Specialty Insurance Statistics and Insights 2026

Specialty insurance does not move in one direction. A growing premium base can sit beside deteriorating underwriting results. A quiet catastrophe year can coexist with substantial uninsured exposure. Better accident frequency can still leave insurers facing severe individual losses.

That is why useful specialty lines statistics need to be read at product level.

This article examines eight areas: critical illness, disability, medical malpractice, flood, title, professional liability, crop and aviation insurance. It uses specialty insurance products broadly, to include focused protection products and specialist risk classes. These products span life and health as well as property and casualty; they do not form a single, consistently defined global market.

The figures below reflect research available in September 2026. Most describe 2025 experience, with newer 2026 market indicators and clearly dated historical comparisons where useful.

Specialty lines statistics: a quick view

These statistics should not be added together. They include premiums, paid claims, estimated insured losses, renewal pricing and accident frequency. Their value lies in the different questions they help answer.

Protection lines: the benefit definition is what gets tested

Aviva’s March 2026 disclosure provides a recent insurer-level example. In 2025, its UK individual critical illness, children’s benefit and total permanent disability category paid 5,586 claims, with a 90.7% acceptance rate. The company said most declines in this category occurred because the diagnosis did not meet the policy definition. Source: Aviva’s 2025 protection claims release.

The category matters. These are neither critical-illness-only figures nor an industry-wide UK acceptance rate.

For product teams, the practical issue is the relationship between the condition customers believe they have insured and the contractual trigger. The diagnosis, severity threshold and applicable wording must be visible throughout the claim assessment.

An overall protection payout percentage is too broad for that work. Break critical illness reporting down by condition, product generation, benefit type and reason for decline. A recurring disagreement over one definition should be identifiable without searching manually through hundreds of files.

Claims experience can then inform both the next product version and how existing coverage is explained.

Disability insurance statistics: premium growth and policy growth tell different stories

Gen Re’s 2025 U.S. Individual Disability Market Survey, published in May 2026, covered 16 carriers. Participants reported $491.8 million in new sales premium, up 2.7%, and 285,377 new policies, up 0.7%. In-force premium grew 1.9%, while the number of in-force policies remained around 3 million. Source: Gen Re.

Premium growth outpaced growth in new policy count. That supports a closer look at average premium and business mix; it does not, by itself, prove that prices increased or that more workers obtained meaningful coverage.

The survey covers individual disability product categories and should not be presented as a combined total for employer group disability and Social Security Disability Insurance.

Claims reporting needs similar care. An income protection claim can involve recurring benefits, changing capacity for work and rehabilitation support, so the count of payments is not necessarily the count of people making new claims.

For example, Aviva separately reported £63.6 million in UK individual income protection benefits paid to 4,154 new and existing claimants in 2025, with a 90% acceptance rate. Source: Aviva.

For disability insurers, a useful dashboard therefore pairs new-claim incidence with duration, recurring benefit payments, rehabilitation activity and return-to-work outcomes. Annual payout value alone cannot show whether claims are becoming more frequent, lasting longer, or carrying higher monthly benefits.

Liability lines: payment pressure and softer pricing at once

Milliman’s 2025 analysis followed a composite of 160 U.S. medical professional liability specialty companies. Their direct MPL payments reached $4 billion, more than 14% above 2024. The composite’s combined ratio was 108.6%, marking its ninth consecutive year of underwriting losses. Source: Milliman, April 2026.

The composite is a defined group of specialty writers, rather than every insurer writing medical malpractice. Some participants also write other lines, which is relevant when interpreting company-level financial results.

For claims and actuarial teams, the combination of rising payments and an underwriting deficit warrants attention to severity, development and defense costs. A payment made in 2025 may relate to care delivered and a claim reported years earlier.

Calendar-year payments therefore should not be divided by newly reported claims and described as the average cost of a new malpractice case.

A stronger view tracks paid and incurred severity by maturity, alongside claim frequency, closure patterns and reserve development. That makes it easier to identify whether current profitability rests on improving underwriting, favorable development, investment returns, or a mixture of the three.

The distinction matters in a long-tail line where a profitable financial year and an underwriting profit are not interchangeable.

Flood insurance statistics: a lower loss year does not remove the protection gap

Swiss Re Institute estimated global flood-related insured losses at $3.4 billion in 2025, compared with a $15.4 billion average over the preceding five years. These are estimated losses, not simply cash claim payments made during the calendar year. Source: Swiss Re Institute, March 2026.

A lower annual total deserves context. Flood insurance statistics measure insured exposure as well as the underlying hazard: low insurance penetration can leave substantial economic damage outside claims datasets.

Commenting on natural catastrophe results more broadly, Swiss Re’s Head Catastrophe Perils, Balz Grollimund, said:

“The below-trend natural catastrophe losses seen in 2025 are the result of favourable variability rather than any easing of underlying risk.”

Source: Swiss Re.

U.S. NFIP data adds a longer-term perspective. FEMA’s August 2025 guidance says almost one-third of NFIP claims from 2014–2024 came from outside current high-risk flood areas. It also reports an average flood claim payment of almost $64,000 for 2020–2024. Source: FEMA’s flood insurance guide for real estate professionals.

For distribution teams, the implication is that a flood-zone label should not become the entire coverage conversation. For underwriting teams, portfolio analysis should consider location-level characteristics and accumulation, rather than assuming a recent quiet year represents the range of future outcomes.

