singles differ significantly (D14) | bequest motives are a key determinant of aggregate retirement wealth (D14) |