NAHB's immigration-enforcement claim checks out on its own numbers — but 3 of 4 figures are the trade group's own research, and it cites its lowest housing-shortage estimate without saying so

Summary

NAHB's Sept. 25, 2026 post is fact-checked as largely accurate in its own arithmetic but built almost entirely on the trade association's own research and its most favorable-to-its-argument figures, presented without that framing. The BLS JOLTS construction job-openings figure (~300,000) checks out exactly against AskAmerica's own live warehouse query (326,000 in July 2026). The Home Builders Institute/University of Denver study figures (1.98 months, 19,000 homes, $10.806B rounded to "$10 billion") are transcribed accurately from the primary press release. The "more than one-quarter" (26.3%) immigrant-workforce-share figure is corroborated by independent trade-press coverage of NAHB's own Census ACS analysis, though it could not be independently recomputed in this warehouse (no industry-by-nativity crosstab is loaded here). The one figure that most needs context and doesn't get it is the "housing shortage of about 1.2 million units" — this is NAHB's own estimate, using a narrower, self-described "lower-bound" methodology, and it sits at the low end of a range that runs to 3.7-4.5 million from Freddie Mac, Zillow, and Up For Growth. Overall grade: 2 Pinocchios on claims accuracy (significant one-sided framing, no invented numbers) and 1 Pinocchio on fidelity (NAHB represents its own and HBI's sources accurately).

Claim-by-claim verdicts

1 true · 1 mostly true · 1 partially true · 1 not checkable here · 4 assertions checked

Verdicts: true, mostly true, partially true, mostly false, false, not checkable here (no table carries the measure), stale vintage (the article cites a release this corpus has not loaded — a freshness gap, not a falsehood).

Claim detail

SELECT "year", "date", metric_type, value FROM econ.jolts_industry WHERE industry_code='230000' AND metric_type IN ('JOL','JOR') AND "year">=2025 ORDER BY "year" DESC, "date" DESC FETCH FIRST 30 ROWS ONLY

SELECT SUM(total_population) AS total_pop, SUM(foreign_born) AS total_foreign_born FROM census.acs_nativity WHERE "year"='2023' AND geography='state'

Claim 1 — "A housing shortage of about 1.2 million units"

Verdict: Partially true. This is NAHB's own figure from its Feb. 2026 Housing Outlook. NAHB's own published methodology (Eye on Housing, Feb 2026) describes it as a narrow, self-described "lower-bound" estimate: the additional vacant units needed to restore 2024 metro vacancy rates to long-run equilibrium, explicitly excluding pent-up household-formation demand and the need to replace obsolete housing stock. Independent estimates from the same period run substantially higher: Freddie Mac ~3.7 million (Q3 2024), Up For Growth ~3.78 million (2025 report), Fannie Mae ~4.4 million, Zillow ~4.5 million. Only Zonda's ~1.0 million estimate is lower than NAHB's. The claim is not fabricated — it is a real, methodologically documented NAHB estimate — but presenting it as "the" housing shortage, rather than the lowest of several major published estimates and NAHB's own stated floor, is a materially misleading framing for a reader who doesn't already know the range.

Claim 2 — "Government data indicate construction is short about 300,000 workers"

Verdict: Mostly true. AskAmerica's own live warehouse query against econ.jolts_industry (BLS JOLTS, series JTS230000...JOL, industry_code 230000 = Construction) confirms construction job openings of 326,000 in July 2026 (298,000 in June, 291,000 in May) — independently verifying the question's own hypothesis and NAHB's rounded "~300,000" figure almost exactly. This is genuine, correctly-cited BLS government data, verified directly by re-querying the primary source series in this session (not taken on NAHB's word). The one methodological wrinkle a careful reader should know: JOLTS "job openings" counts positions employers are actively recruiting for in a given month, which is a widely used but imperfect proxy for a structural "shortage" — normal frictional vacancies get counted the same way unmeetable demand would. NAHB's phrasing ("short about 300,000 workers") elides that distinction, but the underlying number is real, current, and correctly sourced to BLS.

