Property and Casualty Insurance Industry Statistics 2026

A US property and casualty insurer that closed its 2025 books found a 92.9% combined ratio, a $68.7 billion underwriting gain and the best result in years. It also found net income down on the year before.

Both are true, and the gap between them is the whole problem with reading an industry at a glance.

A profitable year can come from better pricing, from favourable development on old claims, or from a quiet catastrophe season. Each implies something different about the next renewal.

These property and casualty insurance industry statistics cover market size, underwriting performance, expenses and the pricing trend running through 2026. US statutory results provide the financial benchmark, with global research and broker renewal data alongside. A completed financial year and a forecast are different kinds of evidence, so reporting periods are stated throughout.

Research updated 9 September 2026. Reporting periods are stated alongside the figures.

Property and casualty insurance industry statistics at a glance

The table mixes scale, performance and pricing. Premiums are not claims payments, surplus is not annual profit, and a renewal-rate index does not measure growth in the market as a whole.

How big is the P&C insurance market?

Swiss Re puts the global market at roughly $2.4 trillion, after a doubling in nominal size over two decades. Its historical series runs through 2024, and that reporting date should travel with the number into any 2026 presentation.

Before setting it against another estimate, check three definitions.

• Business included. P&C, non-life and all insurance are not interchangeable categories, and some non-life datasets sweep in health business.
• Premium basis. Direct, gross and net written premium describe different positions in the insurance and reinsurance chain.
• Growth basis. Nominal growth carries price and currency effects; real growth strips inflation out.

An insurer can report higher premiums without writing a single additional policy. Rate increases, rising insured values and a shift in business mix all inflate the premium base.

Separating those drivers matters more than it sounds. Growth from new customers carries acquisition and servicing costs; growth from higher replacement costs on the existing book carries claims exposure instead.

Which property casualty insurers are largest?

AM Best’s March 2026 compilation ranks these groups highest by 2025 US net premiums written.

These are group figures from a US statutory population. They are not a ranking of global consolidated revenue, and they are not direct premium market share.

Size also says very little about whether a niche is worth entering. A large personal lines portfolio and a specialist liability book face different loss trends, different distribution economics and different capital requirements.

For an MGA or a smaller carrier the useful comparison is a closely matched segment: same line, same customer type, same territory, same channel. Anyone shortlisting systems for that work will find the same fragmentation among P&C insurance software vendors.

Underwriting improved, but profit did not move in a straight line

The NAIC reports a 92.9% combined ratio for 2025, against 96.9% in 2024 and 101.7% in 2023.

A preliminary study from Verisk and APCIA adds the part the ratio hides. In its private US insurer dataset net earned premiums rose 6.3% while incurred losses and loss adjustment expenses fell 0.4%. Net income after tax still dropped from about $169 billion to $148 billion, largely on lower realised capital gains.

Better underwriting and lower net income, in the same year.

Why the two reports disagree

Verisk and APCIA estimated a $63 billion underwriting gain. The NAIC reported $68.7 billion. Neither is wrong; the reporting populations and data cutoffs differ.

The practical rule is to stay inside one series. Splicing this year from one source onto last year from another produces a growth rate that describes nothing.

What does the insurance industry combined ratio actually show?

In the US statutory presentation it combines the loss, loss adjustment expense, underwriting expense and policyholder dividend ratios. Loss-related ratios generally use earned premium; the underwriting expense ratio uses written premium. NCCI’s 2026 guide sets out the mechanics.

Below 100% is an underwriting profit on the reported basis. Above 100% is a loss. Investment returns sit outside it entirely.

Reading 93% as a 7% profit margin skips tax, investment results, other income and the accounting differences inside the calculation. It also skips claims maturity, which is the larger omission.

A calendar-year ratio absorbs changes in estimates for older claims. An accident-year measure groups losses by when the events happened. An underlying ratio strips out catastrophes and prior-year development, according to whoever is publishing it. Put the label next to the number, because two insurers can quote the same ratio while describing different views of the risk they are writing today.

What is a typical expense ratio for property casualty insurance?

For the aggregate US industry the benchmark is about a quarter of net written premium. The NAIC reported underwriting expense ratios of 24.9% in 2023, 25.3% in 2024 and 25.8% in 2025.

