The 2024 Permit Crash That Wasn't: A Reporting-Lag Case Study
Data Quality
The raw data looked like a 36% permit collapse. A matched reporting window showed 1.6% down. The difference was backfill, not the economy.
The Level Index
The correction
An earlier version of this article treated incomplete 2024 public permit records as a market contraction. That conclusion was wrong.
The raw full-year series made 2024 permit volume look about 36% lower than 2023. But the second half of 2024 was still being backfilled into the source. The national ratio of second-half to first-half records fell to about 0.35 in 2024, compared with roughly 1.1 in complete years. That is a completeness warning, not an economic signal.
When we compare the same complete January through June window in both years, permit volume was down 1.6%, essentially flat. The independent Census Building Permits Survey reported permitted housing units down 2.2%, from 1,514,772 in 2023 to 1,482,048 in 2024. The two sources do not measure exactly the same permit universe, but both reject the collapse story.
| Measure | 2023 to 2024 result | Valid use |
|---|---|---|
| Raw full-year public permit records | about -36% | Do not use, recent backfill was incomplete |
| Matched January-June permit records | -1.6% | Comparable national trend |
| Census Building Permits Survey units | -2.2% | Independent residential check |
The corrected benchmark and full method are available in the permit-market dataset.
Why a caveat was not enough
The old article added a paragraph saying that 2024 might be partial, but it left the decline in the title, state table, risk framing, and recommendations. That still told the reader to act on a number we knew might not be comparable.
A disclosure does not repair a decision claim. If incomplete data drives the headline, color, risk label, density calculation, or recommendation, the page is still wrong. The incomplete field has to be removed from the decision path.
That is why Level no longer publishes those raw state declines as market risk. We also retired the old state and trade pages whose cells could not be reproduced from a committed query and named source.
Free benchmark review
See how your operating numbers benchmark.
We compare your numbers against real benchmarks from companies like yours and turn the gaps into a 90-day fix plan.
What contractors can conclude
The national 2024 permit signal was roughly flat, not a broad collapse. That does not mean every local market or project type was flat. A national aggregate cannot establish demand in Houston, multifamily starts in Florida, or a commercial mechanical contractor's backlog.
For a local forecast, use measurements at the right grain:
- signed backlog and expected start dates
- quote volume, win rate, and quote age by work type
- local Census or permitting-agency releases with complete comparable periods
- County Business Patterns establishment growth
- serviceable building stock and customer concentration
- billing speed, collections, and cash timing on work already won
Level's state benchmark pages publish reproducible labor, building-stock, serviceable-demand, and employer-firm measures across all 50 states plus DC. The metro building market adds commercial structures and plumbing and HVAC establishment counts for 100 metros, with the limits shown beside the figures.
The operating lesson
Bad external data can create the same failure as bad books: it makes a precise recommendation look better supported than it is.
Before putting a recent-vintage series into a forecast, ask four questions:
- Is the reporting window complete?
- Are both periods measured at the same grain?
- Does an independent primary source show the same direction and magnitude?
- Would the decision change if late records arrived tomorrow?
If the answer to the first two questions is no, stop. Do not turn the field into a risk score.
The contractor decision is not "prepare for 36% fewer jobs." It is to build the forecast from current backlog, direct local releases, and operating data the company can reconcile. Use market data as context, not as a substitute for the pipeline and cash model.
FAQ
Did US building permits fall 36% in 2024?
No. That was the appearance in an incomplete public permit-record extract. A matched January-June comparison was down 1.6%, and Census Building Permits Survey units were down 2.2%.
Can the matched national result be used as a state forecast?
No. It corrects the national reporting-lag error. A state or metro forecast still needs a direct, complete local series at the same grain.
Why keep this URL live?
Because the correction is more useful than silently removing the earlier claim. Existing links now land on the full explanation, the corrected figures, and the method.
Where should a contractor start?
Use the cash-gap calculator for a project-level decision, or review the Level Index for operating benchmarks and reproducible state context.
Get the next one
Want next week's benchmark in your inbox?
One email a week. Real numbers from 2,200+ service businesses. No fluff. Unsubscribe anytime.
Related reads
Business Growth
What EBITDA Multiple Do Electrical Contractors Sell For? (2026)
Electrical contractors sell for 5.5-8x EBITDA in the lower-middle market, but data-center and mission-critical specialists pull 8-12x+. Why revenue mix now matters as much as size.
Business Growth
What EBITDA Multiple Does an HVAC Company Sell For? (2026)
HVAC companies sell for ~9.5x EBITDA on average, 4-8x for tuck-ins up to 15-18.5x for service-dense platforms. What moves you up the range, and how to find where you sit.
Business Growth
Owner-Direct Revenue Is A Data Channel For Contractors
Owner-direct relationships are valuable because they expose asset history, service frequency, callbacks, renewal margin, and pull-through revenue.
About the author
Sam Yang
Founder & CEO
Founder of Level, the AI operating layer for contractors and skilled trades, and the other operating businesses where scarce labor is the constraint. Ex-CFO across trades, SaaS, and service businesses. 4 years as Director of Growth Product at BuildOps, building financial tooling used by 1,000+ commercial contractors. Four years in PE and investment banking rolling up and acquiring service businesses, $2.5B in total transactions including M&A and IPOs. Stanford MBA, Brown undergrad. The Level founding team's analysis of 2,200+ contractors ($13.25B in revenue) across operating, private-equity, and CFO roles anchors the Level Index benchmark research.
LinkedIn