Hurricane disaster declarations show no significant rising trend since 2010; homeowners premiums rose steadily anyway

Summary

No. The frequency of federally declared hurricane disasters in the 18 Atlantic/Gulf coastal states has not shown a statistically significant rising trend over 2010-2025 (OLS regression of annual count on year: slope +0.097 state-disasters/year, p=0.845, n=16, r-squared=0.003) -- it is instead extremely volatile, from zero declarations in 2014-2015 to 30 in the record 2021 season. Homeowners insurance premiums, meanwhile, rose steadily and substantially over the same broad period regardless of any single year's hurricane count: the national average premium climbed 43% in nominal dollars (14% after inflation) from 2013 to 2022, with growth accelerating sharply in 2021 (+7.6%) and 2022 (+11.2%). A same-year correlation between coastal hurricane-declaration counts and premium growth is weak and not significant (r=0.18, p=0.65); a one-year-lagged version (this year's premium growth vs. last year's hurricane count) is moderate and marginally significant (r=0.70, p=0.038) but rests on only 9 annual observations and should not be read as proof of a lag effect. In short: premiums are tracking rising rebuilding/reinsurance costs and claim severity broadly, not the count of declared hurricane events specifically.

Step 1 -- Has hurricane disaster frequency actually risen?

Using FEMA's OpenFEMA disaster-declarations data (disasters.disaster_declarations), disasters with incident_type = 'Hurricane' were counted as distinct disaster-number x state combinations across 18 Atlantic and Gulf coastal states (AL, FL, GA, LA, MS, NC, SC, TX, VA, MD, DE, NJ, NY, CT, RI, MA, NH, ME) for 2010-2025 (2026 is only partially in the window and excluded). The annual series is: 2010: 5, 2011: 21, 2012: 5, 2013: 19, 2014: 0, 2015: 0, 2016: 1, 2017: 19, 2018: 13, 2019: 16, 2020: 16, 2021: 30, 2022: 8, 2023: 8, 2024: 7, 2025: 4.

An OLS regression of this count on year (run via this session's ols_regression tool) gives a slope of +0.097 disasters/year, standard error 0.488, t=0.199, p=0.845, r-squared=0.003 -- statistically indistinguishable from zero, and the tool itself flags this as a small-n (n=16), single-predictor estimate with wide standard errors. The series is dominated by extreme year-to-year swings (two consecutive zero years in 2014-2015, then a record 30 in 2021) rather than a monotonic trend. This matches the broader meteorological literature: total U.S. landfalling hurricane counts have not shown a clear long-term increasing trend, even though hurricane intensity and associated rainfall/storm-surge damage have trended up in many studies. It is also worth flagging that a FEMA declaration is a policy outcome (a governor's request, approved by the President) as well as a weather outcome -- it can shift with administrative practice, not just storm activity -- so this measure is at best a proxy for hazard frequency, not a pure meteorological count.

Step 2 -- Has homeowners premium growth tracked that frequency?

This corpus (askamerica) does not carry a homeowners-insurance premium table -- search_catalog confirmed "homeowners" as an unmatched term against every schema; the closest analogues (NFIP flood policies, crop insurance) cover different risks. National average homeowners premium data therefore comes from NAIC data as published via the Insurance Information Institute (Triple-I): $1,096 (2013) rising steadily to $1,569 (2022), a 43.2% nominal increase. Adjusted to 2022 dollars using BLS CPI-U (via this session's adjust_inflation tool), the real increase over the same period is 14.0% ($1,377 to $1,569) -- still a genuine real increase, but roughly a third the size of the nominal figure, with inflation accounting for the rest. Growth accelerated most in the final two years of the series: +7.6% in 2021 and +11.2% in 2022 (nominal), the two years following the record-active 2020 and 2021 hurricane seasons and a period of rising reinsurance and rebuilding costs broadly.

NAIC's more recent (2018-2024) regional analysis, cited via Triple-I and CNBC coverage of the report, found inflation-adjusted premium growth of 18% in the Northeast, 25% in the Midwest, 27% in the Southeast, and 43% in the West over that period. The Southeast region -- which contains most of the Gulf/Atlantic hurricane-exposed states in this analysis -- had the second-highest regional growth rate, behind only the West (where wildfire, not hurricanes, is the dominant peril). That regional pattern is at least directionally consistent with hurricane exposure being one factor behind faster Southeast premium growth, though it is far from the largest one in the data (the West's non-hurricane-driven growth was highest).

Step 3 -- Does frequency track growth statistically?

Aligning the coastal hurricane-declaration count with the national premium series for the years both are available (2013-2022): the same-year correlation between hurricane count and the premium level is weak (r=0.28, p=0.43, n=10). Correlating same-year hurricane count against that year's premium percent-change is also weak and insignificant (r=0.18, p=0.65, n=9). Correlating the prior year's hurricane count against the following year's premium percent-change -- allowing for the real-world lag between a bad storm season and an insurer's next rate filing being approved by a state regulator -- gives a moderate, marginally significant relationship (r=0.70, p=0.038, n=9). Given the very small sample (9 annual data points) this lagged result should be read as suggestive, not as established causation; it is also consistent with, but does not prove, the standard insurance-industry account that rate increases take one to several years to work through state regulatory approval after a loss year.

The more robust finding is the asymmetry: premium growth was continuous and substantial across nearly every year in the series -- premiums rose every single year from 2013 to 2022 -- while hurricane-declaration frequency bounced between zero and 30. Premiums did not fall in the quiet 2014-2015 seasons, and did not spike or dip in lockstep with any single active or quiet hurricane year. That pattern says premium growth is driven mainly by cumulative and broad-based cost pressure -- rebuilding/materials costs, reinsurance pricing, general claim severity (wind/hail claim severity rose from $14,747 per claim in 2019 to over $19,700 by 2022 per ISO/Triple-I data) and litigation/loss trends -- rather than by the count of hurricanes declared in any given year.

Every query behind this report

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

SELECT year, state_fips, count(distinct disaster_number) as hurricane_disasters
FROM disasters.disaster_declarations
WHERE incident_type = 'Hurricane'
  AND state_fips IN ('01','12','13','22','28','37','45','48','51','24','10','34','36','09','44','25','33','23')
GROUP BY year, state_fips
ORDER BY year

SELECT year, count(distinct disaster_number || '-' || state_fips) as hurricane_state_disasters
FROM disasters.disaster_declarations
WHERE incident_type = 'Hurricane'
  AND state_fips IN ('01','12','13','22','28','37','45','48','51','24','10','34','36','09','44','25','33','23')
GROUP BY year
ORDER BY year

SELECT CAST(year AS INT) as yr, count(distinct disaster_number || '-' || state_fips) as hurricane_state_disasters
FROM disasters.disaster_declarations
WHERE incident_type = 'Hurricane'
  AND state_fips IN ('01','12','13','22','28','37','45','48','51','24','10','34','36','09','44','25','33','23')
GROUP BY CAST(year AS INT)

SELECT * FROM (VALUES (2013,1096),(2014,1132),(2015,1173),(2016,1192),(2017,1211),(2018,1249),(2019,1272),(2020,1311),(2021,1411),(2022,1569)) AS t(yr, premium)

SELECT * FROM (VALUES (2010,5),(2011,21),(2012,5),(2013,19),(2014,0),(2015,0),(2016,1),(2017,19),(2018,13),(2019,16),(2020,16),(2021,30),(2022,8),(2023,8),(2024,7),(2025,4)) AS t(yr, hurricane_count)