Long-Term Care Funding: Self-Insure, Traditional, or Hybrid, and Which Dollars Should Pay

The Risk, Sized Honestly

Start with what the exposure actually is, because both the fear-based version and the dismissive version are wrong.

The CareScout Cost of Care Survey, the industry’s standard benchmark, puts 2025 national medians at roughly $35 per hour for non-medical home care, which works out to about $80,080 per year at 44 hours a week, around $74,400 per year for an assisted living community, and about $129,575 per year for a private nursing home room. Texas runs cheaper than the national picture: assisted living medians in the state sit near $5,666 per month, a meaningful discount to the national figure, though costs in Austin, Dallas, and Houston metros run above the state median and memory care typically adds a premium on top of assisted living.

Now add time and inflation, because care costs compound and the exposure sits decades away for most people reading this. A $120,000 annual blended cost today becomes roughly $263,000 per year in twenty years at 4% care inflation, and a three-year care episode beginning then totals about $821,000. That is the shape of the tail risk: not a certainty, not a rounding error, and timed to arrive at the exact end of the plan, after sequence risk and longevity have already taken their turns.

Two facts temper the scare math. Most care needs are shorter and cheaper than the worst case; many people need months of help, not years of nursing care. And the risk is not symmetric across a couple: women live longer, do more of the unpaid caregiving for the first spouse who declines, and then face their own care need alone, which connects this topic directly to the survivor dynamics we covered in our widow’s penalty piece. When we model care risk, we are usually modeling it hardest for the second spouse.

One more fact, stated plainly because so many households plan around a mistaken assumption: Medicare does not cover long-term custodial care. It covers short skilled nursing stays after hospitalization and limited home health in narrow circumstances. The help-with-daily-living care that constitutes the actual risk is not a Medicare benefit at any income level. Medicaid does cover custodial care, but only after assets are substantially spent down, generally at facilities of its choosing, and the program’s fiscal direction is tightening: the 2025 budget law (OBBBA) made a series of Medicaid changes, including revisions to long-term care eligibility rules, and paused federal minimum staffing standards for nursing facilities into the next decade. For the households we work with, Medicaid is the backstop of last resort, not a plan, and its terms are set in Washington and Austin, not by you.

Path One: Self-Insurance, Done as a Decision

For households with substantial assets, self-insurance is genuinely viable. Viable is not the same as automatic, and it is definitely not the same as what most self-insuring households actually do, which is nothing.

Real self-insurance is a named commitment: this asset, or this slice of the portfolio, is the care reserve, sized against a modeled episode, invested appropriately for a liability that is decades away and then suddenly immediate, and mentally fenced so the plan does not quietly spend it twice, once as “our care fund” and once as “the kids’ inheritance.” The failure mode of informal self-insurance is exactly that double-counting: the same dollars assigned to legacy, late-life spending, and care simultaneously, which works right up until the care bill arrives.

Self-insurance has real advantages. There is no premium, no insurer to negotiate claims with, no use-it-or-lose-it resentment, and full flexibility over what kind of care gets bought, including the private-pay home care arrangements that insurance sometimes reimburses grudgingly. If care is never needed, the reserve passes to heirs intact.

It has one honest weakness: the reserve must exist in full from day one of the risk, and the risk’s worst version, an early, long, expensive dementia episode, is exactly the scenario where a portfolio-funded plan hurts most, both financially and because the healthy spouse is now managing the portfolio, the care, and the household alone. Whether your balance sheet can absorb an $800,000 tail event without breaking the surviving spouse’s plan is not a feeling. It is a number, and it comes out of the same multi-year projection we use for everything else.

Path Two: Traditional Long-Term Care Insurance

Traditional LTC insurance is a pure risk transfer: premiums in exchange for a pool of benefits that pays when you can no longer perform two of six activities of daily living or have a severe cognitive impairment, after an elimination period that commonly runs 90 days. Policies are defined by four dials: the monthly or daily benefit, the benefit period or pool size, the elimination period, and inflation protection, which is the dial that matters most for anyone buying in their 50s or 60s, since the claim is likely decades away.

