What NAIC auto insurance data measures

The NAIC's Auto Insurance Database Report is the main official source for what drivers spend on car insurance in each state. It reports average expenditures and average premiums per insured vehicle, plus claim frequency, claim cost and loss ratios by coverage, along with traffic, theft, income and state law data. The latest edition, the 2022/2023 report adopted in December 2025 and released February 13, 2026, covers premiums through 2023 and claims through 2022. It is reliable for trends and broad comparisons, but it lags by about two years and does not tell you what any individual driver pays.

Who publishes it

The NAIC, the National Association of Insurance Commissioners, is the organization of state insurance regulators; see The NAIC explained. Its Casualty Actuarial and Statistical Task Force prepares the report from data collected by statistical agents that gather premium and claim data from insurers: the American Association of Insurance Services (AAIS), Insurance Services Office (ISO), the National Independent Statistical Service (NISS), Independent Statistical Service (ISS), the Massachusetts Commonwealth Automobile Reinsurers and the Maryland Automobile Insurance Fund, plus the California and Texas insurance departments (report).

The headline measures

A "car-year" is one vehicle insured for 365 days. Average expenditure assumes every insured car has liability coverage but not necessarily collision or comprehensive, so it estimates what consumers actually spent per vehicle. The combined average premium approximates the cost of a policy with all three coverages. The median state average expenditure in 2023 was $1,114 and the median state combined premium $1,268, according to the report; the NAIC summarized the edition in a February 2026 release.

"Liability" here is broad: it includes bodily injury and property damage liability, uninsured and underinsured motorist, medical payments and no-fault personal injury protection.

Claims data

The second half of the report uses earned premium, incurred losses and incurred claims for accident years 2020 through 2022, by coverage and state, separately for the voluntary market and the residual market (drivers insured through state plans because they could not get coverage otherwise). From these it calculates:

• Frequency: claims per 100 insured car-years.
• Severity: average loss per claim.
• Pure premium: losses per insured car-year.
• loss ratio: incurred losses as a share of earned premium, before insurer expenses.

Highlights from the 2022/2023 edition: total liability incurred losses reached $120.5 billion in 2022, up 9.4% from 2021; the average collision claim rose 17.6%, from $6,113 in 2021 to $7,191 in 2022; and the countrywide liability loss ratio rose from 61.4% in 2020 to 81.5% in 2022. These are the numbers behind the rate increases of 2023 and 2024; see why car insurance got more expensive, and where prices stand in 2026.

The context tables

The report also collects state data that helps explain cost differences: fatal accident rates, traffic density, vehicle thefts, population density, disposable income, and state laws on rate and form filing, tort thresholds, minimum liability limits, PIP and uninsured motorist requirements, seat belts, drunk driving and distracted driving. The NAIC notes that urban population, miles driven per mile of highway, and disposable income per person are correlated with state premiums.

Limitations to keep in mind

The NAIC is explicit about the caveats:

• No adjustment for who is insured. Averages do not account for driver classifications, vehicle types, limits or deductibles, so a state where people buy higher limits or drive pricier cars shows a higher average.
• State comparisons need caution. In the report's words, "direct comparisons between state results should be treated with a high degree of caution."
• Different state laws. No-fault states require personal injury protection; minimum limits vary; some states' figures are shaped by unique systems.
• State-specific notes. California's 2022 and 2023 data is preliminary. New Jersey insurers have historically paid two to four times the national average in policyholder dividends, which the report does not subtract. Texas comprehensive exposures are estimated. Massachusetts figures reflect its Safe Driver Plan credits and surcharges.
• What is excluded. Motorcycles, motor homes, RVs, antique autos and similar vehicles are not included, so totals differ from insurers' financial statements.
• Lag. Premium data runs about two years behind, and claims data about three. Historical figures are revised in each edition, so numbers can differ from earlier reports.

The NAIC also explains why its spending figures rise more slowly than the Consumer Price Index for car insurance: when prices go up, some drivers drop coverage or raise deductibles. From 2019 to 2023, average expenditure rose 19.24% while the CPI auto insurance component rose 36.56%.

How to use the report

1. For your state's level and trend, use average expenditure across several years rather than one year's rank. Our the most and least expensive states for car insurance guide summarizes the 2023 rankings.
2. For a policy with full coverage, look at the combined average premium, remembering it is an approximation.
3. To understand why your state costs what it does, compare frequency, severity and the context tables with neighboring states.
4. For current prices, pair it with the monthly CPI and your own quotes; see how to read car insurance rate studies and average-cost claims and the average cost of car insurance, and what the numbers mean.
5. To evaluate an insurer, use different NAIC tools, such as the complaint index; see The NAIC complaint index: how to look up an insurer.

Key takeaways

• The NAIC Auto Insurance Database Report is the official source for state-by-state auto insurance spending and claims data.
• Average expenditure ($1,281.92 in 2023) measures what drivers spent per vehicle; combined average premium ($1,438.60) approximates a policy with liability, collision and comprehensive.
• Claims tables show frequency, severity and loss ratios that explain why rates move.
• The data lags about two years and does not adjust for coverage choices, so use it for context, not as a quote.