Most retirement plans are built around a number. For many people, that number is roughly 85 — a reasonable midpoint between average life expectancy figures they have seen cited over the years. Here is the problem with that: planning for the average means planning to run out of money if you live longer than average. And statistically, many of you will. This is not a warning designed to frighten you. It is a data problem that needs to be corrected, because the numbers the financial industry commonly cites are the wrong numbers to plan with.
When people hear that life expectancy in the United States is roughly 78 or 79 years, they sometimes use that as a planning horizon. That figure is life expectancy at birth — which includes infant mortality, accidents in early life, and deaths in middle age. For someone who has already survived to 65, that number is not relevant to their planning. The relevant figure is life expectancy at age 65 — what the Social Security Administration's 2025 Trustees Report calculates as remaining life expectancy for someone who has already reached retirement age. According to those period life tables, a 65-year-old man can expect to live an additional 17.5 years, reaching roughly age 82 to 83. A 65-year-old woman can expect an additional 20.2 years, reaching roughly age 85. But even these figures understate the planning horizon, because they are midpoints. Half of all 65-year-olds will live longer than the average. Planning for the average means planning to run out of money precisely when you might need it most.
For a married couple, the relevant planning metric is not individual life expectancy — it is joint longevity. Joint longevity asks: what is the probability that at least one of us will still be alive at a given age? For a 65-year-old couple today, there is approximately a 47 percent probability that at least one spouse will live to age 90. There is roughly a 20 percent probability that at least one will survive past age 95. These are not outliers — they represent the planning reality for nearly half of all couples entering retirement today. This matters for three financial reasons. First, the survivor typically loses one Social Security check when the first spouse dies. Second, the surviving spouse often faces higher per-person living costs than the couple did together. Third, long-term care needs tend to be highest in the final years of life, which — for a couple — may be a very long time away from today.
A 30-year retirement is mathematically different from a 20-year retirement in ways that are easy to underestimate. Inflation is the first issue. At 3 percent annual inflation, the purchasing power of a dollar is cut roughly in half over 25 years. A fixed income stream that felt adequate at 65 may cover substantially less by age 85. This is why Social Security's annual cost-of-living adjustment — tied to the Consumer Price Index — is more financially significant than most beneficiaries realize. It is one of the only retirement income sources that automatically adjusts for inflation. Sequence of returns risk extends across a longer window. The standard recommendation to hold more bonds and less equity as you age may need recalibration for a 35-year horizon — a portfolio that becomes too conservative too early may not generate enough growth to sustain three-plus decades of withdrawals. Healthcare cost escalation is back-loaded. The annual inflation rate for medical care has exceeded general inflation for decades. And healthcare costs are not evenly distributed across retirement — they tend to accelerate sharply in the final years of life, precisely when portfolio assets may be most depleted.
Traditional retirement planning uses average life expectancy as a planning horizon. Financial gerontologists and actuaries increasingly argue that planning for the 90th or 95th percentile of longevity is the only way to genuinely mitigate longevity risk. Longevity risk is the risk of outliving financial assets. It is a tail risk, meaning it affects a minority of the population — but that minority is substantial, and the consequences of underestimating it are severe. A couple who planned to age 85 and finds themselves at 92 with a depleted portfolio faces few good options. Social Security delay is one of the most direct tools for managing this risk at the individual level. Every year of delay past full retirement age adds approximately 8 percent to the monthly benefit, up to age 70. For a person who lives to 90, that additional monthly income — inflated annually by the cost-of-living adjustment — compounds into a significant difference in cumulative lifetime benefits. Age 85 ~55% (women), ~42% ~73% (men) Age 90 ~34% (women), ~22% ~47% (men) Age 95 ~15% (women), ~8% ~20% (men) Source: SSA 2025 OASDI Trustees Report period life tables.