A return-on-equity spread has two terms. AM Best's global reinsurance composite earned a median 16.3% on equity in 2025 against a 9.6% cost of equity, a 6.7-point cushion and marginally below the record 2023 result (AM Best, August 2026). The second term is the one that moved.
Cost of equity rose for a fourth consecutive year. The weighted average cost of capital went from 7.67% in 2024 to 8.23% in 2025, then to 8.63% in the first quarter of 2026. That last step, 40 basis points in three months, landed while the Guy Carpenter Global Property Catastrophe Rate-On-Line Index was on its way to a 16% decline for the year.
Key Takeaways
- 6.7 points separate the composite's 16.3% median return on equity from its 9.6% cost of equity. The return is a near-record; the cost of equity is a four-year high. Only one of those two facts gets reported.
- 96 basis points of weighted average cost of capital increase between 2024 and the first quarter of 2026, from 7.67% to 8.23% to 8.63%. Forty of those basis points arrived in a single quarter.
- 4.77% on the 10-year Treasury in early September 2026, roughly 69 basis points above a year earlier, against a federal funds target of 3.50% to 3.75%. The curve steepened, so the leg that anchors a cost of equity was rising.
- 11.4% is Gallagher Re's cost of equity for its own reinsurance composite, 1.8 points above AM Best's 9.6%. The hurdle quoted as a fixed benchmark is a modelling choice worth several points of apparent cushion.
- Under 1.0% gross premium growth for AM Best's US and Bermuda composite in 2025, down from 11.7% in 2024, with the combined ratio deteriorating to 90.2 from 88.5. The return held because the book stopped growing.
Two Rate Legs, Moving Apart
AM Best's summary is that the higher cost of equity raised the cost of capital despite lower interest rates. That is true of the policy leg. The Federal Reserve cut through 2025 and has since held the federal funds target at 3.50% to 3.75% for five consecutive meetings.
It is not true of the leg that discounts an equity claim. The 10-year Treasury closed at 4.77% on 3 September 2026 (FRED, September 2026), roughly 69 basis points above where it sat a year earlier. Short rates fell, long rates rose, and the curve steepened between them.
That changes the attribution. A cost of equity is built off a long risk-free rate, not the overnight target, so across the most recent leg of the climb the risk-free component was pushing the number up by itself. The common read, that a rising cost of equity must mean investors repricing reinsurance earnings volatility, assumes a falling risk-free rate that the long end does not show.
AM Best makes the same point in passing elsewhere, describing reinsurers accumulating capital "in the current relatively high-interest rate environment" (Artemis, August 2026). High and falling are different claims about different tenors.
The scale of the four-year move is clearest against this site's earlier coverage of the same AM Best work, when the CAPM median cost of equity for global reinsurance sat near 7.5% and economic value added was compressing toward 6.7%. The reported figure is now 9.6%.
The Hurdle Rose in the Quarter the Price Fell
Take the arithmetic literally. Using Aswath Damodaran's start-of-year inputs, a 4.18% risk-free rate and a 4.23% implied equity risk premium (Damodaran, January 2026), a 9.6% cost of equity only arrives at a beta near 1.28. AM Best does not publish its build, so that is a statement about magnitude rather than a reconstruction of its method: the market is pricing reinsurance equity as roughly a quarter more volatile than the index.
A hurdle rate is not a scorekeeping device. It is the discount rate sitting under a pricing decision, and it sets the underwriting margin a treaty has to clear before it adds economic value. Gallagher Re puts the premium-to-capital ratio for reinsurance groups at 86%, so 96 basis points of additional capital cost is worth roughly 1.1 points of combined ratio. Against the 9.8 points of underwriting margin implied by a 90.2 combined ratio, that is about an eighth of the margin, consumed before a single loss pick is revisited.
The timing is the awkward part. Guy Carpenter's global index fell 12% at January 1 and reached 16% down for the year after the April and mid-year renewals, with Asia-Pacific down 19% and Europe down 15% (Artemis, June 2026). The US index posted its steepest annual fall since 2014, the global index its steepest since the late 1990s, and the 40-basis-point cost-of-capital step sits inside that window.
Supply explains the direction. Aon put global reinsurer capital at a record $800 billion at 30 June 2026, up $15 billion in the half, with an annualised sector return on equity of 15.5% (Global Reinsurance, September 2026). Capital sets the marginal price, and it is still growing into a falling rate-on-line while the return that capital demands goes up.
Which Hurdle, and Against Which Return
The spread is quoted as though both terms were observed. Neither is.
