the annuity puzzle

Here is a genuine mystery in economics. Standard theory says a retiree who does not know how long they will live should usually convert a big chunk of savings into a lifetime annuity — it is the cleanest way to avoid both running out of money and dying with a hoard you never enjoyed. Yet in the real world, very few people voluntarily buy annuities. This gap between what theory recommends and what people actually do is called the annuity puzzle.

The classic result (Yaari, 1965) showed that under simple assumptions a rational retiree should annuitize nearly all of their wealth, because annuities pay a 'mortality credit' that no ordinary investment can match. The puzzle is why real demand is so low. Researchers point to many partial explanations: people want to leave a bequest; they fear losing control and liquidity (the lump sum is gone); they already have annuity-like income from state pensions; they distrust insurers or fear inflation eroding fixed payments; they misjudge their own life expectancy; and behaviourally, handing over a large lump sum 'feels' like a loss, while spending from a visible balance feels safer.

For actuaries and policymakers the annuity puzzle matters enormously, because the shift from defined-benefit pensions (which annuitize automatically) to defined-contribution savings (which do not) means more retirees must choose to annuitize — and most do not. This drives product design (partial annuitization, deferred 'longevity' annuities, default annuitization in pension plans) and public policy. The honest caveat: 'puzzle' overstates it — many reasons not to fully annuitize are perfectly rational, so the real question is why so few people annuitize even partially.