A nursing home private room costs $127,750 a year.
That's the national median for 2024, per the Genworth and CareScout Cost of Care Survey — not a worst-case estimate, not a premium facility in a high-cost city. The median.
Someone turning 65 today has almost a 70% chance of needing some type of long-term care services and supports before they die. Women who end up needing care use it for an average of 3.7 years. Men, 2.2 years.
Most people tracking their net worth have a target number. A portfolio goal. They know their savings rate and their projected trajectory, and they have a reasonable sense of when they'll hit "enough." What they rarely have is a floor — a defined, dollar-specific minimum that accounts for the worst-case expenses most likely to arrive before they die.
The portfolio number tells you when you can theoretically stop working. The floor tells you whether you can actually spend what's above it without flinching at every withdrawal.
The Number That Isn't a Floor
Retirement planning has converged on a single benchmark: the 4% rule. In his landmark 1994 paper, William Bengen analyzed every 30-year retirement window in the historical data and found that a portfolio could sustain "a first-year withdrawal of 4 percent, followed by inflation-adjusted withdrawals in subsequent years" without running out in any 30-year period he examined.
It was a powerful finding. It was not a floor.
The 4% rule tells you how much a diversified portfolio can sustain under normal distribution conditions. It says nothing about what happens if you need three years of full-time nursing care. Or if the market drops 40% in your first two years of retirement and you're forced to sell equities at the bottom. Or if your spouse dies and a Social Security income stream disappears permanently.
A withdrawal rate is a throughput calculation. A floor is a threat model.
Wade Pfau, in Safety-First Retirement Planning, distinguishes between a probability-based approach — using a diversified portfolio and hoping market returns cooperate — and a safety-first approach that begins by covering core expenses with something reliable. The floor is the safety-first starting point: the amount that must be secured before any surplus can be freely spent.
The floor isn't pessimism. It's precision. Once you know what the floor costs, everything above it stops being ambiguous. Every dollar in surplus is mathematically free, because the worst-case scenarios have already been priced.
Long-Term Care: The Variable Most Plans Omit
The single largest driver of retirement plan failure isn't market volatility or overspending. It's long-term care costs landing on a portfolio that never planned for them.
A 2025 study from the Morningstar Center for Retirement & Policy Studies found that 43% of baby boomers will incur long-term care costs, with the average total bill coming to $242,373. When those costs are included in retirement viability analysis, 41% of households that need care are projected to run out of money before they die — compared to 26% without long-term care costs factored in.
The failure rate is 58% higher — tracing back to a single variable most retirement plans never quantify.
Medicare provides almost no protection here. According to Medicare.gov, it "doesn't cover any type of long-term care, whether in a nursing home, assisted living community or at home." It covers short-term skilled nursing care after a qualifying hospital stay, for rehabilitative purposes. It does not cover the custodial care — help with bathing, dressing, eating, mobility — that constitutes most long-term care needs.
The 2024 Genworth and CareScout Cost of Care Survey puts the national median annual costs in concrete terms: assisted living runs $70,800; a semi-private nursing home room, $111,325; a private room, $127,750.
Three years in a nursing home, at median rates, totals roughly $383,000. Out of pocket.
The floor isn't your portfolio target. It's the specific amount you'd need if the worst-case scenarios all arrived at once. The distance between those two numbers is your actual financial safety margin.
There are three practical ways to fund the long-term care piece of the floor: build a dedicated cash or investment reserve, purchase a hybrid life/LTC or annuity/LTC policy that pays benefits only if needed, or carry traditional long-term care insurance. All three count. The method matters less than the decision to quantify the risk and lock something down for it.
The Healthcare Bill Before Long-Term Care
Long-term care is the catastrophic wildcard. But sitting beneath it is a more predictable cost: ordinary Medicare-covered healthcare over the full length of a 20-to-30-year retirement.
Fidelity Investments releases an annual estimate of this figure. The 2025 estimate found that a 65-year-old can expect to spend an average of $172,500 in health care and medical expenses throughout retirement — not including long-term care. For a couple, the estimate is $345,000.
