Protection is coming loose from the policy

Our industry has a settled habit when it looks ahead. We spot the trend, translate it into a price, renew the policy. The AXA 2026 Foresight Report, the Atlas of New Futures steps outside that habit. It makes no forecast. Over a fifteen-year horizon it builds five fictional regions out of signals visible today, then describes daily life inside each one: a bioeconomy grown on coral reefs, city dwellers moving into emptied countryside, a women-led nomadic solidarity network, a city sold by subscription, and a new frontier opening in the Arctic.

The method interests me as much as the stories. The team developed thirteen candidate scenarios, kept five so that different logics of adaptation would sit side by side, and matured some of them with a climate research cluster at the University of Hamburg. They are not trying to know the future. They are trying to break the templates we use when we think about it.

I read all five. Insurance sits at the centre of none of them. It shows up in a corner of every one, doing a different job each time. That is the part I care about. I read that job under four headings below; two scenarios (the Amadare Network and Studio Odgrod) sit together because they land on the same lesson.

Protection is coming loose from the policy

Harmony Harbor is the scenario that looks straight at the insurer. Cities built on vast platforms near ports, supplied directly by sea, switching on a floating reactor when the national grid fails. Residents are not adapting to the climate. They are buying continuity. The report calls it Resilience-as-a-Service.

No insurer makes the promise of protection in this model. A state-backed logistics company makes it, and it delivers not by paying claims but by making sure the loss never lands. The forecasting system sees the flood weeks in advance, container ships move the threatened living units, life carries on uninterrupted. The customer profile is stated plainly: older, tired of adapting, wealthy enough to afford what the report calls simulated normality.

For an insurer, the competitor here does not carry an insurance licence. It is an operator that meets the customer’s need for peace of mind before we do, and bills for it monthly. Drawing on UBS’s 2025 wealth report, the AXA authors note that trillions of dollars will change hands through inheritance over the next twenty-five years. In unequal societies that money tends to buy private alternatives to public systems rather than strengthen the public ones. Who sells those alternatives is a strategic question, not a technical one.

The premium selects the future

Marseille 2 describes a resilience hub in the Arctic. The region warms roughly four times faster than the rest of the planet, retreating ice opens new routes, and Greenland holds twenty-five of the thirty-four items on the European Union’s critical raw materials list. Cold air gives data centres free cooling, and thirty-two of them already run in the Arctic today.

One sentence in this chapter is, for me, the sharpest line in the whole document. The Northern Sea Route cuts the distance between Northern Europe and East Asia by thirty to forty per cent compared with Suez. It still cannot scale. The reasons given are ice-class vessel certification, thin port infrastructure and high insurance costs.

The route is physically open and economically closed. Our price is part of what closes it.

We usually position insurance as an industry that reacts to what has already happened. Here the arrow points the other way. Marine premiums set the pace at which a trade corridor becomes real. The same mechanism runs everywhere. What can be insured gets built; what cannot gets postponed. The same hand shapes where a plant goes up, whether a housing project finds financing, how soon a technology reaches the field. I would treat that as the strongest lever we hold.

Investing in the source of the risk

In the Ayokarang Reef Region the insurer sells no policy. It invests in the reef. The fiction runs like this: port companies pay in for protection, biotech firms for verified genetic diversity, insurers to reduce the risk, and all of it goes directly into the ecosystem. Local communities take a guaranteed share of the proceeds for as long as they keep the reef healthy.

Part of it has already left fiction. The report points to Surf Ecosystem Insurance, developed by the Save The Waves Coalition with support from the Ocean Risk and Resilience Action Alliance, a financial instrument that funds rapid recovery of a wave ecosystem after extreme weather. Surfing contributes at least 194.7 million dollars a year to the local economy in Santa Cruz, and without adaptation measures all thirty-one surf spots there are expected to degrade. Barcelona’s network, which turned libraries, schools and museums into official climate shelters, works on the same logic. The city repurposed what it already had instead of building something new.

None of this is a claims payment. It is spending that stops the loss from forming. When the institution writing the cover also invests in the condition that produces the risk, both the balance sheet and the customer relationship change shape. An annual renewal conversation turns into a partnership that manages risk together.

The lives insurance never sees

The Amadare Network rests on women-led nomadic communities moving through arid geographies. Their protection system is not insurance. It is remittance income and mutual aid. The figures carry weight. Remittances to low and middle income countries reached 685 billion dollars in 2024, more than foreign direct investment and official development assistance combined. Access to M-PESA in Kenya lifted roughly 194,000 households out of poverty, with the effect strongest in households led by women.

The chapter’s list of indicators to watch includes one line that stopped me: insurance models for mobile and non-traditional ways of living. Studio Odgrod’s list carries a sibling of it, insurance access for small-scale farming. In both scenarios insurance appears as a gap nobody has filled yet, not as a product on offer.

Odgrod’s backdrop explains the gap. In the European Union the average farmer is fifty-seven, the share under forty is twelve per cent, and sixty-one per cent of soils count as unhealthy. Between 2021 and 2023, climate-driven extreme weather cost the Union 162 billion euros. A small, diversified farm survives these conditions on the cover it can reach as much as on its crop. The report lists this not as a product opportunity but as a precondition that decides whether the scenario happens at all.

Reading that, I thought about our own architecture. A policy is written against a fixed address, a fixed asset, a recorded income. Lives that move, that have no address, that earn irregularly stay invisible to the system. Climate keeps adding to their number. The World Bank projection quoted in the report puts internal climate displacement at up to 216 million people by 2050. The population we cannot see is growing.

Reading it from a market like Turkey

None of the five scenarios describes Turkey. Four of them touch it.

On earthquake, a market like Turkey set up its mandatory pool years ago, yet the pool is still far from full. According to figures released by DASK in August 2026, nationwide penetration stands at 58 per cent, from a high of 65 per cent in the Marmara region to a low of 45 per cent on the Black Sea coast; the institution’s legal target is 100 per cent. Even at full penetration the real exposure would remain, because it sits in the building stock rather than in the cover. The Ayokarang logic is clear enough here: pricing the premium correctly, and even filling the pool, is not the job; standing on the side that finances retrofitting is. In agriculture, the gap between what the state-backed insurance pool covers and total production is Odgrod’s insurance access line in local form. Seasonal farm labour, informal employment and internal migration put Amadare’s question to us far earlier than to most. In logistics, the country already lives what Marseille 2 underlines: part of every route and investment decision is taken inside an insurance price.

I do not read these reports as forecasts. I read them to test the assumptions I am working with today. One sentence came out of the five scenarios for me. Protection is coming loose from the insurance product, and wherever it comes loose somebody else takes the role. A logistics operator, a community fund, a municipality, a technology company.

Rather than trying to stop that separation, I would prefer to stand where the function of protection is going. A policy is a product. Protection is a function. Our share of the future will be decided by how far we follow the function, not by how well we defend the product.

Sources: AXA Foresight Report 2026, “Atlas of New Futures”. Earthquake insurance penetration: DASK / Anadolu Agency, 16 August 2026. Images: AXA Foresight Report 2026.