Insurance Market Analysis: Premiums, Claims, and Loss Ratios

Insurance is a business built on estimates that later meet reality. A premium is a promise, a claim is the bill, and the loss ratio is where the two shake hands. Read all three together or you are reading marketing.

Market Survey Analysis view: This guide is built for the decision of whether to trust an insurance market number enough to act on it, whether that means entering a line, repricing a book, or backing a growth forecast for an insurer or a segment. Start with the boundary, then test the evidence chain. For related market intelligence and research workflows, keep the definition, source and decision in one review record.

How to read the result

Read every insurance figure in four layers. First, the definition: is this written premium, earned premium, or net premium after reinsurance. Second, the source and period: a regulatory filing, a rating agency report, or a trade association estimate, and the accident year or calendar year it covers. Third, the segment: personal lines, commercial lines, one state, one country, one product. Fourth, the decision the number is meant to support. Skip any of the four and the number becomes a slogan.

A loss ratio travels fast once it leaves its filing. It moves from a regulator's table into a slide, then into a pricing memo, and along the way it tends to lose its line of business, its accident year, and its confidence interval. Track a headline number back to where it started before you let it move a plan.

Written premium vs earned premium

Written premium is the amount an insurer books when a policy is issued or renewed. Earned premium is the portion of that amount the insurer has actually held over the coverage period so far. A twelve-month policy sold in November has earned only a sliver of its premium by December. Growth headlines built on written premium can outrun the business the insurer has actually carried risk for.

Decision-check: when a market report claims strong premium growth, confirm whether the figure is written or earned, and whether it is gross of reinsurance or net. The gap between the two tells you how much of the growth is booking activity and how much is risk actually held.

Loss ratio and combined ratio

The loss ratio compares claims paid and reserved against premium earned. The combined ratio adds underwriting expenses to that picture, so it captures the full cost of running the business, not just the cost of claims. A combined ratio under 100 means the insurer made an underwriting profit before investment income; above 100 means underwriting alone lost money, and the company is relying on investment returns to close the gap.

Decision-check: before comparing two insurers or two lines, check whether the loss ratio quoted is incurred (claims paid plus reserves set aside) or paid only. A paid-only ratio understates cost for lines where claims take years to settle, such as liability or workers' compensation.

Underwriting cycle: hard market vs soft market

Insurance pricing moves in cycles. In a hard market, capacity tightens, prices rise, and terms get stricter, usually after a period of heavy losses or reduced reinsurance capacity. In a soft market, competition intensifies, prices fall, and terms loosen. A market analysis that reports rising premiums without naming the cycle phase risks reading a temporary hardening as permanent structural growth.

Decision-check: ask where the market sits in its cycle before projecting current premium trends forward. A forecast built during the peak of a hard market and extended five years out is a common way plans get built on a number that was never going to hold. Our guide on how to read a market forecast before you use it covers this pattern in more depth.

Claims frequency vs severity

Frequency is how often claims happen. Severity is how much each claim costs once it happens. A rising loss ratio can come from more claims, costlier claims, or both, and the fix for each cause is different. Rising frequency often points to underwriting selection or exposure growth. Rising severity often points to inflation in repair or medical costs, or to a shift in the mix of claims toward larger, more complex events.

Decision-check: when a loss ratio is flagged as deteriorating, ask for the frequency and severity trend separately, not just the combined result. Pricing and underwriting actions that address frequency will not fix a severity problem, and the reverse is also true.

Reinsurance and catastrophe exposure

Reinsurance is insurance for insurers. It spreads large or concentrated losses, particularly from catastrophe events, across other carriers. A primary insurer's own loss ratio can look calm in a bad year simply because reinsurance absorbed the peak losses, and it can look volatile in a quiet year if reinsurance costs rose sharply after a prior bad year elsewhere. Reading a loss ratio without knowing the reinsurance structure behind it means reading half the picture.

Decision-check: for any line exposed to catastrophe risk, such as property or crop insurance, check whether the reported figures are gross or net of reinsurance, and whether reinsurance pricing itself is rising or falling in that period. Governance around who can access and cite which version of a claims figure matters here; our note on market research governance covers how to keep claims data and access aligned across a team.

