Aging US housing stock → structural repair/remodel spend → does Home Depot (not the homebuilders) capture it via Pro + exclusive house brands?

A net-new chain from DAILY step-2 #10 consumer (2026-09-15). Distinct from nvr-asset-light-exemption-homebuilder-pair-trade (new construction) and from consumer-trade-down-to-private-label-manufacturer-treehouse (grocery PL; THS already delisted). Thin vertical; AI-infrastructure is already overweight.

The chain (one paragraph)

The median age of a US home stayed ~23 years from 1940 to 1980 (post-war construction kept adding new homes), then rose — 25 by 1990, 30 by 2000, 33 by 2010, and 42 years today. The installed base needs repair and remodel even when housing starts are weak. Home Depot's Pro + DIY stack, plus exclusive house brands at an estimated 15–25% of sales (Behr, Hampton Bay, Ryobi, Ridgid, Husky), is how that spend is captured. Lowe's is the competitor, not the structural exemption. (⚠ unverified gap: a Census/ACS median-age-of-stock print plus HD's disclosed Pro / house-brand mix in a 10-K — the podcast cites "Arvind" research and industry estimates, not a filing.) From 2026-09-15-acquired-the-home-depot.

Candidate tickers

• HD — the named retailer; Pro + exclusive brands. Why this one: Acquired's load-bearing claim is HD, not a grocer.
• LOW — the competitor that reopened in the late-1990s housing bubble; a pair/relative, not the primary.

What evidence would convert this to an active thesis

• A Census / ACS (or HUD AHS) print that the median age of US housing stock is still rising, with a dollar figure for repair/remodel vs new construction.
• HD 10-K / earnings: Pro mix and/or house-brand mix disclosed, and Pro growing when housing starts are not.
• A valuation pass vs LOW and vs the builders showing HD is not already priced for the aging-stock story.

What would kill it

• Repair/remodel tracks housing turnover / mortgage lock-in, not age-of-stock — then HD is a rates/turnover name, not a structural one.
• House-brand mix is already in the multiple, or exclusive brands lose share to Amazon / club.

Sources (initial)

• 2026-09-15-acquired-the-home-depot — Gilbert: median age ~23 years 1940–1980 then aging; house brands 15–25% vs Kirkland ~1/3; Pro running room. Rosenthal: US ownership + tax incentives vs China housing stock (large market, terrible for home improvement).
• 2026-09-15-event-day-gates-fomc-day-1-statement-tomorrow — grocery PL record restates the other consumer spine; this chain is the net-new one.

Update (2026-09-18) — August building-materials sales −0.2%. Not a conversion.

From 2026-09-18-bucket-consumer-transport-retail-sales-dat: Census August retail sales, building materials and garden equipment were the only major category down (−0.2%). That is a same-month caution on the spend leg, not a Census/ACS median-age-of-stock print and not HD Pro mix. Conversion bar unchanged. Stay hypothesis. Do not chase HD.

Update (2026-09-21) — Airbnb $250M housing accelerator. Not HD Pro mix.

From 2026-09-19-feed-construction-physics-reading-list-09-19-2026 (partial: true): Airbnb announced a $250 million “last-dollar financing” housing-accelerator for multifamily projects. Adjacent housing-supply color, not a Census/ACS median-age print and not HD Pro / house-brand mix. Conversion bar unchanged. Stay hypothesis. Do not chase HD.

Related

• home-depot
• nvr-asset-light-exemption-homebuilder-pair-trade
• consumer-trade-down-to-private-label-manufacturer-treehouse
• farm-income-squeeze-to-ag-equipment-downcycle

Update (2026-09-21) — Census/AHS + HD 10-K retrieved. Mix still undisclosed. Stay hypothesis.

Gage early-chain. August building-materials −0.2% and Airbnb $250M remain not conversion (already filed). This pass fetched age-of-stock primaries/secondaries and the HD FY2025 10-K. Pro mix and house-brand mix are still not in the filing. Do not invent them. Do not chase HD.

Age-of-stock (what was actually retrieved)

• Census primary (owned homes, older vintage): 2021 AHS — "the median age of all owned homes was 41 years." Owned only; not the 2023 all-stock print. (Census library story)
• HUD/Census AHS 2023, via MBA (secondary): MBA Chart of the Week (8 Oct 2024), citing HUD + Census AHS: "The median age of the 2023 housing stock was 44 years (versus 39 years a decade ago in the 2013 AHS data)." 145.3 million units. (MBA Newslink) ⚠ Not the AHS Table Creator itself — Table Creator did not render a number this pass. Treat 44 as MBA's restatement of AHS 2023, not as a Census HTML table I opened.
• ACS B25035 (median year structure built): data.census.gov table ACSDT5Y2023.B25035 timed out; Census API returned no series without a key. No ACS median-year-built number is filed from this pass. Do not invent 1980 or convert it to an age.

The podcast's "42 years today" sits between Census 2021 owned-homes 41 and MBA's AHS 2023 44. Direction (stock older than the post-war 23-year plateau) is not contradicted. Conversion still wants a Census/ACS (or HUD AHS Table Creator) print plus a repair/remodel vs new-construction dollar figure — not a retailer podcast.

HD FY2025 10-K — Pro named, mix not disclosed

Retrieved hd-20260201 (period ended 2026-02-01).

• Two customer groups: consumers (DIY/DIFM) and Pros. Pro initiatives, Pro Xtra, Pro Trade Credit, SRS (FY2024) and GMS via SRS (FY2025) — wallboard, ceilings, steel framing. No Pro % of sales. No house-brand % of sales.
• Proprietary / exclusive brands named (HDX, Husky, Hampton Bay, Home Decorators Collection, Glacier Bay, Vigoro, Everbilt, Lifeproof). "These proprietary products help differentiate us from other retailers and generally carry higher margins than national brand products." No 15–25% (or any) mix. That figure stays an industry estimate on 2026-09-15-acquired-the-home-depot, not a 10-K fact.

Conversion bar (still unmet)

• Census/ACS or HUD AHS table for median age / year-built of the full stock, plus repair/remodel $ vs starts — partial (story + MBA secondary; ACS table not retrieved).
• HD 10-K/10-Q Pro mix and/or house-brand mix, and Pro growing when starts are not — not in FY2025 10-K.
• Valuation pass vs LOW / builders — not run.

Dated checkpoints

• Next ACS 1-year / 5-year B25035 national median year built (need a retrieved table, not an API miss).
• 2025 AHS (biennial; 2023 is the last MBA-cited vintage) via Census Table Creator or HUD release.
• HD next 10-K / 10-Q if they ever split Pro % or proprietary-brand %.
• Census monthly building-materials retail sales — already a same-month caution, not this bar.

Stay hypothesis. Do not chase HD. Not a sleeve.