As delegates departed Monaco following the annual Rendez-Vous de Septembre, Anthony Hobley co-lead insurance for the UN High-Level Climate Champions, CEO of Carbon Tracker and an adjunct professor at the Singapore Institute of Technology says the risk market has fundamentally changed from that one which the event was first founded.

In 1957, André Roux founded the Rendez-Vous de Septembre for the market to take stock before reinsurance treaties renew on 31 December. It was a sound design for a world in relative equilibrium. Hazard was broadly stationary, and a twelve-month look-back was a fair proxy for a twelve-month look-forward. Annual repricing was an elegant solution to a stable problem.

The market arrived in Monaco off a benign first half, with the lowest insured natural catastrophe losses since 2020. It also arrived on a Mediterranean coastline at the close of a French fire season that burned three times the long-run average.

Two facts, one season, one room. The renewal will be priced off the first.

The most sophisticated risk-pricing institution on earth is structurally prevented from pricing the risk visible from its own balconies. A contract repriced every twelve months need never hold a view beyond twelve months. As the Bank for International Settlements noted in Too Hot to Insure, most insurers do not explicitly price climate change, relying instead on annual repricing that misses chronic physical risk. A world in equilibrium could be insured twelve months at a time. A world in rapid transition cannot.

What I am not asking for

I am not asking for thirty-year fixed-premium contracts—no reinsurer would stand behind them. Nor am I asking insurers to forecast their own prices, which depend on unknowable future capital costs and competition. What can be projected is the price of risk: expected loss from a changing hazard at defined return periods for a defined place. Keep the annual contract; add the forward curve behind it.

Pensions already do this. Your annual statement shows current pot value, but pairs it with a standardized projection to retirement. It isn’t a promise, but it gives the saver time to act while acting is still cheap.

A pension provider that sent only the current valuation would be failing in a duty of care to the saver. I think it is worth asking, out loud in Monaco, why we believe we owe the people who depend on us any less. The Insurance sector needs to find its version of such a projection for its insured clients.

We have already drawn the curve. Once.

Some say the long horizon cannot be modelled, yet we have done it. In May 2025, fi-compass published Insurance and Risk Management Tools for Agriculture in the EU for the European Commission and EIB. It produced a forward climate risk curve using risk metrics common to global capital markets. Annual average loss for EU27 agriculture—now EUR 28.3 billion—is projected to rise by 42–66% by 2050 depending on emissions pathways. Holding farming practices constant measures the cost of inaction, creating an actionable investment schedule rather than a countdown to abandonment.

Yet no equivalent curve exists for Mediterranean wildfires, UK floods, or long-lived energy assets—not because we lack capacity, but because no one has commissioned it.

The data is available. Swiss Re warns heat in Europe is becoming a chronic exposure, and reinsurers are reconsidering whether wildfire is merely a secondary peril. Munich Re data shows Europe accounted for just 5% of global wildfire losses between 2016 and 2025—the historical baseline annual models calibrate on, and the number that will be wrong. Across Europe, only a quarter of climate-related catastrophe losses since 1980 were insured.

Every other major commodity—power, gas, freight, credit—has a forward curve. Physical climate risk is the sole major asset class without one, despite our sector being best equipped to draw it.

What the missing curve costs

An annual pricing model can only raise prices or exit. It cannot warn a decade ahead while adaptation is possible. The signal arrives not as a manageable price curve, but as a non-renewal letter—as Tenbury Wells in Worcestershire discovered, and as communities across Galicia, the Aude, and the Var are experiencing now.

The consequence is financial. Insurability is a gate: as the COP30 Insurance Communiqué noted, what is not insurable is neither bankable nor investable. Foresight is the raw material of adaptation, and insurance controls the supply.

A twelve-month price on a forty-year asset

The horizon problem also dictates what we underwrite. Capital markets read insurance pricing as a verdict on risk, unaware it is a twelve-month snapshot silent on long-term litigation, demand destruction, regulatory restriction, and cumulative physical impacts.

When we write a twelve-month policy on a new fossil fuel facility, we publish a number that may misleads markets about medium- and long-term risk. New LNG capacity—like TotalEnergies’ Alaska LNG, Venture Global’s CP2, or Commonwealth LNG—is sanctioned on long-dated revenue contracts. Lenders take a twenty-year view of revenue, while insurance takes a twelve-month view of risk.

The precedent is troubling. Asbestos and tobacco were profitable annual risks before becoming multi-decade liabilities through the courts. A twelve-month model offers no protection against long-tail liability; it merely prevents us from seeing it coming.

The uncomfortable arithmetic

You cannot publish a twenty-year forward climate risk curve for the Mediterranean while underwriting the fossil fuel expansion driving that risk. I do not believe any honestly drawn forward curve supports underwriting new oil, gas, and coal exploration based on our own numbers over asset lifetimes. Not on ethical grounds. Rather, on our own numbers, over our own asset lives.

Even if a curve permitted it, accounting for Earth system climate sensitivities would reveal a vastly different risk profile. Is our twelve-month model mispricing the assets we feel most confident about?

This is not a call to become environmental regulators. It is a mandate to finish the pricing job we proved capable of delivering in EU agriculture.

What I am asking for

• Treat forward climate risk curves as public infrastructure. Authorities should commission 2050 forward curves for wildfire, flood, and heat, using the EU agriculture study as a template. Supplying this analytics represents a commercial revenue line, not a sacrifice.
• End abrupt exit letters. Insurers holding forward portfolio views should not jump from quoting to non-renewal without warning. We must share existing forward risk insights with clients before withdrawal.
• Match capital horizons to asset lives. Apply multi-decade views, ORSA, and reverse stress testing (such as the Bank of England’s SS5/25 framework) to thirty- or forty-year assets.
• Require regulatory standardization. First-mover disadvantages prevent voluntary compliance. Supervisors must mandate standardized long-horizon disclosures, building on established supervisor climate frameworks.

Back to 1957

André Roux built the Rendez-Vous around the 1 January renewal because in 1957 that horizon sufficed. It no longer does. As RVS drew to a close, the expertise gathered in Monaco recognised how climate-driven risks are growing more complex, but offered no structural solutions. Now all eyes turn to New York for Climate Week where the issue is very much on the agenda. The insurance industry can point their analytical prowess at 2050, and give society a framework it can actually act on.