Why a care projection is not just today's price
Long-term care is the largest uninsured liability most retirement plans carry, and it is the one people size wrongly in the most predictable way: by looking up the current daily rate for a nursing home in their state and multiplying by a number of years. That answer is right only if care begins today.
Two compounding effects sit between now and the claim. The first is the years before care starts, over which the daily rate keeps rising. The second is the episode itself, during which the rate keeps rising again — a three-year stay does not cost three times the first year's price, it costs slightly more, and over a long stay the difference is substantial.
The rate at which care prices rise is the assumption that dominates the whole projection. Care is a labour business: the cost of a nursing home bed is mostly the cost of the people staffing it around the clock, so it tracks caregiver wages in a chronically short labour market rather than the basket of goods in the consumer price index. Planners therefore use a care-specific inflation rate above general inflation. This calculator makes it an input rather than picking one for you, because the honest answer depends on the setting and the region — home care and facility care do not inflate at the same pace, and neither is uniform across states.
What comes out is not a prediction. It is a sized liability: the number your plan has to be able to absorb, expressed in the dollars of the years the care is actually delivered. That is the figure to hold against insurance, savings and home equity.
The three pieces of arithmetic
Compound the daily rate to the claim date. dT = d0 × (1 + g)T, where T is the years between now and the start of care. This is ordinary compounding, and it is the step people skip. At 4.5% a year, twenty-two years multiplies the rate by 2.63.
Sum the episode. Year one costs dT × D, where D is days of care per year. Year two costs that figure times (1 + g), and so on. A partial final year is prorated. The total is the sum of those yearly figures, each in the dollars of its own year — no discounting, because these are amounts that will actually be paid.
Discount back to the claim date to find what you need. This is the step that separates a sensible answer from an alarming one. You do not need the whole episode total in hand on the first day of care, because the money you have not spent yet keeps earning. The sum genuinely required at the start of care is each year's cost discounted at the return on the assets: R = Σ ck ÷ (1 + r)k. Compare R with what the earmarked assets will have grown to, and the difference is the funding gap.
Notice what happens when the care inflation rate and the asset return are equal. Each year's cost grows by exactly the factor the discounting removes, so every term in the sum equals the first year's cost, and the required amount is simply the first year's cost multiplied by the number of years. That is a useful sanity check, and it also explains the structure of the whole problem: what matters is not the level of either rate but the gap between them.
Worked example: planning at 60 for care at 80
You are 60. You plan around care starting at 80 and lasting three years, in a facility charging $300 a day today. You assume 5% care inflation, you have earmarked $250,000, and you expect 5% on it.
- Years of compounding. 80 − 60 = 20 years.
- Inflation factor. 1.0520 = 2.65330.
- Daily cost when care begins. $300 × 2.65330 = $795.99 a day.
- First year of care. $795.99 × 365 = $290,536.10. (Equivalently, $300 × 365 = $109,500 today, × 2.65330.)
- Second and third years. $290,536.10 × 1.05 = $305,062.91, and × 1.05 again = $320,316.05.
- Total episode cost. $290,536.10 + $305,062.91 + $320,316.05 = $915,915.06.
- Sum required at the start of care. Because the 5% return exactly matches the 5% care inflation, each year discounts back to $290,536.10, so R = 3 × $290,536.10 = $871,608.30.
- Projected assets at 80. $250,000 × 1.0520 = $250,000 × 2.65330 = $663,324.43.
- Funding gap. $871,608.30 − $663,324.43 = $208,283.87.
- Monthly saving that closes it. Over 240 months at 5% a year, a monthly deposit accumulates 411.03 times itself, so $208,283.87 ÷ 411.03 = $506.74 a month.
Three observations. The episode costs $915,915 but you need $871,608 on day one, and the $44,307 difference is the return earned on money you have not spent yet. Nearly a million dollars for three years is not an error — it is $300 a day compounded for twenty years. And the whole gap closes for about $507 a month, which is a far more tractable number than the headline and is the one worth acting on.
How to read the gap, and what actually moves it
Test the duration before you test anything else. Duration is the input the total is most sensitive to, and it is the one people are most confident about for no good reason. A three-year assumption and a six-year assumption differ by more than doubling, because the later years are the most expensive ones. If your plan only survives at three years, it is not funded.
Then test the setting. Home care with a five-day-a-week aide, assisted living, and a private nursing-home room differ by a large multiple, and most episodes move through more than one of them — often starting at home and ending in a facility. Running the calculator twice, once for each phase, gives a more honest total than picking a single average setting.
