Research

We publish research because the restoration trade is covered almost entirely by vendors pricing their own services and by subscription market reports. Neither publishes a method. These do.

The instruments

Flood loss instruments

Built from the complete set of 626,886 FEMA NFIP claims with a loss year of 2015 or later, retrieved 4 October 2026.

• The Water Depth Damage Curve: 626,886 NFIP Claims Analysed
  Median building damage rises 3.3 times across the first two feet of water, then flattens. Computed from 626,886 FEMA NFIP claims filed since 2015.
• Flood Claims by State: Where Water Losses Concentrate
  Florida, Texas and Louisiana hold 67.8% of NFIP building claims since 2015, but Louisiana leads per head. Computed from 491,235 FEMA claims.
• The Insured Flood Loss Gap: Claimed Against Paid
  The median NFIP building claim settles at 95.8% of assessed damage. Deductibles, the $250,000 coverage limit and depreciation explain the gap. Computed from 479,868 FEMA claims.
• Flood Loss by Building Type: Homes, Rentals and Commercial
  Non-residential NFIP claims run 1.26 times residential ones at the median and 1.65 times at the mean, and 26.4% of claims come from outside high-risk flood zones. Computed from 626,886 FEMA
• The Contents Restoration Report: What Is Inside the Building
  Median contents damage is 28% of building damage on an NFIP flood claim, and only 47% of building claims record contents at all. Computed from 626,886 FEMA NFIP claims filed since 2015.
• The Flood Zone Exposure Report: Claims Outside the Mapped Risk
  28.6% of NFIP flood claims come from outside the mapped high-risk zones, and on 7.5% of claims the rated zone and the current map disagree. Computed from 626,886 FEMA NFIP claims filed since
• Inflation-Adjusted Flood Damage: A Decade of NFIP Claims in 2024 Dollars
  In 2024 dollars the median NFIP flood claim swings 3.9 times across 2015 to 2024, tracking the year’s largest flood rather than a steady rise. Computed from 477,160 FEMA claims and the BLS C

Storms, freezes, fire and flood insurance prices

Built from the NOAA Storm Events Database (2015 to 2025), 3,555,151 FEMA NFIP policies in force on 31 March 2026, FEMA housing registrations from the 2021 Texas winter storm and Census AHS 2023 pipe leak items, retrieved 6 to 8 October 2026.

• Flood Insurance Cost by State: 3.56 Million NFIP Policies Priced
  Median NFIP flood insurance for a single-family home: $1,022 a year, from $367 in Alaska to $1,467 in Connecticut. 3,555,151 FEMA policies.
• Hurricane Damage by State: NOAA Storm Records, 2015 to 2025
  Florida logged $38.4 billion of $81.5 billion in NOAA hurricane damage, 2015 to 2025; five episodes hold 58%. State rankings and FEMA comparison.
• Tornado Damage by State: 14,271 Tornadoes in NOAA Records, 2015 to 2025
  Texas logged $2.34 billion of $13.3 billion in reported US tornado damage, 2015 to 2025; Missouri leads per resident. From 16,489 NOAA records.
• Wildfire Damage by State: NOAA and FEMA Records, 2015 to 2025
  California holds 63% of $30.2 billion in NOAA wildfire damage, 2015 to 2025; one fire holds 56%. Only 20% of records are priced.
• Frozen Pipe Damage Statistics: The 2021 Texas Freeze and US Homes
  Texas 2021 freeze: 434,568 home insurance claims, NOAA damage $275 million. 3.7% of US homes report a pipe leak a year. FEMA, Census, NOAA data.

Water, mold and leaks in American homes

Built from the Census Bureau and HUD American Housing Survey 2023 national public use file, 48,527 occupied sample homes weighted to 133.2 million, with margins of error from the survey’s 160 replicate weights.

• How Common Are Water Leaks in Homes? About 3 in 20 Homes a Year
  15.4% of US homes, about 20.5 million, reported a water leak in the past 12 months. Pipes lead inside leaks, roofs lead outside leaks. Census AHS 2023.
• Percentage of Homes With Mold: What the American Housing Survey Shows
  An estimated 3.0% of occupied US homes, 3.9 million, reported visible mold in the past 12 months. Homes with a leak: 10.6%, 6.7x the no-leak rate. Census AHS 2023.

Storms, loans and the insurance gap

Built from the NOAA Storm Events Database (2015 to 2025), SBA Disaster Loan Data (FY2015 to FY2022) and 6,367,334 FEMA housing registrations from eight large disasters, retrieved 6 October 2026.

• Flash Flood Deaths by State: 11 Years of NOAA Storm Event Records
  Texas logged 338 of 1,089 US flash flood deaths, 2015 to 2025; West Virginia has the highest rate per million. From 45,915 NOAA records.
• How Many Disaster Victims Have Insurance? 6.37M FEMA Registrants
  Of 6.37 million FEMA disaster registrations, 56.4% of owners had homeowners and 9.8% of all had flood insurance. 73.7% of flooded owner homes had no flood policy.
• SBA Disaster Loan Statistics: Verified Losses Against Loans Approved, FY2015 to FY2022
  SBA approved $14.9bn in disaster home loans against $31.3bn of verified loss, FY2015 to FY2022: 47.7%. Contents were 24.8% of the loss.

Flood insurance, mitigation and federal money

Built from four OpenFEMA datasets retrieved in full on 5 October 2026: county flood insurance take-up, the NFIP multiple loss property list, Hazard Mitigation Assistance projects and the FEMA disaster financial summaries, joined to the declarations file.

