Research
Research
Studies from the HyreRoof Research Desk. Every claim is sourced to a named primary document, a statute, a licensing board, a federal dataset, with the date it was retrieved. Where a source could not be reached, the page says so instead of filling the gap.
Last updated .
The short version: eleven studies, each built from primary federal or state records rather than from other people’s summaries. They cover who is allowed to roof (licensing, by state and by city), what the code actually requires (the ice-barrier line), what the work costs over time (material against labour inflation), who does it (shortage and fatality data), and where the law protects a homeowner after a storm. Every figure carries its source and the date it was retrieved, and gaps are published as gaps.
Published studies
Eleven studies are published. Each row links to the study, says what it reads and when it was last updated; the method, the sources and the limitations live on the study page itself.
| Study | What it reads | Updated |
|---|---|---|
| Is Roofing Even Licensed Where You Live? | Roofing is regulated under six structurally different regimes in the United States, and "licensed or not" turns out to be the wrong question. Eight states verified against the statutes and licensing authorities themselves, with what each state actually checks before it issues. | September 3, 2026 |
| Is There Really a Roofer Shortage? | The shortage is asserted constantly and measured almost never. Built from BLS employment data against Census housing counts, including the self-employment gap that makes most published roofer counts too low. | September 3, 2026 |
| What Actually Happened to the Price of a Roof | Materials and labour, decomposed and tracked against general inflation from federal price series. Producer prices are not what a homeowner pays, and the page is explicit about the difference. | September 3, 2026 |
| America’s Oldest Housing Stock, City by City | The hundred largest US cities ranked on the Census Bureau’s median year structure built, with the margins of error and the three cities the scale literally cannot separate. Housing age is a proxy for roof age, and the page says how weak a proxy it is before it shows you the table. | September 3, 2026 |
| Homes Entering the Roof Replacement Window | A cohort model over Census year-built data, run at 20, 25 and 30-year assumptions rather than presented as one number, then checked against the 8.3 million roofing projects the American Housing Survey actually counted. | September 3, 2026 |
| The Cities That License Your Roofer When the State Will Not | Where no state roofing licence exists, the city decides, and the cities disagree completely. Ten municipal programmes across Colorado and Texas, read at their own codes and application forms, from an ICC exam plus $1m of cover to a $120 fee and a mailing address. | September 3, 2026 |
| Does Your Roofer Have to Be Insured? | "Licensed and insured" is four separate legal instruments compressed into three words. Eight jurisdictions verified at the statute, the agency or the municipal code: including the places that require none of them. | September 3, 2026 |
| The Roofing Fatality Atlas | Roofers died at 48.7 per 100,000 full-time equivalent workers in 2024, about fifteen times the all-worker rate. Plus eight years of OSHA enforcement. | September 3, 2026 |
| The Ice Barrier Line | IRC R905.1.2 does not draw a line on a map. It defers to a YES/NO cell each jurisdiction fills in itself. Code text, seven determinations, 62 NOAA stations. | September 3, 2026 |
| After the Storm: The Laws Written to Control Roofing Contractors Who Follow the Weather | The post-storm contractor problem is legislated state by state: cancellation rights, deductible offences, adjusting bans. Nine states read at the statute. | September 3, 2026 |
| Inside the Florida Roofing Market | Florida rewrote its roofing and insurance law four times in nineteen months. We read every statute and analysed 9,402 DBPR roofing licences. | September 3, 2026 |
What is not here yet
Two more studies are planned, roof age and insurance rules, and hail exposure by county, but neither is published, and this page will not list a study before it exists.
Both depend on datasets we have not yet confirmed we can retrieve. Until we can, they stay unwritten rather than being assembled from secondary summaries of the data.
How this research is made
Research is attributed to the HyreRoof Research Desk, not to a named individual, because it is organisational work. Primary sources are preferred in every case where one exists: the statute rather than a summary of it, the licensing board rather than a directory that copied the board. Where our own analysis goes beyond what a source states, the page says which is which.
The full policy, sourcing, what we will not publish, how we expect to make money, and how corrections work, is on the editorial policy page.
Where to go from here
The research explains the landscape. If you have a roof in front of you and a decision to make, the arithmetic is more use than the analysis: the roof cost calculator prices a job line by line, and the repair vs replacement calculator puts a threshold under the choice. Neither asks for your details. When you are ready to talk to someone, tell us about the project.
Common questions
Who writes HyreRoof research?
The HyreRoof Research Desk. Research here is attributed to the desk rather than to a named individual, because it is organisational work rather than one person’s reporting.
What counts as a source on these pages?
A primary document, preferred in every case where one exists: the statute rather than a summary of it, the licensing board rather than a directory that copied the board, the federal dataset rather than an article about it. Every claim carries the date the source was retrieved. Where our own analysis goes beyond what a source states, the page says which is which, and where a source could not be reached, the page says so instead of filling the gap.
Why are planned studies not listed here?
Because this page will not list a study before it exists. Two of the planned studies depend on datasets we have not yet confirmed we can retrieve, and until we can they stay unwritten rather than being assembled from secondary summaries of the data.
Where is the full editorial policy?
On the editorial policy page, at /editorial-policy/. It covers sourcing, what we will not publish, how we expect to make money, and how corrections work.
Primary sources behind these studies
- Employment, wages, fatality and price series, U.S. Bureau of Labor Statistics. Checked September 18, 2026.
- American Housing Survey, U.S. Census Bureau. Checked September 18, 2026.
- Storm Events Database, NOAA National Centers for Environmental Information. Checked September 18, 2026.
- Model code text, International Code Council. Checked September 18, 2026.