How P&C Insurance Float Creates a Structural Advantage

Underwriting profit is not what makes property and casualty insurance one of the best capital-allocation engines in finance. Float is. A P&C insurer collects premiums today for claims it might not pay for years, and in the gap between the two, that money belongs to nobody in particular — the policyholder hasn't been paid yet, and the insurer hasn't earned it yet either. It just sits there, invested, compounding.

I've found that investors who fixate only on combined ratio miss the more important number sitting one line down: how much of other people's money an insurer gets to invest, and for how long. A carrier with a mediocre combined ratio but a large, steadily growing float can still be a superb business. A carrier with a pristine combined ratio and a shrinking premium base is compounding a smaller and smaller pile of capital. Our income-investing coverage lays out the basic mechanics of float in how insurance companies invest; the focus here is different — how the three competitive roles inside the sector actually work, not portfolio lessons for individual holders.

That distinction sorts the P&C industry into competitive roles that look similar on the surface — everyone writes policies, everyone pays claims — but run on different economics underneath. Specialty underwriters compete on pricing discipline in hard-to-place risk. Large diversified carriers compete on scale and distribution. Reinsurers compete on balance-sheet capacity to absorb losses nobody else will take. Float is the thread that runs through all three, but it behaves differently in each.

Float: The Structural Core of P&C Economics

Float is the sum of premiums an insurer has collected but not yet paid out as claims, net of the money it's already spent acquiring and servicing the business. It shows up as a liability on the balance sheet — loss reserves, unearned premium reserves — but it functions as investable capital. The insurer holds it, invests it in bonds and equities, and earns a return until a claim comes due. Berkshire Hathaway is the cleanest public illustration of what happens when float compounds for decades. Per its 2025 annual report, Berkshire's insurance float stood at approximately $176 billion at year-end 2025, up from $171 billion at year-end 2024 and from $88 billion at year-end 2015. That's not underwriting profit sitting idle. It's an investment portfolio, funded largely by other people's premiums, that's grown alongside the businesses it helped Berkshire buy.

The cost of that capital is set by the combined ratio — the ratio of losses plus expenses to earned premiums. A combined ratio under 100% means the insurer is effectively being paid to hold float; premiums exceed claims and expenses, so float has a negative cost of capital before a dollar of investment income shows up. A combined ratio above 100% means float costs something, and the insurer needs investment returns to make the overall economics work. Berkshire's insurance operations posted an 87.1% combined ratio for full-year 2025 — float that got cheaper to hold even before the investment portfolio did anything at all.

Three Roles, One Float Mechanism

Group insurers by what they actually compete on, the same way U.S. banking sector structure groups lenders by competitive role, and the float mechanism looks different in each bucket even though the balance-sheet math underneath is identical.

• Specialty and niche underwriters — Kinsale Capital Group. Writes excess-and-surplus lines that standard carriers won't touch, pricing hard-to-place risk with underwriting discipline as the entire value proposition. Kinsale's FY2025 combined ratio was 75.9%, down from 76.4% in FY2024, with gross written premium up 5.7% to $2.0 billion and net investment income up 27.9% to $192.2 million — a small, fast-growing float earning an unusually cheap cost of capital.
• Large diversified carriers — Progressive and Berkshire Hathaway's primary insurance operations. Compete on scale, data, and distribution across enormous, granular books (personal and commercial auto, homeowners). Progressive's companywide combined ratio was 89.0% for FY2025, with an underwriting profit margin of 12.6%, up from 11.2% in 2024. The float base is orders of magnitude larger than a specialty writer's, and the edge comes from pricing precision applied at massive scale rather than from avoiding hard risk altogether.
• Reinsurers — RenaissanceRe Holdings. Sits above primary insurers, absorbing tail risk that specialty and diversified carriers lay off rather than hold themselves. RenaissanceRe posted an 87.2% combined ratio for FY2025 (85.4% on an adjusted basis) and $1.3 billion of underwriting income, even after large-loss events — including Hurricane Melissa — added 15.3 points to the ratio. Its float is the most volatile of the three, and its investment book, $3.0 billion in total investment results for the year, is what absorbs that volatility.

None of these roles is fixed in place. Combined ratio moves with the underwriting cycle as much as with company-specific discipline. A hard market — shrinking reinsurance capacity, rates rising across the industry — tends to pull combined ratios down for almost everyone at once. A prolonged soft market does the opposite: abundant capacity chases the same premium dollars, pricing discipline erodes, and ratios drift up industry-wide regardless of any individual underwriter's skill. That cycle is turning right now. Property catastrophe rates fell roughly 16% at the July 2026 midyear reinsurance renewal as record dedicated capital chased a smaller pool of demand, and RenaissanceRe's own Q2 2026 results showed gross premiums written down 12.5% year over year to $3.0 billion — a deliberate pullback from business it no longer found attractively priced, not a demand problem. That's a cyclical headwind or tailwind sitting on top of each company's structural role, and it's worth separating from the company-specific discipline that determines who still beats peers within whatever part of the cycle the industry happens to be in.

