Insurance Desk

The Cycle Margaret Ennis

S&P's retrocession call is a classic late-cycle tell. When reinsurers start buying retro in volume — specifically because pricing looks favorable and they want to cap their cat exposure heading into a softening market — you are watching the market hedge against its own excesses. The retro market was effectively hollowed out in the hardest years; now, with capital having returned to the primary and reinsurance layers, retro is reconstituting itself. That is not a bullish sign for the hard market. It is a sign that underwriters sense the peak.

The CAD 923 million Saskatchewan and Manitoba revision is the other side of that coin. Secondary perils — severe convective storm, hail, inland flood — have been systematically surprising to the upside in this cycle, and this three-month revision from CAD 850 million is textbook loss development on an event that most aggregate covers and industry-loss-warranty triggers were not structured to capture cleanly. Every upward revision like this erodes the euphoria of the post-renewal period and reminds cedents why they need protection that actually attaches.

The S&P note that collateralized tail protection remains 'stable' is the caveat that matters. The cat-bond layer is holding, but the middle market — working layers, aggregate covers, retro — is where the softening pressure is being felt first. The capital is back; it is just not yet back in the cheapest form. That will come. Hard markets sow their own undoing, and right now the seeds are in the soil.

S&P's retrocession-comeback signal marks the late phase of the hard market: reinsurers are hedging their own cat exposure because they sense softening ahead, while secondary-peril loss development on events like Saskatchewan/Manitoba proves why the protection is needed.

Bias flag — Mean-reversion lens may cause Margaret to underweight the structural regime shift: if climate non-stationarity means secondary-peril frequency is permanently elevated, 'the capital comes back' thesis produces a soft market that is mispriced relative to the new frequency environment.

Cat Bond Desk Soren Vaeth

The Artemis dashboard puts the cat-bond market yield at 8.86% — a 5.05% insurance risk spread over 3.81% collateral yield — against an outstanding-market expected loss of 2.5%. That is a spread-over-EL multiple of just over two times on the aggregate book. For context, the market is absorbing $18.9 billion in YTD issuance across 94 deals at an average size of $136 million without apparent indigestion. The Porch Group's Harbor Crest Re ($100 million, multi-peril including named storm, wildfire, winter storm) and Hannover Re's 3264 Re ($200 million, US and Canada named storm and earthquake) are the most structurally interesting recent deals: both are broadening the peril set and issuer diversity, which is exactly what you want to see if you are worried about concentration.

Margarett Ennis is right that the retro comeback is a late-cycle signal, but I would push back on her framing slightly. The S&P note specifies that collateralized tail protection is 'stable' — meaning the cat-bond layer is not softening in tandem with the traditional retro market. That is a structural feature, not a cycle feature. ILS investors have a different capital cost and return target than Lloyd's syndicates buying retro; the spread-over-EL at the 5.05% risk spread level remains attractive relative to high-yield credit (HY OAS is 2.67% per the live quant snapshot), which explains why capital keeps arriving.

The CAD 923 million Saskatchewan/Manitoba revision is a secondary-peril event, and the critical question for cat-bond positioning is whether it was within the modeled loss range for aggregate cat bonds covering North American severe convective storm. If the CatIQ figure triggers any industry-loss-warranty resets on aggregate covers, we will see that in secondary pricing over the next 30 days. Watch for spread widening on multi-peril aggregate structures specifically.

At 5.05% insurance risk spread against 2.67% HY OAS, cat bonds remain attractively priced relative to credit alternatives, sustaining $18.9B YTD issuance — but the CAD 923M Saskatchewan/Manitoba upward revision is a secondary-peril test case for aggregate ILS structures that the secondary market will reprice over the next month.

Bias flag — Soren reads the 5.05% risk spread as adequate relative to HY OAS, but this comparison assumes the cat-bond expected-loss figure (2.5% market-level) is correct; if secondary-peril models are systematically understating frequency — as Ravi argues — the spread-over-EL multiple is flattering and real risk-adjusted return is lower than the headline suggests.

Modeled Loss Dr. Ravi Chandrasekar

The CatIQ revision from CAD 850 million to CAD 923 million — an 8.6% upward move at the three-month mark — is squarely within the documented pattern of secondary-peril loss development. Severe convective storms in Canadian prairie provinces are a peril where the event catalog is thin relative to the Gulf Coast hurricane record; exceedance-probability curves for Saskatchewan and Manitoba are built on a shorter and less uniform observational record, which means the model confidence interval is wider and the upward development risk is higher than users typically discount.

The NOAA confirmation that the contiguous U.S. just recorded its hottest August, hottest summer, and hottest year-to-date in history is the structural backdrop against which this specific event revision must be read. Record heat is not just a wildfire preconditioner for California — it is a moisture-flux driver that intensifies the convective available potential energy (CAPE) environment for Great Plains and Canadian prairie severe weather. The model is a hypothesis built on historical temperature-moisture relationships; if those relationships are shifting, the EP curve is already stale by the time the event season ends.

