Introduction

Property insurance is a vital component of real estate markets, as lenders require insurance policies for mortgage loans and homeowners rely on payouts following disasters. This research identifies homeowners insurance premiums as a leading price signal for changing climate-related physical risks. However, understanding this market has been hampered by existing public data that is often coarse and out-of-date. The paper introduces a new measure of homeowners insurance premiums based on mortgage escrow data to provide evidence on the factors driving recent growth in the sector.

The residential homeowners insurance market

Property and casualty (P&C) insurance protects homeowners from physical and liability damages. In 2024, insurers in the United States collected $173 billion in gross premiums for homeowners insurance products. Approximately 94 per cent of homeowners maintain insurance coverage. Most owner-occupied residential policies are “HO-3” contracts, which cover structures, contents, and legal expenses, but specifically exclude certain perils such as floods and earthquakes. To manage risk, insurers diversify losses across multiple business lines and frequently purchase reinsurance from global providers to limit downside risk from catastrophic events.

Data

The study utilizes a new dataset developed from mortgage escrow accounts provided by CoreLogic’s Loan-Level Mortgage Analytics (LLMA). This dataset includes over 74 million observations from 2014 to 2024. Disaster risk data is sourced from the National Risk Index (NRI) and the First Street Foundation (FSF), covering 18 physical hazards. Housing and insurance coverage data are supplemented with zipcode-level estimates from the Federal Housing Finance Administration (FHFA) and US Census data. Reinsurance market tightness is tracked using the Guy Carpenter Rate-on-Line Index for the U.S. property catastrophe market.

Measuring insurance expenditures

Insurance premium expenditures are inferred from mortgage servicer data on borrower payments into escrow accounts. For most borrowers, the total monthly payment consists of principal, interest, property taxes, and insurance (PITI). The researchers decompose these payments to isolate property insurance expenditures. The methodology employs a repeat-loan specification, similar to a home price “repeat sales index”, which follows changes in premiums for the same loan over time. This approach effectively holds fixed all time-invariant characteristics of the underlying property, borrower, and location.

Homeowners premiums across time and geography

The analysis finds that average nominal annual premiums increased from $1,690 in 2017 to $2,750 in 2024. This represents a significant rise, particularly since 2021, with average premiums increasing by over $900 in just three years. There is substantial geographic variation; the riskiest coastal areas along the Gulf of Mexico and the East Coast, as well as tornado and hail-exposed regions in the Great Plains, face the highest insurance burdens. In some cases, insurance expenditures account for more than 40 per cent of monthly housing service outlays.

Insurance expenditure correlates and dynamics

Housing structure values and disaster risk explain most of the variation in insurance premiums. The research documents that the increase in construction costs explains 35 per cent of the rise in homeowners insurance premiums after 2017. However, the largest increases were concentrated in zipcodes with the highest disaster risk. The association between a one standard deviation increase in disaster risk and premiums rose from $220 in 2017 to $615 by 2024. This steepening risk-to-premium gradient explains 20 per cent of the growth in premiums during this period.

Capital markets and the cost of disaster risk

The rise in homeowners insurance premiums coincided with a doubling of U.S. property catastrophe reinsurance prices between 2018 and 2024. This “reinsurance shock” increased 2024 annual premiums by an average of $425 among zipcodes in the top decile of catastrophe exposure, explaining approximately 45 per cent of the rising pass-through of disaster risk in those areas. The study also notes that higher reinsurance costs have impacted insurance availability, leading to policy non-renewals in highly exposed counties.

House price effects and implications of the rising price of catastrophic risk

Premium increases induced by the reinsurance shock have been capitalised into home values. In zipcodes in the top decile of catastrophe exposure, home values were reduced by an average of nearly $44,000. This represents a relative home price decline of 11 per cent in the most exposed areas where disaster risk is expected to continue increasing. These findings indicate that housing market participants treat these premium increases as non-transitory, reflecting a major repricing of climate-exposed housing assets.

Conclusion

The cost of property insurance serves as a leading indicator for adapting to climate-related risks. The research established that recent premium growth has been concentrated in disaster-prone zipcodes and driven significantly by global capital market pricing of catastrophic risk. As reinsurance and catastrophe bond markets reprice these risks, the impacts will increasingly affect household budgets and real estate values in climate-exposed regions.