That answers the question at industry level. It sets no target for an individual carrier.

A high-volume direct motor insurer, a broker-distributed commercial carrier and a specialist writing technically demanding risks run different acquisition and operating models. A smaller insurer also spreads fixed costs across a narrower premium base.

So an expense comparison has to hold constant the commission structure, the product complexity, the scale and growth stage, the reinsurance arrangements and whether the figure is an underwriting expense ratio or a broader operating-cost measure. Loss adjustment expenses need particular care: they belong with claims costs in the combined-ratio framework, and adding them to the expense ratio counts them twice.

On a hypothetical $100 million net written premium base, one point of expense ratio is $1 million, holding the denominator constant. That is arithmetic, not a projection of what a technology programme will save.

The management question is where the point would come from: fewer repeated data entries, cheaper distribution, less rework, a different mix. Cutting cost without understanding the activity tends to damage the underwriting result it was meant to improve.

Catastrophe volatility still matters after a profitable year

Swiss Re Institute estimated $107 billion in global insured natural catastrophe losses for 2025, with secondary perils, meaning wildfires, severe convective storms and floods, contributing 92% of the total. Swiss Re published the figure in March 2026.

Those are global catastrophe estimates, not total P&C claims and not calendar-year cash payments.

Inside the US result, one factor did most of the work.

“A near 90 percent decline in hurricane-related claims in 2025 materially reduced catastrophe losses”
Saurabh Khemka, president of Underwriting Solutions, Verisk

He attributed it to limited US landfall. Which is to say the improvement was weather, not underwriting.

Concentration in hail-prone counties, vulnerable construction, understated replacement values and a shifting wildfire footprint are all invisible in a quiet year and expensive in a loud one. The exposure data is what makes them visible: correct locations, current values, reliable construction detail and a clear view of accumulation across the portfolio. The same point runs through the 2026 climate reports for insurers.

Workers’ compensation shows why calendar year and accident year differ

NCCI reported a 91% calendar-year combined ratio for private workers’ compensation carriers in 2025, a twelfth consecutive underwriting gain. The 2025 accident-year combined ratio was 102%.

Eleven points, same book, same year.

The same report recorded a 2% fall in lost-time claim frequency alongside 4% increases in both medical and indemnity severity. Fewer claims, each one costing more, with favourable development on older years holding up the calendar-year number.

For reserving and pricing teams the task is separating the part of the result that reflects this year’s exposure from the part that reflects a changed opinion about the past.

The 2026 pricing trend is softer, but casualty is the exception

Marsh’s Q2 2026 index recorded these average renewal rate changes across its global commercial client portfolio.

US casualty rates rose 7%, while casualty fell in every other region covered.

These are renewal-pricing observations across Marsh’s client mix, not changes in total premium written, and any individual insured’s outcome varies with exposure, limits, retentions and loss history.

The portfolio implication is blunt. An instruction to grow commercial lines means one thing in property at -12% and something quite different in casualty at +2%, which is the two-speed soft market in a single table.

Swiss Re’s July 2026 world insurance research forecasts 0.6% global real non-life premium growth in 2026 and 1% in 2027, expecting competitive pricing, plentiful capital and easing reinsurance costs to keep markets soft into 2027. That is a real-terms forecast for the broader non-life category, and it should not be applied mechanically to the nominal market-size estimate above.

Slower premium growth raises the value of knowing exactly where profit comes from. A book carried by successive rate increases needs a different plan once competition takes them away.

Turning industry statistics into operating decisions

Industry figures earn their keep when they prompt sharper questions about your own book. Where is growth actually coming from? Which lines show improving current-year loss experience? How much of the underwriting result depends on reserve releases? Are acquisition and servicing costs moving faster than premium?

Answering those needs premium, exposure, claims, expense and distribution data connected at the level where decisions get made, with the gross and net basis, the accident and calendar year, and the observed and forecast figures kept apart rather than blended.

Most carriers can produce each piece; far fewer can produce them together on the same insurance business dashboard without a month of spreadsheet work. Openkoda exists for that gap, with configurable products and pricing over a policy, claims, billing and reporting core that you own outright.