Skip to content
STI
← Back to Blog

If Your Retirement Portfolio Drops 30%, Does Aging in Place Still Beat Assisted Living? The September 2026 NPV Math

The $300 bonus and the $460,000 decision

Every September, NerdWallet publishes a good faith reminder that free coffee exists on National Coffee Day and that switching banks for a $300 signup bonus might be worth the paperwork. People spend real hours comparing Klatch Coffee's free-cup terms against Dunkin's, or reading the fine print on a Chase checking bonus, because it feels like "found money" and the math is simple enough to hold in your head.

Then the same people back into a $400,000+ decision about aging in place versus assisted living, memory care, or a nursing home — for themselves or a parent — using a gut feeling and whatever their cousin did. The dollar amounts are 1,000x bigger and the math is harder, so somehow it gets less rigor, not more.

This post is about closing that gap for one specific, underexamined variable: what happens to your aging-in-place vs. facility-care math when the market does something like Mr. Money Mustache's September 2026 piece worried about — a serious AI-bubble-style correction — right in the middle of your care years. It changes the two paths differently, and most people never model that asymmetry.

The baseline: a worked example (your numbers will differ)

Here's a hypothetical family — call them a composite, not a real case — to make the math concrete.

The person: 78 years old, currently at 1 ADL (activities of daily living) loss, home is paid off, $850,000 in a retirement portfolio, $2,200/month Social Security, needs 20 hours/week of in-home care today.

The care needs escalation curve: Based on typical ADL decline patterns, this person's care hours climb roughly like this over 5 years — 20 hrs/week (Y0) → 30 (Y1) → 35 (Y2) → 45 (Y3, crossing into 3 ADL losses) → 60 (Y4, essentially round-the-clock).

September 2026 wage input: the Bureau of Labor Statistics' August 2026 report showed average hourly earnings up $0.10 for the month and 4.1% unemployment — a tight labor market that's kept home health aide wages growing faster than headline wages. We'll use $34/hour for a home health aide, rising roughly 3%/year, consistent with the wage pressure covered in At $33/Hour for Home Health Aides in 2026, Here's Exactly When Aging-in-Place Costs More Than Assisted Living.

Facility comparison set: assisted living at $73,800/year, memory care at $94,800/year, nursing home (semi-private) at $116,400/year — all 2026 figures, inflating at 3%/year.

One-time home modification cost: $45,000 (stairlift, bathroom remodel, grab bars, ramp) — needed in Year 0 for aging in place, not needed for a facility move.

Running the NPV, year by year

Using a 4% real discount rate (reflecting a blended conservative return assumption, not a crash scenario yet):

YearADL lossesCare hrs/wkAging-in-place annual costFacility annual cost
0120$98,360 (incl. $45k modification)$73,800 (assisted living)
11–230$73,140$76,014
2235$84,616$78,294
3345$106,249$103,600 (memory care)
43+60$138,819$131,000 (nursing home)

Discounted to present value at 4%: aging in place comes out to roughly $460,000 over 5 years. The facility path comes out to roughly $423,000. That's a $37,000 NPV gap favoring facility care — the reverse of the popular assumption that aging in place is always cheaper. It flips because the escalation curve gets steep: by Year 4, 60 hours/week of paid aide time at $38/hour costs more than a nursing home bed.

This is the exact dynamic covered in more detail in Aging in Place vs Assisted Living vs Memory Care vs Nursing Home: The NPV Gap Ranges From -$116,000 to +$298,000 Over 10 Years — the ADL decline rate is the single biggest lever on which side of zero you land on. Slower decline, aging in place usually wins. Faster decline, it usually doesn't, because you're paying full custodial-care wages for hours that a facility bundles more cheaply through shared staffing.

This is the kind of analysis Dorevanti runs for you — so you don't have to build the spreadsheet yourself, rebuild it every time BLS releases new wage data, and re-run it every time your parent's ADL status changes.

Now overlay the market crash

Here's the part almost nobody models: it's not just what the care costs — it's how it gets funded, and funding source risk hits the two paths asymmetrically.

Mr. Money Mustache's September piece on the AI bubble made a point worth sitting with: it's not the size of a market drop that wrecks a retirement plan, it's when it happens relative to your withdrawals. A 30% correction in year one of a drawdown period does permanent damage that the same 30% correction in year ten wouldn't, because you're pulling money out of a smaller base right when it needs to be growing back. This is sequence-of-returns risk, and it applies directly to care funding.