Gallagher Re puts the cost of equity for its own composite at 11.4%, 1.8 points above AM Best's 9.6%, and its underlying return at 13.8% once catastrophe experience, prior-year development and investment gains are stripped out. On that pair the cushion is 2.4 points, not 6.7.
| Return measure | Source | ROE | Spread vs 9.6% | Spread vs 11.4% |
|---|---|---|---|---|
| Median ROE, global composite, 2025 | AM Best | 16.3% | 6.7 pts | 4.9 pts |
| Aggregate net ROE, US and Bermuda, 2025 | AM Best | 16.8% | 7.2 pts | 5.4 pts |
| Annualised sector ROE, H1 2026 | Aon | 15.5% | 5.9 pts | 4.1 pts |
| Reported ROE, H1 2026 | Gallagher Re | 19.9% | 10.3 pts | 8.5 pts |
| Underlying ROE, H1 2026 | Gallagher Re | 13.8% | 4.2 pts | 2.4 pts |
The composites and periods differ, so the rows are not strictly comparable with each other. The point is the range: the same sector, in the same year, shows a cushion anywhere between 2.4 and 10.3 points depending on which return definition and whose hurdle is used. The site covered the return half of that spread when Gallagher Re's 19.9% resolved to 13.8% underlying; the hurdle half moves too, and moves in the wrong direction.
The return term has a second problem. AM Best's US and Bermuda composite of seven groups, including Arch Capital, Everest, General Re, RenaissanceRe and Transatlantic, held aggregate net return on equity at 16.8% across both 2024 and 2025. Gross premium growth over the same span fell from 11.7% to under 1.0%, and the combined ratio deteriorated to 90.2 from 88.5 (Reinsurance News, September 2026).
That deterioration was produced with lower catastrophe losses and greater favourable reserve development than the prior year, which means the accident-year margin eroded by more than the 1.7 points the headline shows. The return was held flat by declining business, not by underwriting better.
Greg Dickerson, a director at AM Best, expects "top-line growth for the composite to be similarly muted in 2026, given the acceleration of rate decreases seen in property reinsurance and a slowing of price improvement in US casualty lines after several years of strong gains" (Reinsurance News, September 2026).
Declining to write business is a lever that works once. Fixed expenses do not fall with premium, so holding volume flat through another 16% of rate reduction puts the expense ratio to work against the same margin the cost of capital is already eating. The hurdle has risen in each of the last four years, and the composite has now spent one of its two defences against it.
Further Reading on actuary.info
- Soft Cycle Could Push Reinsurers Below Cost of Capital by 2027 - the earlier reading of the same AM Best cost-of-capital work, when the CAPM cost of equity sat near 7.5%.
- Gallagher Re: Reinsurers' 19.9% Return Is 13.8% Underlying - the return side of the same spread, decomposed line by line.
- Property Cat Rate-on-Line Down 16% and the Net Cost of Reinsurance - what the falling index does to the primary carriers buying the cover.
- AM Best: Record Capital Tests Reinsurance's 87% Combined Ratio - the reserve-cushion question across the wider peer set.
- Fitch: Big Four Reinsurers' Record 21.5% ROE Masks Price Cuts Still to Earn Through - the European peer view of the same half.
- Reinsurance Illiquidity and the Cost of Record Capital - what record capital does to the return investors will accept.
Sources
- AM Best, "Reinsurers' Returns Exceed Cost of Capital Despite Softening Market" (August 12, 2026) - the 9.6% cost of equity, the 16.3% median ROE, and the 7.67% / 8.23% / 8.63% weighted average cost of capital series.
- "US and Bermuda reinsurers maintain profitability as premium growth slows: AM Best" (Reinsurance News, September 1, 2026) - the seven-company composite, the 16.8% aggregate net ROE, the 90.2 combined ratio and the Greg Dickerson quote.
- "Global and US property cat rates down 16%, APAC 19% after July renewals in 2026: Guy Carpenter" (Artemis.bm, June 30, 2026) - the Global Property Catastrophe Rate-On-Line Index movements and regional splits.
- Federal Reserve Bank of St. Louis, 10-Year Treasury Constant Maturity Rate (DGS10) - the 4.77% close on September 3, 2026.
- Federal Reserve, Open Market Operations - the 3.50% to 3.75% federal funds target range and the 2025 cutting cycle.
- Aswath Damodaran, "Data Update 2 for 2026: A Testing Year (2025) for US Equities" (January 23, 2026) - the 4.18% risk-free rate and 4.23% implied equity risk premium at January 1, 2026.
- Gallagher Re, "Reinsurance Market Report: Results for Half-Year 2026" (September 2026) - the 11.4% cost of equity, 19.9% reported and 13.8% underlying ROE, and the 86% premium-to-capital ratio.
- "Aon: Record capital set to drive more flexible reinsurance market in 2027" (Global Reinsurance, September 2026) - the $800 billion capital figure and the 15.5% annualised sector ROE.
- "Reinsurance competitive pressures to rise in 2027, discipline can keep returns attractive: AM Best" (Artemis.bm, August 18, 2026) - the interest-rate environment characterisation and the 2027 capacity outlook.