Distribution channel mix: agent, broker, and direct

How a policy reaches a customer, through a captive agent, an independent broker, or a direct digital channel, shapes acquisition cost, retention, and often the loss profile of the book itself. Direct channels tend to carry lower distribution cost but can carry different risk selection than an experienced broker's book. A market share number that does not break down by channel hides which distribution model is actually driving growth or margin.

Decision-check: when comparing insurers on growth or profitability, check whether the difference tracks to underwriting skill or simply to a different distribution mix. The two get confused often, and they call for very different responses.

A practical evidence table

This table is a frame for organizing evidence, not a scoreboard to fill in with the first number you find. Each cell is a prompt to go back to a named source, a named period, and a named segment before writing a figure into it. A table with confident numbers and no named sources is worse than an empty one, because it looks finished when it is not.

What this analysis does not prove

A snapshot of premiums and loss ratios tells you where a book of business stood over a defined period. It does not tell you what will happen at the next renewal cycle, after the next catastrophe season, or once a regulator changes a rate filing rule. Insurance results are backward-looking by nature, built from claims that may still be developing, and treating a current loss ratio as a fixed input for a five-year plan is a common way otherwise careful analysis goes wrong.

It also does not prove causation. A falling loss ratio might reflect better underwriting, a milder claims year, a change in reserving assumptions, or a shift in the business mix toward less risky policies. Naming the true driver takes more digging than the top-line ratio alone can offer, and a market report that skips this step is handing you a conclusion, not evidence.

Review checklist before publication

• Confirm whether premium figures are written or earned, and gross or net of reinsurance.
• Confirm whether loss and combined ratios are incurred or paid, and for which accident year.
• Name the underwriting cycle phase the data was collected in.
• Separate claims frequency trends from severity trends wherever both are cited.
• Note the distribution channel mix behind any market share or growth claim.
• Trace every headline figure back to a named, dated source before it goes into a brief.

Frequently asked questions

What is a good loss ratio for an insurance line?

There is no single good number; it depends on the line and the expense structure behind it. A line with low distribution and administrative costs can sustain a higher loss ratio and still be profitable, while a line with high acquisition costs needs a lower loss ratio to reach the same result. Compare the combined ratio, not the loss ratio alone.

Why do insurers report both gross and net loss ratios?

Gross figures show total claims before reinsurance recoveries; net figures show what the insurer actually retained after reinsurance. The gap reveals how much catastrophe or large-loss risk has been transferred elsewhere, which matters for judging an insurer's real exposure.

How often should premium and claims data be reviewed for a market analysis?

At minimum, at each renewal or filing cycle for the line in question, since claims reserves develop over time and early figures often change. A single snapshot without a follow-up review invites stale conclusions.

Does a rising combined ratio always mean an insurer is losing money?

Not necessarily. A combined ratio above 100 means an underwriting loss, but many insurers offset this with investment income on the premium they hold before claims are paid. Check both figures before judging overall profitability.

Where can I find reliable insurance market statistics?

Start with regulatory filings, rating agencies, and established industry associations, and always note the period and segment covered. For a broader library of research method notes, see the Market Survey Analysis blog.

Sources and method notes

This guide draws on standard insurance industry reporting conventions and general research governance practice. It is a framework for reading and testing insurance market claims, not a substitute for the primary filing or report behind any specific figure.

• Insurance Information Institute Industry Overview. U.S. property-casualty and life insurance industry facts and statistics. Checked on 2026-09-19.
• ICC/ESOMAR International Code 2025. Ethics, transparency, accountability, privacy and human oversight in research and analytics. Checked on 2026-09-19.
• Office for National Statistics Methodology. Official-statistics production, quality, concepts and methodological practice. Checked on 2026-09-19.

Next step

Use this framework to define a focused brief, test the evidence and identify the next decision. If the boundary or source base needs work, request a custom research discussion rather than forcing a weak number into a plan.