In our example, guaranteed income is $2,200/month Social Security ($26,400/yr) plus a VA Aid & Attendance benefit — assume the person is an eligible veteran, at roughly $2,795/month ($33,540/yr) in 2026. That's $59,940/year in income that doesn't touch the market at all. Everything above that has to come out of the $850,000 portfolio.

Aging-in-place funding gap by year: roughly $38,000 (Y0) → $13,000 (Y1) → $25,000 (Y2) → $46,000 (Y3) → $79,000 (Y4). That's an average draw approaching $40,000/year, climbing sharply.

Now: what if the AI-bubble scenario MMM describes actually happens, and the portfolio drops 30% in Year 1 — from $850,000 down to roughly $595,000 — right while you're also pulling out $13,000–$79,000/year for escalating in-home care? You're now withdrawing from a permanently smaller base during the years the market needs to recover. Even if the market fully bounces back by Year 4, your portfolio doesn't, because the withdrawals during the crash locked in the loss.

Run that same shock against the facility path, and the exposure looks different. Assisted living and memory care draw on the same portfolio in the early years — so the sequence risk is similar at first. But by Year 4, when the nursing-home stage is reached, there's a backstop that aging in place doesn't have: Medicaid. Once countable assets fall below the state limit (commonly $2,000 for an individual, with a community-spouse resource allowance around $130,380 in many states in 2026), Medicaid picks up nursing home costs. There is no equivalent universal backstop for paying a home health aide 60 hours a week — Medicaid's Home and Community-Based Services (HCBS) waivers exist, but they're state-specific, often capped, and frequently have waitlists.

That asymmetry is the real insight: a market crash is more dangerous to an aging-in-place plan than to a facility-care plan, because the facility path (specifically nursing home) has a means-tested floor that in-home custodial care generally doesn't. If you're funding either path heavily from an equity portfolio and you're worried about a correction like the one MMM's piece discusses, that's a reason to model your Medicaid spend-down timeline before the crash happens, not after your aide hours have already escalated to 60/week. The Aging in Place vs Nursing Home crossover analysis walks through exactly where that floor sits at different ADL loss levels.

VA Aid & Attendance doesn't change the crossover — it changes the cushion

Worth being precise here: the $33,540/year VA Aid & Attendance benefit in this example reduces the net cost of both paths roughly equally, since it's a fixed monthly stipend independent of where the care happens. It doesn't shift the $37,000 NPV crossover point much — it just makes both numbers smaller and, more importantly, makes both numbers less exposed to the market, because it's guaranteed income, not portfolio drawdown. If you're eligible and not currently claiming it, that's the single highest-leverage move available before you touch the investment account at all — see the detailed stacking math in Aging in Place vs Assisted Living: The $59,109 NPV Gap VA Aid & Attendance Can't Close for why it closes part of the gap but rarely all of it.

Life expectancy is the multiplier on everything above

Every number above assumes a 5-year horizon. If life expectancy is closer to 3 years, the $45,000 home modification cost never fully amortizes — you're paying for a stairlift you'll use for 36 months, which tilts the math toward facility care or a shorter-term rental modification. If life expectancy is closer to 8-10 years, the early aging-in-place years (before ADL losses escalate) look comparatively cheap for longer, but you're also exposed to market volatility for a longer stretch, raising the odds that a downturn lands during your drawdown window at some point. Neither horizon is right or wrong — it's just a variable that has to be plugged in with an honest, individualized estimate, not a national average.

What to actually do with this

None of this says "aging in place is bad" or "facility care is safer" as a blanket rule — the math above flips in either direction depending on your specific ADL decline speed, your portfolio size and allocation, your VA and Medicaid eligibility, and your life expectancy assumption. What it does say is that a market-risk conversation belongs inside your care-cost model, not next to it. If most of your funding sits in equities, you're carrying a second layer of risk that a pure "cost per year" comparison misses entirely.

You can model this for your specific situation — including a market-shock stress test on your own portfolio, ADL curve, and benefit eligibility — at Dorevanti, rather than building five years of spreadsheet scenarios by hand every time the Fed, the BLS, or the stock market moves.

Sources

Ready to compare care options?

Compare Care Options Free