The honest history: the traditional market earned its scars. Early generations of policies were underpriced, insurers won regulatory approval for repeated premium increases on existing policyholders, and many carriers left the market entirely. The product that survives is more conservatively priced, which means more expensive, and premium increases on in-force policies remain legally possible. Anyone evaluating traditional coverage should stress the plan against a future premium increase rather than assuming today’s price is a contract for life; it is not.

What traditional coverage still does better than anything else is leverage. Because there is no death benefit and no cash value, every premium dollar buys more care benefit than any alternative structure. For a household that wants maximum care protection per dollar and can tolerate use-it-or-lose-it economics plus premium-increase risk, traditional remains the efficient tool, and the tax treatment below sweetens it meaningfully for some buyers, especially business owners.

Path Three: The Hybrids

Hybrid policies, life insurance or annuities with long-term care riders under the tax rules that govern qualified care coverage, now dominate new sales, and the reason is behavioral as much as financial: they answer the objection that kills traditional sales, “what if I never use it.” A hybrid pays for care if care is needed, pays a death benefit to heirs if it is not, and typically locks premiums by contract, removing the increase risk that haunts traditional policies. Many are funded with a single premium or a short premium schedule, which suits households sitting on cash or an old underperforming life policy; existing life insurance or annuity value can often be repositioned into a hybrid through a tax-free exchange, turning a legacy asset into care protection without triggering tax.

The cost of all that comfort is leverage. The same dollars buy meaningfully less care benefit through a hybrid than through traditional coverage, because part of every dollar is buying the death benefit and the premium guarantee. Hybrids are, in effect, self-insurance with training wheels and a floor: less efficient than pure insurance if care happens, less efficient than pure investment if it does not, and valuable precisely because they behave acceptably in both futures. For households that would otherwise choose nothing, that is not a small virtue. Our job when we model one is simply to make the internal pricing visible, because the packaging is designed to feel free and it is not.

The Tax Mechanics That Change the Answer

This is the layer most funding discussions skip, and it moves real money.

Premiums. Tax-qualified long-term care premiums count as medical expenses up to age-based limits, per person, set annually by the IRS. For 2026: $500 at age 40 or under, $930 for 41 to 50, $1,860 for 51 to 60, $4,960 for 61 to 70, and $6,200 over 70. For most employees those amounts only help if total medical expenses clear the 7.5% of AGI floor on Schedule A, which they usually don’t. But two doors matter for our readers. Self-employed owners can generally deduct qualified LTC premiums up to the age-based caps through the self-employed health insurance deduction, no itemizing and no 7.5% floor required, and certain entity structures can do better still, which makes the years you own a business the cheapest years you will ever buy this coverage. And HSA dollars can reimburse qualified LTC premiums tax-free up to those same age-based limits, which converts the HSA strategy we’ve written about, fund it, invest it, don’t spend it, into a designated care-premium engine in retirement. One caveat before you count on any of this: these premium deductions apply to standalone tax-qualified policies, and most hybrid life-plus-LTC policies do not qualify, since the life insurance component is never a deductible medical expense. The tax-free benefit treatment below still applies to qualified hybrid riders; it is the premium deduction that mostly does not.

Benefits. Benefits from qualified policies arrive income-tax-free, with one wrinkle for per-diem style contracts: tax-free treatment is capped at the greater of actual care costs or $430 per day in 2026. Reimbursement-style policies simply pay care costs tax-free.

