Over the past several years deploying capital in the insurance capital markets, I’ve found that most coverage rarely goes beyond surface-level headlines, outcomes, and broad narratives. 

This newsletter takes a different approach - a perspective from inside the market - the pricing cycles, the capital flows, what’s moving markets, and what’s behind the headlines.

I’ve spent the last few weeks on the road - stops in London, New York, Dallas. Several investors have asked me a version of the same question: as the interest rate environment changes, how do you expect insurance returns to change with it?

It's a fair question. Jerome Powell is on the way out, and the idea of central bank independence is under pressure. Anyone who's come across Pradhan and Goodhart's The Unanchored Central Banker — or listened to Patrick Boyle's episode on it (cheat code) — has reason to think carefully about what the structural rate environment looks like over the next decade. 

Pradhan and Goodhart argue that the conditions that let central banks deliver low, stable inflation for thirty years are fading. Aging societies push deficits up faster than private savings can absorb. China's long export of goods deflation is reversing. High public debt restricts how aggressively a central bank can hike. The real rate, in their view, is biased structurally higher from here.

Whether they're right or wrong, I should be able to say something coherent about what it means for insurance returns. So I went deeper.

Quick note before going further: when I say 'rates,' I mean interest rates. When I say 'insurance pricing,' I mean what an insurer charges per unit of risk. Easy to mix up.

Intuitively, you’d expect insurance returns to be extremely sensitive to interest rates. Insurers collect premiums upfront, invest those premiums as ‘float’, and earn income from those assets while claims develop over time. Higher rates should mean structurally higher returns. One would then expect that extra investment income to be eventually competed away through lower insurance pricing.

What do the textbooks say? In equilibrium, over a long enough horizon, insurance returns are roughly the same whether rates are at one percent or six. There are two routes to the same conclusion - the underwriting perspective and the investment perspective.

From the underwriting perspective: fair insurance pricing equals the present value of expected loss payments, discounted close to the risk-free rate. When rates rise, the PV of future claims falls, and fair pricing falls with it. When rates fall, the PV rises, and fair pricing has to rise to cover the same expected loss. The market adjusts to whatever the rate environment dictates. Through a full cycle, an insurer's underwriting return ends up roughly the same whether rates are at one percent or six.¹

From the investment perspective: Premiums collected sit on the balance sheet as investable assets until claims are paid - the float - Buffett famously described this as an interest-free loan from the policyholder. If the underwriting result is break-even or better, the float costs nothing, and any investment return on it is profit generated on capital the insurer didn't have to raise. When rates are high and float is more valuable, competition pushes insurance pricing down. When rates are low and the float earns less, less competitive pressure pushes insurance pricing up. The cost of float gets competed away in proportion to its value. What remains is a real return on capital.²

Across enough insurers and a long enough window, the empirical work supports this.³ The interesting question - why does the realised experience so rarely look like that?

Capital can flow in the opposite direction. The underwriting cycle isn't driven by anyone running fair-premium calculations. It's driven by capital supply: the volume of insurance capital available to write risk at any given moment; when capital is plentiful, insurance pricing falls. When capital exits, insurance pricing rises. 

This means capital can flow in the opposite direction from what the principles of rate neutrality would suggest if the alternative available to investors is less attractive. In other words, when rates fall and yields elsewhere look unattractive, capital often moves into insurance - exactly when the textbook says it shouldn't.

The 2010s are the clearest example: with rates near zero for most of a decade, alternative capital deployed in reinsurance grew at roughly 15% per year⁴ causing a significant fall in pricing - running counter to the theory of rate neutrality for the better part of a decade. When rates moved sharply upward from 2022, the pattern inverted - capital pulled back as other asset classes looked more attractive, and pricing rose. Again, the mechanism to deliver rate neutrality operated in reverse.

Adjustment is lagged and violent, not smooth. The 1980s provides good historical context. With U.S. rates in the high teens, P&C insurers wrote at combined ratios above 110% - 10 cents of underwriting loss on every dollar of premium. High rates enabled insurers to generate a yield on the float that more than covered the underwriting losses. 

Then rates fell. The math that had been working stopped working. By 1985, a lower yield on the float created significant real losses for insurers as underwriting losses could no longer be offset. Insurance pricing eventually reset higher - but only after a multi-year stretch in which insurers were trapped carrying underpriced books while losses accelerated. Several became insolvent, leading to the 1985 liability crisis⁵. The rate neutrality offset worked but it took years, breaking many insurers in the process.

Different lines of insurance behave differently. Short-tail property and cat are one thing: claims settle quickly. A hurricane either damages the building or it doesn't. The rate washes through at the next renewal cycle. The offset works fairly cleanly.

Long-tail casualty is a different beast. General liability, professional indemnity, commercial auto and other long-tail lines carry reserves that run for years - and the loss costs move with the same macro forces that drive yields: labour costs, medical costs, auto repair costs, etc. 

Swiss Re's 2024 analysis of social inflation put U.S. liability claims inflation at roughly 5% per year against general inflation closer to 4%.⁶ Their conclusion: the impact of liability inflation on casualty business will outweigh the interest rate benefit within one to two years. Ironically, the lines where float is largest and rate sensitivity is highest are often the lines where the rate-neutrality offset works least cleanly.

Three things have stayed with me since working through this.

The textbook is right about the destination but it’s silent on the journey. The conventional intuition isn't wrong - it's a statement about equilibrium. What it's silent on is what it costs to get from one equilibrium to another. For an insurer, the journey can include insolvency.

High rates can hide a lack of discipline for a long time. When float earns well, an insurer can write less-disciplined business and show consistent profit. When rates fall, the slack can disappear quickly. Underwriting cycles can run fifteen to twenty years peak-to-peak - long enough that senior underwriters can spend entire careers inside a single rate regime. What looks like discipline may be calibrated to a regime that underwriters have never seen change. 

Rates do shape how returns show up in any given period. But historical returns aren't the test that tells you whether you're backing a good manager - the key test in this context is whether historical returns have been supported by disciplined underwriting that will survive a regime change. 

Whether Pradhan and Goodhart are right or wrong, the period ahead will provide (another!) test of underwriting discipline.

The next issue will look at the difference between insurance-linked securities and specialty insurance, and why it matters for capital allocation. The distinction is more consequential than most LP materials suggest.

If you've been sitting on a question you want me to look at - about a product, a structure, an idea - send it over. Some of the better issues come from the things I haven't thought to ask yet.

Until then.

Charlie Evans, CAIA
Chief Investment Officer, Zephyros Investments

¹ https://www.casact.org/sites/default/files/database/forum_03fforum_03ff515.pdf
² https://www.berkshirehathaway.com/2024ar/2024ar.pdf
³ Doherty, Neil A., and James R. Garven (1995). "Insurance Cycles: Interest Rates and the Capacity Constraint Model." Journal of Business 68(3).
⁴ https://www.mckinsey.com/industries/financial-services/our-insights/insurance/alternative-capital-in-property-and-casualty-a-way-forward
⁵ https://pubs.aeaweb.org/doi/pdf/10.1257/jep.5.3.115
⁶ https://www.swissre.com/dam/jcr:6bc7d3b7-0f42-4209-a01a-e22787b98685/sri-sigma4-2024-litigation-costs-claims-inflation-final.pdf