Claim 3 — Immigrants are "more than one-quarter" (26.3%) of the construction workforce, a 2024 record high

Verdict: Not independently checkable in this warehouse, but corroborated externally. Multiple independent trade-press outlets (Scotsman Guide, ConstructionOwners.com, Structural Building Components Association) reporting on NAHB's own analysis in 2026 all cite the identical 26.3% figure, attributed to NAHB's Natalia Siniavskaia's analysis of Census ACS microdata, with trade-level breakdowns (57% of drywall/ceiling installers, 56% of plasterers/stucco masons, 53% of roofers, 43% of construction laborers, 35% of carpenters). AskAmerica's own census.acs_nativity table confirms the population-wide foreign-born share was 13.9% nationally in 2023 (46.1 million of 332.4 million) — consistent context for a much higher concentration specifically within construction trades, which have long relied heavily on immigrant labor. However, AskAmerica's census.acs_industry table carries industry employment counts with no nativity breakdown, so the precise 26.3% construction-specific figure could not be recomputed directly from this corpus; it is a genuine, confirmed data gap here, not a defect. The figure traces entirely to NAHB's own research, though it is standard, replicable ACS-PUMS methodology and no contradicting figure was found anywhere in the trade or general press.

Claim 4 — HBI/University of Denver study: 1.98-month delay, 19,000 fewer homes, $10.8 billion annual cost

Verdict: True, accurately transcribed. Fetched directly from NAHB's June 10, 2025 primary press release ("New Study Reveals Significant Economic Impact of Housing Industry Labor Shortage"): the study reports an aggregate economic impact of $10.806 billion/year (direct carrying-cost impact $2.663B; lost production of ~19,000 single-family homes in 2024, an $8.143B loss) and an unweighted average construction-time increase of 1.98 months. NAHB's Sept. 25, 2026 post rounds $10.806B down to "$10 billion," which is defensible, and correctly attributes "two months" and "19,000 fewer homes." The material caveat: this is research commissioned by HBI (NAHB's own workforce-training affiliate) and "supplemented with an analysis by NAHB" — conducted by University of Denver researchers, but not independent, unaffiliated academic work, and not peer-reviewed. NAHB accurately represents what its own commissioned study found; the question is whether that study's own methodology holds up, which is outside what this fact-check can verify independently.

The sourcing pattern, stated plainly

Of the four load-bearing figures in this post, three (the housing-shortage estimate, the immigrant-workforce-share analysis, and the HBI/Denver economic-impact study) trace directly back to NAHB or its affiliate HBI's own commissioned research, not an independent government or academic source. Only the ~300,000-worker JOLTS figure is a genuine third-party government statistic, and even that is used somewhat loosely (job openings treated as "shortage"). None of the four figures were found to be fabricated, contradicted, or materially misstated relative to their own primary sources — NAHB is faithful to its own research. But a trade association funding the studies it then cites as evidence for the policy position it is advocating (looser immigration enforcement for its members' workforce) is a standard, material caveat for a fact-checker to name explicitly, not bury in a footnote. This is advocacy research accurately reported, not neutral government data as the post's framing ("government data indicate...") implies for the whole picture.

What This Report Does Not Answer

• Verify immigrant share of construction workforce (26.3%) directly against census.acs_employment or equivalent ACS tables if available: No industry-by-nativity crosstab exists in this warehouse (census.acs_industry has no nativity breakdown, census.acs_nativity has no industry breakdown) — confirmed gap, reported as such rather than guessed around. Population-wide foreign-born share (13.9%, 2023) was computed as directional context; corroborated instead via independent press coverage of NAHB's own ACS analysis.

Every query behind this report

6 warehouse calls ran in this session, in order. Each is reproducible against the same snapshot.

SELECT industry_code, industry_name FROM econ_reference.jolts_industries WHERE industry_name ILIKE '%construction%'

SELECT state, geo_name, "year", total_population, foreign_born FROM census.acs_nativity WHERE state='00' OR geo_name ILIKE '%United States%' ORDER BY "year" DESC LIMIT 5

SELECT geography, COUNT(*) FROM census.acs_nativity WHERE "year"='2023' GROUP BY geography

SELECT "year", "date", metric_type, value FROM econ.jolts_industry WHERE industry_code='230000' AND metric_type IN ('JOL','JOR') AND "year">=2025 ORDER BY "year" DESC, "date" DESC FETCH FIRST 30 ROWS ONLY

SELECT SUM(total_population) AS total_pop, SUM(foreign_born) AS total_foreign_born FROM census.acs_nativity WHERE "year"='2023' AND geography='state'

SELECT SUM(construction) AS total_construction_employed FROM census.acs_industry WHERE "year"='2023' AND geography='state'