Title insurance claims statistics: evaluate the claims and the work before issuance

The U.S. title insurance industry generated $18.5 billion in premiums in 2025, an increase of 13.8%. Claims payments exceeded $667 million, compared with $676 million in 2024. Source: ALTA’s 2025 market-share and premium report, May 2026.

Those figures show premium growth alongside relatively stable claims payments. They do not mean that the difference between premiums and paid claims is profit.

Title operations also involve examining ownership records and addressing problems before a policy is issued. A financial assessment must account for that work, distribution and operating expenses, as well as claim costs and their timing.

There is also a cohort problem. Claims paid this year can arise from policies issued in earlier years, so comparing current cash payments with current premiums is not a measure of the eventual profitability of this year’s business.

For insurers and settlement partners, useful operational reporting links a claim back to the transaction, policy, relevant documents and issue that caused the loss. That can help distinguish a recurring process problem from an isolated defect - and identify opportunities to prevent similar claims before the next closing.

Professional liability insurance statistics: softer pricing does not establish lower risk

Marsh’s Global Insurance Market Index reported a 3% decrease in financial and professional lines renewal rates in Q2 2026, following a 5% decline in Q1. The index reflects Marsh’s client portfolio and combines multiple financial and professional products. It is not an E&O-only market-size or claims-frequency statistic. Source: Marsh Global Insurance Market Index.

That distinction is especially useful when considering professional liability insurance statistics. A composite pricing movement can conceal different conditions for architects, engineers, lawyers and other professions.

For architects and engineers, the NSPE’s account of its joint 2025 carrier survey describes a competitive market alongside concerns about increasing claim severity, particularly in complex, multi-party projects. Source: National Society of Professional Engineers.

For an underwriter, the question is whether the account’s terms and price reflect its own exposure. Compare profession, project type, contract scope, jurisdiction, limits and retention before treating a market-wide rate decrease as the appropriate renewal outcome.

Claims feedback is particularly valuable here. Allegations involving advice, design, documentation or scope of services can help identify patterns that a premium trend will never reveal.

Crop insurance statistics: use the crop year and the valuation date

USDA’s Risk Management Agency offers unusually detailed public reporting. Its September 7, 2026 Summary of Business showed the following for the 2025 crop year, excluding livestock business:

Source: USDA RMA Summary of Business, page 1, combined non-livestock business. The source reports dollar values in thousands; the table converts them to billions.

The valuation date belongs beside the crop year. Reporting continues to develop after harvest, so a 2025 crop-year figure retrieved in September 2026 can differ from an earlier published estimate.

The loss ratio also needs context. It compares indemnities with total premium under the program’s reporting basis. It is not an insurer combined ratio or a complete measure of participating carriers’ profitability.

Crop insurance statistics are particularly useful when broken down by crop, location, plan and cause of loss. A national aggregate can conceal very different experiences across regions and products.

For teams building reports from RMA data, save the retrieval date and dataset version. Comparing one fully developed year with a partial current crop year can create a misleading improvement simply because claims have not yet developed.

Aviation insurance statistics: frequency and severity can move apart

IATA recorded 51 accidents across 38.7 million commercial flights in 2025, an all-accident rate of 1.32 per million flights, compared with 1.42 in its revised 2024 data. Yet 2025 also recorded eight fatal accidents and 394 onboard fatalities, versus seven and 244 in 2024. Source: IATA’s 2025 Safety Report release.

These are aviation safety statistics. They provide exposure context for insurance, but they do not measure insured hull losses, liability settlements or total claims costs.

The distinction is essential. An accident count gives equal weight to events that may have very different consequences for an insurer. Aircraft values, damage extent, liability exposure and policy participation determine how an event translates into a loss.

General aviation also needs separate treatment from commercial airline operations. Gallagher’s Q2 2026 general aviation update describes continuing competition for well-managed risks, with more selective underwriting where loss histories or geopolitical exposures are challenging. Source: Gallagher Specialty, July 2026.

For aviation insurers, a useful portfolio view combines operational exposure, asset values, coverage and geography with claims development. Improving fleet-level safety cannot substitute for understanding the accumulation created by individual aircraft, airports or territories.

What specialty insurers should take from the 2026 evidence

The common thread across these specialty lines statistics is the need for more specific portfolio information.

Separate volume from severity. More premium does not necessarily mean more policies. Higher claims payments do not necessarily mean more new claims. Identify which component changed before deciding how to respond.

Keep the contract connected to the claim. A critical illness definition, a professional liability policy trigger and a crop insurance plan each require different evidence. Product version, endorsements and decision history should be available when the claim is assessed and when its outcome is analyzed.

Make dates part of the data. Crop-year development, long-tail liability payments and current-quarter renewal pricing cannot be compared as though they describe the same period. Recording the observation date makes reporting reproducible.

Use automation around the specialist work. Document collection, data extraction, task routing and reporting are practical areas to improve. Evaluate them against fewer errors, less rework and better access to evidence, while preserving the specialist judgment each line requires.

For insurers and MGAs bringing these processes together, Openkoda supports configurable products and workflows alongside policy, claims and reporting functions. The aim is to give each line the information and process it needs while retaining a consistent view of the wider portfolio.

Specialty insurance rewards detail. The strongest benchmark is one that helps an underwriter, claims specialist or product manager ask a better question about the business they actually write.