This covers Medicare Part B premiums, Part D prescription drug costs, out-of-pocket deductibles, and services original Medicare excludes. It does not include long-term care — that number gets added separately.
The two together form the healthcare half of the floor. The Fidelity figure — $345,000 for a couple's non-LTC medical costs — is one anchor. A self-funded long-term care reserve is the second, and considerably larger, layer.
These numbers aren't designed to alarm. They're designed to exist on paper. An uncalculated risk doesn't disappear — it surfaces as anxiety every time you consider spending on something you want.
The Sequence-of-Returns Buffer
The third floor component has nothing to do with costs. It's about timing.
Sequence of returns risk — the danger that poor market performance in the early years of retirement permanently impairs a portfolio, even if markets eventually recover — is precisely what Bengen's 4% rule was calibrated to survive. His research identified the worst-case historical scenario as a hypothetical retiree who stopped working in 1966, just as the stagflation decade was beginning. That retiree, withdrawing inflation-adjusted amounts from a 50/50 portfolio, faced a maximum sustainable starting withdrawal of 4.15% — the SAFEMAX, as Bengen called it.
Retirees who began withdrawing in years with favorable early sequences, with the same portfolio allocation, had dramatically better outcomes — not because they were smarter, but because the order of returns went their way.
The mechanism is irreversible. When you sell portfolio assets to fund living expenses during a market downturn, those shares are gone. They don't participate in the recovery. A 30% portfolio loss in year one means you're funding withdrawals from a permanently smaller base. The math doesn't just pause during a bear market; it resets on worse terms.
A sequence buffer is typically one to two years of non-discretionary living expenses held in cash or short-term bonds — money available to spend during a market downturn without selling equities at the bottom. It doesn't improve your average return. It prevents you from locking in your worst return at the worst possible moment.
The buffer's dollar size depends on your non-discretionary spending level, but it's a meaningful floor component because it's not an investment. It's an insurance policy against bad luck compounding with bad timing.
Building the Number
The floor has four components. Put them together once, and the result is a specific number rather than a vague sense of security.
The healthcare reserve accounts for the predictable Medicare gap across a full retirement. Using Fidelity's estimate as a planning anchor, $150,000 to $200,000 per person represents a defensible target — money that will be spent, gradually, on premiums and out-of-pocket costs over 20-plus years.
The long-term care reserve is the largest variable. Given the 70% probability and the median nursing home cost of $127,750 per year, a self-insured reserve of $200,000 to $400,000 per person represents a realistic range for full self-insurance — though a hybrid LTC policy can reduce the dedicated reserve significantly while still covering the catastrophic tail.
The sequence buffer is one to two years of non-discretionary spending, liquid and separate from the investment portfolio. It's deployed during downturns and replenished during recoveries.
The survivor gap is the piece most plans miss. When one spouse dies, the household permanently loses the smaller of the two Social Security payments. Quantify that income reduction specific to your situation, then capitalize it in the floor — either as an annuity income stream or as additional portfolio cushion.
Add the four numbers. That sum is the floor. It belongs in every retirement plan as a named, explicit line item, not as a vague margin.
What the Floor Gives You
The most expensive thing about an undefined floor isn't the risk it leaves uncovered. It's the spending it makes impossible above it.
A saver who has built a $2 million portfolio but never quantified the healthcare, long-term care, sequence, and survivor risks will spend that portfolio like $1.5 million might not be enough — because the uncounted risks feel like they could cost anything. The anxiety isn't irrational. It's the natural response to an open variable.
Define the floor at, say, $700,000 — specific, documented, with a plan for each component — and the math changes. The remaining $1.3 million isn't ambiguous surplus. It's the freedom number. It's the Portugal trip and the grandchildren's college fund and the renovation you've been deferring. It belongs to you, not to an unnamed future emergency.
Wade Pfau's safety-first framework rests on this insight: a retiree who hasn't secured core expenses will always be spending with one eye on the uncovered risk. The floor is what changes that. Not the size of the portfolio — the specificity of the accounting.
One afternoon. Four numbers. What would you spend differently if you knew the floor was real?
Ready to see your full financial picture?
Try Pro free for 30 days. No bank login required. No credit card.
Create your free dashboard