The care year itself. Here is the mechanic that changes late-life planning: qualified long-term care costs are themselves deductible medical expenses, and a serious care year produces a deduction so large that it can neutralize the tax on the retirement account withdrawals funding it. A household pulling $300,000 from a traditional IRA to fund a $300,000 care year clears the 7.5% floor at $22,500 and deducts roughly $277,500, wiping out most of the withdrawal’s taxable income. Pre-tax dollars, ordinarily the most tax-burdened dollars on the balance sheet, become nearly tax-free in exactly this scenario. Which is one concrete reason “convert everything to Roth” has never been our position: a care episode is the one common future in which traditional dollars outshine Roth dollars, and a projection that models a care year will often deliberately leave a pre-tax balance unconverted as the care-funding tranche. The HSA plays the same role one better, since qualified care costs come out of it tax-free with no floor at all.

Sequencing. Which account pays for care is a withdrawal-ordering question layered on the framework from our sequencing piece: HSA first for qualified costs, traditional dollars into the big-deduction years, Roth preserved for the survivor. The right order in a care year is nearly the reverse of the right order in an ordinary year, and knowing that in advance is worth real money.

How We Run the Decision

We treat long-term care as a scenario inside the multi-year projection, not a product conversation. The care shock gets modeled explicitly: an episode for each spouse, at ages and durations worth stress-testing, at local costs inflated to the years they would actually occur. Then the three funding paths compete inside the household’s real numbers: what self-insuring does to the survivor’s plan, what a traditional policy’s premium stream costs against the leverage it buys, what a hybrid’s guarantees are really priced at, and how the tax mechanics above tilt the financing. Sometimes the model says the balance sheet carries the risk comfortably and the honest move is to name the reserve and stop worrying. Sometimes it says a policy is cheap protection for the exact scenario that breaks the plan. Usually it says something in between, like insuring one spouse, or covering a base layer and self-insuring the tail.

We lean away from the two slogans that dominate this topic. “Insurance is a ripoff, self-insure” ignores that the un-modeled version of self-insurance is just hoping, and hoping is hardest on the surviving spouse. “Everyone needs LTC insurance” is what people say when they sell it. We don’t sell it, or anything else. The right answer falls out of your numbers, your health picture, your family’s caregiving reality, and what you want the last chapter to look like, which is, as always, where the money conversation was really pointed the whole time.

Common Questions

At what age should we look at long-term care coverage?

The evaluation window is generally the mid-50s through mid-60s. Earlier, and you pay premiums for decades against a distant risk; later, and pricing rises steeply while health underwriting starts disqualifying people. The right time to model the funding question, though, is whenever the retirement projection gets built, because the self-insure-versus-insure answer shapes how much of the portfolio is truly spendable.

Does Medicare or a Medicare supplement cover any of this?

No. Medicare covers short skilled nursing stays after a qualifying hospitalization and limited skilled home health, not ongoing help with bathing, dressing, eating, or supervision for cognitive decline. Medigap plans follow Medicare’s coverage rules, so they exclude custodial care too. The gap is the whole reason this article exists.

Can I pay long-term care insurance premiums from my HSA?

Yes, for tax-qualified policies, up to the age-based annual limits, which for 2026 run from $500 at age 40 or under to $6,200 over age 70, per person. Actual qualified long-term care costs are also HSA-eligible without those caps, which is a major reason we treat an invested HSA as one of the best care-funding assets a household can build.

What triggers benefits under a long-term care policy?

Tax-qualified policies pay when a licensed practitioner certifies you cannot perform at least two of six activities of daily living (bathing, dressing, eating, toileting, continence, transferring) for an expected 90 days or more, or that you have a severe cognitive impairment requiring supervision. Most policies then apply an elimination period, commonly 90 days, before benefits begin, which is effectively a deductible measured in time.

Is it too late if one of us already has health issues?

Sometimes for insurance, never for planning. Traditional and hybrid policies both underwrite health, and conditions like diagnosed cognitive decline are generally disqualifying. But couples are underwritten individually, so covering the healthy spouse often remains available and is frequently the more important policy anyway, and the self-insurance, tax-sequencing, and reserve-naming work in this article requires no underwriting at all.