Markel Group complicates the tidy three-way split, and deliberately so. Its insurance segment posted a 95% combined ratio on $9,353 million of operating revenue in FY2025 — closer to a diversified carrier's economics than a specialty underwriter's — but the float it generates funds Markel Ventures, a permanent-capital operating-company portfolio built explicitly on the Berkshire model. Same mechanism, different destination for the capital.

Combined Ratio and Float Growth, Read Together

Here's the trap: a low combined ratio in isolation tells you underwriting was disciplined this year. It doesn't tell you whether the float pool behind that discipline is big enough, or growing fast enough, to matter. Kinsale's 75.9% combined ratio is excellent. But its net written premium — $1.6 billion in FY2025, up 9.4% — is a rounding error next to a carrier the size of Progressive. Cheap float on a small base still compounds a small base.

Reserve development is the other variable that quietly distorts the headline number. Markel's FY2025 results included $484.0 million of favorable development on prior-accident-year loss reserves, up from $454.9 million in FY2024 — money released back into current-year income because claims came in cheaper than originally reserved. That's real, but it's backward-looking. It reflects how conservative the reserving was two or three years ago, not how well the current book is being priced. So a combined ratio benefiting heavily from reserve releases isn't necessarily a signal that this year's underwriting is better than last year's. It might just mean the actuaries were cautious in 2022 or 2023.

Where the Model Breaks: Cat Losses and Float Risk

Float is only a structural advantage as long as the insurer isn't forced to liquidate its investment portfolio at the wrong moment. RenaissanceRe's FY2025 combined ratio absorbed 15.3 points of large-loss impact — Hurricane Melissa alone added 4.0 points — and the company still finished the year with underwriting income. That's the model working as designed: the float base and the balance sheet were large enough to absorb a genuinely bad catastrophe year without needing to sell assets into a falling market to pay claims.

It doesn't always work that way. Markel absorbed a smaller, more contained hit — roughly $61.9 million from the Southern California wildfires early in 2025 — inside a more diversified specialty book, which is a very different risk profile than a reinsurer whose entire business model is concentrating tail risk other insurers won't hold. Concentrated cat exposure is the mechanism by which float, an asset in ordinary years, becomes a liability in a bad one: claims accelerate exactly when a reinsurer's bond and equity portfolio is most likely to be under pressure from the same macro shock. It's not so different from what a bank stress test is checking for on the other side of financials — can the balance sheet survive the tail scenario, not just the average one.

There's a duration risk layered on top that gets less attention. Float invested in fixed income locks in a yield at the time of purchase; if claims payouts accelerate while the bond book is still earning yesterday's lower rates, the float's investment return lags what a fresh dollar could earn today. I'm less confident than I'd like to be about how exposed any single insurer is to this, since duration and asset-allocation detail lives deep in the investments footnote of each 10-K and varies name to name — Berkshire and Markel run meaningfully more equity-heavy float portfolios than a reinsurer like RenaissanceRe, which skews toward fixed income to match the shorter-tail nature of catastrophe claims.

How to Read Float Economics When Evaluating a P&C Insurer

A practical framework, in the order I'd actually apply it:

1. Look at combined ratio over five-plus years, not one. A single good year tells you little about underwriting discipline through a full cycle.
2. Strip out prior-year reserve development to see the accident-year combined ratio — the closest approximation of how this year's business is actually being priced.
3. Track float growth alongside net written premium growth. Cheap float on a shrinking or stagnant base isn't a structural advantage; it's a shrinking advantage.
4. Check the investment portfolio's duration and equity allocation. That's where float actually earns its keep, and it's also where the risk in a rate or equity shock shows up.
5. For cat-exposed underwriters and reinsurers, size the position for a bad year's combined ratio, not an average one. It's the insurers that can absorb a 15-point cat hit without touching the float base underneath that are actually built to last through a full cycle.

None of this replaces reading the actual 10-K. But it's the difference between seeing a combined ratio and understanding the industry structure that produced it.

Key Takeaways

• Float — premiums collected before claims are paid — is the structural core of P&C insurance economics, not a footnote to underwriting profit.
• Combined ratio sets the cost of float; float scale and growth determine how much that cheap capital actually compounds. Read both together.
• Specialty underwriters like Kinsale run the cheapest float on the smallest base; diversified carriers like Progressive and Berkshire run larger, slower-growing float at a higher but still-favorable cost; reinsurers like RenaissanceRe carry the most volatile float and the largest tail risk.
• Reserve development can flatter a single year's combined ratio without reflecting current underwriting quality — check the accident-year ratio, not just the reported one.
• Float is an advantage in ordinary years and a liability in a genuinely bad one if the insurer has to sell investments to pay claims at the worst possible moment.