This dissertation contains two essays that examine the role of bequest motives in some of the
most important financial decisions people make during retirement. In the first essay, I study
the “annuity puzzle”: very few retirees buy life annuities even though standard economic life
cycle models indicate that the benefits from doing so are substantial. I find that the bequest
motive is a crucial determinant of the gain from annuities. Whereas people without bequest
motives typically would be much better off buying available annuities, people with modest
bequest motives are likely to gain little from available annuities and in many cases are better
off not annuitizing any wealth at available rates. Bequest motives make self-insurance—and
the bequests that often accompany it—much more attractive. My simulations show that
bequest motives consistent with other facts significantly reduce the number of people who
would otherwise buy annuities.
In the second essay, I study two other major financial decisions faced by retirees: whether
to buy long-term care insurance and how much wealth to save into old age. Few retirees
buy long-term care insurance (less than 10% of retirees in the U.S.), and most retirees hold
much of their wealth into old age. I show that these patterns are consistent with a life cycle
model with a strong bequest motive and inconsistent with models without bequest motives.
As with annuities, people with bequest motives are much more likely than people without
bequest motives to prefer self-insuring their risks in old age because they value the large
bequests that often accompany this strategy.
Together, these essays indicate that some of the most important financial decisions that
people make during retirement are inconsistent with models in which retirees’ exclusive
goal is to secure the highest, most stable level of consumption for themselves. Instead,
retirees’ choices indicate that in addition to valuing their own consumption, they also value
the prospect of leaving wealth to their heirs. By testing alternative theories against a wide
range of facts, this dissertation uncovers strong evidence that bequest motives have important
vii
effects on behavior, including among people who may ultimately leave no bequest.