I want to flag something that Soren raised — whether the Saskatchewan/Manitoba figure triggers aggregate ILS structures. That is the right question, but the harder one is whether the models that priced those structures adequately captured the frequency distribution of multi-storm aggregate accumulations in the Canadian prairie corridor. My read from the structure of this event: probably not fully. The three-month development pattern here is consistent with demand surge (construction costs, labor scarcity in semi-rural prairie markets) on top of a modeled footprint that likely underestimated commercial property exposure in the affected Saskatchewan agricultural and industrial zones.

The CAD 923M Saskatchewan/Manitoba revision fits the secondary-peril development pattern precisely, and NOAA's record-heat data suggests the environmental drivers of severe convective storm intensity are shifting in ways that make historical EP curves increasingly unreliable as pricing anchors.

Bias flag — Ravi's focus on model architecture and EP-curve staleness is correct but may underweight the litigation and social inflation component of loss development — the CAD 923M revision could include claims amplification from public adjuster activity and legal representation, not just physical model error.

Protection Gap Daniela Owusu-Reyes

CAD 923 million in insured losses sounds like a large number until you ask what fraction of total economic losses it represents. CatIQ is an industry loss estimator — it counts what was insured. The NOAA record-heat headline is a reminder that the events driving these loss estimates are accelerating in frequency and intensity, while the communities most exposed to severe convective storms in Saskatchewan and Manitoba are predominantly agricultural and lower-density, exactly the profile where commercial take-up rates for comprehensive property coverage are lower and residential over-insurance is rare.

The retrocession comeback story and the cat-bond market's $65.6 billion in outstanding capital are not relevant to a grain farmer in Manitoba whose crop insurance is federally administered and whose property policy may not cover the full replacement cost of a storm-damaged machine shed. The professional capital markets are getting very good at pricing the upper tail for institutional cedents. The middle of the loss distribution — the working layers that small commercial and residential policyholders depend on — is where softening market conditions and retro retrenchment during hard years have left the most damage to coverage availability and affordability.

Ravi's point about demand surge in semi-rural prairie markets is directly connected to the protection gap question: when construction costs surge post-event in thin-labor markets, insured-to-value ratios collapse. A policy that was at 80% replacement value before the storm may be at 60% after a 25% demand-surge spike in lumber and labor. That gap is invisible in the insured-loss headline and entirely absent from the cat-bond pricing conversation.

The CAD 923M insured figure measures only what the market covered; the protection gap in Saskatchewan and Manitoba agricultural and rural communities — where demand surge erodes insured-to-value ratios post-event — is the number CatIQ cannot produce.

Bias flag — Daniela's framing of rural agricultural underinsurance as pure market failure underweights the role of moral hazard in subsidized coverage zones and the legitimate actuarial challenge of pricing low-density, high-volatility peril regions.

Solvency Watch Eleanor Pryce

The S&P retrocession note carries a specific solvency-relevant signal: reinsurers are buying retro because they are managing their catastrophe limits in a softening market. Translation — the reinsurers most exposed to secondary-peril frequency (Canadian severe weather, U.S. severe convective storm, European flood) are using retro to keep their RBC ratios and Solvency II SCR coverage within board-mandated bands. That is rational risk management, but it also tells you where the stress concentrations are. If retro pricing firms back up unexpectedly — say, because a late-season Atlantic named storm or a second major Canadian severe weather event tightens retro capacity — the reinsurers who have not yet locked in their protection will face the worst of both worlds: softening primary market and firming retro.

The CAD 923 million Saskatchewan/Manitoba figure is not large enough in isolation to threaten any single reinsurer's solvency, but it is a data point in the secondary-peril accumulation story that rating agencies and state departments are tracking carefully. For AM Best and S&P, the question is whether reinsurers' aggregate-year cat budgets — already tested by the U.S. severe convective storm season — are being exceeded before Atlantic peak season is fully resolved. Any reinsurer that entered 2026 with a secondary-peril budget of, say, USD 500 million and has already consumed that budget through U.S. and Canadian severe weather events is running with unbudgeted exposure from now until year-end. That is where rating-watch-negative actions originate.

Reinsurers buying retro to manage cat limits in a softening market signals active aggregate-year budget pressure; if retro capacity tightens unexpectedly before year-end, those without locked-in protection face compounded solvency stress.

Bias flag — Eleanor's focus on aggregate-year budget depletion as a precursor to rating-watch actions may overstate the near-term solvency risk for well-capitalized global reinsurers with diversified books; CAD 923M is a regional event, not a systemic capital event.