The gap between care inflation and the return on the assets is the third lever, and it works in a direction people find counter-intuitive. If care inflates faster than the fund grows, waiting makes the problem worse in real terms, and the calculator says so. If the fund grows faster, waiting helps — but only while the money can stay invested. As care becomes imminent the assets have to move into something liquid and conservative, which is exactly when the assumed return should fall.
Finally, distinguish the two shortfall numbers the page gives you. The funding gap is a single sum needed on the first day of care. The uncovered column in the table is the cash that actually runs out in each later year, in the dollars of that year. They answer different questions and they are not interchangeable: the first sizes a policy or a savings goal, the second tells a family when the money runs out.
If the answer is that insurance should carry part of this, size the benefit with the long-term care inflation rider calculator, which shows how far a policy's daily benefit will have grown by the time you claim. If you are still working, check whether a disability rather than a care episode is the nearer risk with the disability insurance benefit calculator.
How much a daily rate multiplies before care begins
| Years until care | 3% | 4% | 5% | 6% |
|---|---|---|---|---|
| 5 | 1.159 | 1.217 | 1.276 | 1.338 |
| 10 | 1.344 | 1.480 | 1.629 | 1.791 |
| 15 | 1.558 | 1.801 | 2.079 | 2.397 |
| 20 | 1.806 | 2.191 | 2.653 | 3.207 |
| 25 | 2.094 | 2.666 | 3.386 | 4.292 |
| 30 | 2.427 | 3.243 | 4.322 | 5.743 |
The spread across the columns is the reason the inflation assumption matters more than the starting rate. Over thirty years, choosing 6% rather than 3% multiplies the projected daily cost by 2.37 (5.743 ÷ 2.427).
What this projection does not include
- Medicare does not pay for custodial care. Medicare covers a limited period in a skilled nursing facility after a qualifying hospital stay, and it covers skilled home health under conditions. It does not pay for help with bathing, dressing, eating or supervision, which is what almost every long care episode consists of. Assuming otherwise is the single most expensive planning error in this field.
- Medicaid is means-tested and comes with a look-back. It is the largest payer of nursing home care in the United States, but eligibility requires spending down assets, transfers are reviewed over a look-back period, and it constrains which facilities will accept you. It is a floor, not a plan.
- Unpaid family care is not costed. Most care begins with a spouse or an adult child providing it free. That has a real cost in foregone earnings and health, and it usually ends before the episode does.
- Housing costs do not simply disappear. A move into a facility may free up a home, but a couple with one spouse in care still runs a household. Do not net the whole daily rate against current living expenses.
- Tax on the withdrawals is ignored. If the earmarked assets sit in a traditional IRA or 401(k), the sums here are pre-tax and the real requirement is higher. Qualified long-term care expenses may be deductible as medical expenses above the AGI threshold, which offsets part of it.
- One setting at one rate. The model runs a single daily rate for the whole episode. Real episodes escalate through settings, and the escalation usually goes one way.
Get the daily rate from your own market, not a national figure
Care pricing varies more by geography than almost any other consumer service, and a national median is a poor proxy for what you will actually pay. Call three facilities or agencies in the area where care would realistically be delivered — which is often near an adult child rather than where you live now — and ask for the current daily or monthly private-pay rate, the annual increase they have applied for the last three years, and what is excluded from the base rate. Medication management, incontinence care and higher levels of assistance are frequently billed on top of the advertised rate. The Genworth Cost of Care Survey publishes medians by state and setting if you want a starting point, but a local quote beats it.
The four ways a gap of this size gets closed
Self-funding. Viable when the projected gap is small relative to the portfolio, and it keeps the money available for anything else. The risk is asymmetric: an episode at the long end of the distribution can consume assets that were also meant to support a surviving spouse.
Traditional long-term care insurance. Buys a daily or monthly benefit pool for a premium that is not guaranteed — carriers have obtained substantial rate increases on older blocks. What matters is the inflation rider, because a policy bought at 60 and claimed at 82 has to have grown with the cost of care to be worth anything. Size that growth before you size the premium.
Hybrid life or annuity policies with a care rider. These pay a death benefit if care is never needed, which removes the use-it-or-lose-it objection, and the premium is normally guaranteed. They cost more per dollar of care benefit in exchange for that certainty.
Home equity. The largest asset in most households over 70 and the one most often left out of the plan. Selling or borrowing against it converts a fixed asset into care funding at exactly the point when the house is no longer being lived in.
Whichever route you take, size it against the funding gap rather than the headline episode cost, and re-run the projection every few years. The two inputs that drift are the local daily rate, which you can observe, and the age you are planning around, which moves closer every year and shortens the compounding period on both sides of the equation. If Medicare surcharges and retirement health costs are also part of the picture, the retirement healthcare cost calculator and the Medicare IRMAA calculator cover the parts this one deliberately leaves out.