• Percentage of Homes With Flood Insurance: The County Take-Up Map
  FEMA data put the share of US homes with an NFIP flood policy at 2.89%, 43.9% inside the mapped flood zone and 1.25% outside it. County and state take-up for 3,144 counties.
• Repetitive Loss Properties: 240,651 Buildings That Keep Flooding
  FEMA lists 240,651 repetitive loss properties. 56.4% are not NFIP insured today and 98.5% of severe repetitive loss buildings are not mitigated. State, county and ZIP tables.
• FEMA Buyouts by State: Where Hazard Mitigation Money Buys Out Homes
  FEMA mitigation grants have bought out 55,805 properties for $4.28 billion in federal share, led by Texas, Missouri and North Carolina. State, county and program tables from 5,848 funded FEMA projects.
• FEMA Spending by State: 1998 to 2026 Disaster Aid, Ranked
  Outside COVID-19, FEMA put $243.7 billion into disaster aid since 1998, led by Puerto Rico. COVID-19 adds $97.9 billion, ranked separately.
• The Counties With the Most Disaster Declarations, 2015 to 2026
  Owsley County, Kentucky leads US counties with 19 major disaster declarations since 2015, and 34 of the 71 most-declared counties are in Kentucky. Computed from 10,350 FEMA county designations.

Disaster timing and capacity

Built from FEMA disaster declarations, BLS establishment counts and the NOAA billion-dollar disaster archive.

• The Disaster Declaration Lag: How Long Before Help Is Declared
  Major disaster declarations come a median 58 days after an incident begins; fire grants come in 1 day by design. Computed from 1,422 FEMA declarations since 2015.
• The Restoration Seasonality Calendar: When Disaster Work Arrives
  Restoration-relevant FEMA disasters peak in August and bottom out in November, a 5.7 times swing, and every state has its own season. Computed from 1,422 declarations since 2015.
• The Restoration Capacity Index: Disaster Load per Firm by State
  Kentucky carries 0.78 declared disaster events a year per NAICS 562910 establishment, against 0.15 nationally. FEMA declarations set against BLS QCEW firm counts, with the scope limits state
• The Billion-Dollar Disaster Record, 1980 to 2024, and Why It Stopped
  403 NOAA billion-dollar disasters, 1980 to 2024, now frozen. Severe storms are 50% of events, tropical cyclones 53% of cost.

Disaster assistance instruments

Built from 308,360 FEMA Housing Assistance records retrieved 3 October 2026.

• The Restoration Job Size Distribution: Where the Work Actually Sits
  68.3% of owner-occupied FEMA housing inspections record damage between $1 and $10,000, against 2.6% above $30,000. Computed from 7,974,241 inspections.
• The Owner and Renter Damage Gap: Two Different Disasters
  Approval rates sit 46.8% against 45.9%, but rental assistance is 31.0% of renter dollars against 16.2% for owners. Computed from 308,360 FEMA rows.
• The Disaster Grant Ledger: What FEMA Assistance Actually Pays
  The mean approved FEMA housing grant is $3,331 and 0.80% of awards reach the statutory maximum. Computed from 12,036,730 approvals across 630 declared disasters.

Industry structure

Built from BLS wage and establishment series, and FEMA declarations.

• The Restoration Wage Report: Hazmat Removal Pay in All 50 States
  Mean pay for hazardous materials removal workers runs from $31,960 to $71,910 by state. Computed from the BLS OEWS May 2025 file for SOC 47-4041.
• The Remediation Establishment Census: Firms and Employment by State
  9,387 private establishments and 96,727 employees in NAICS 562910, state by state. Also why that code is environmental remediation and not the restoration industry.
• The Restoration Demand Volatility Index: Where Work Is Steady
  Volatility in declared disaster demand spans a 3.3 times range across US states. Computed from 17,942 county-level FEMA declarations, 2015 to 2025.

Regulation and registers

Built from state licensing registers and primary legal sources.

• The State Mold Licensing Map: What 50 States Actually Require
  Only 5 of 51 US jurisdictions issue a state mould licence and 42 have no programme at all. Classified from primary statutes, with corrections for the six states that published lists get wron
• Mold Licensee Density: Texas, Florida and New York Compared
  Texas 32.2, Florida 27.8 and New York 58.6 mould licensees per 100,000 residents, computed from the three state registers. 30% of New York’s contractor register is expired.
• The Asbestos Accreditation Map: AHERA, Waivers and State Routes
  AHERA makes asbestos accreditation effectively universal across 53 jurisdictions. What differs is who runs it: four agency types across the 22 states read, and only 4 of 10 rosters machine r

How these are built

• Public sources only. If a finding cannot be reproduced from a free dataset, we do not publish it as a finding.
• Every report carries a reproducibility block giving the endpoint, the query, the row count and the retrieval date.
• Data quality is published, not hidden. Where a source field is unreliable we show the exclusion counts and the treatment applied.
• Limits come before conclusions. Each report states what it cannot show.
• Corrections are made in public. On 4 October 2026 the flood instruments were rebuilt on the complete NFIP claims set after an earlier retrieval was found to have stopped at 400,000 rows. Medians and ratios moved by under one per cent; counts were corrected upward.
• No market-size estimates. There is no clean official count of the restoration industry, and we explain why rather than inventing one.

Citing this research

Each report carries a ready-made citation line and a CC BY 4.0 notice. Journalists and researchers are welcome to use the figures with attribution to Remediation Insider and a link to the report. If you need a cut of the underlying data that is not published, write to editorial@remediationinsider.com.