Post date 12 Oct 2026 (original LinkedIn date)
Original LinkedIn version· 2,319 words
‘Hiding to nothing’ by West Australian indie rock band Cloning, from their album ‘Vitriol’ released this year. Have a listen:
– love discovering new sounds!
Refreshed after a weekend of team leadership training with mountain rescue colleagues in beautiful north Wales, I’ve been reflecting on an FT piece that I read. Last week I had the distinct pleasure of speaking on a panel at the Local Pensions Partnership Investment’s (LPPI) annual client conference. Our topic was ‘The macro environment: risks and opportunities’, ably moderated by Richard J. Tomlinson, with the hard yards on rates, inflation, liquidity, and AI, being superbly covered by fellow panellists, Barry Kenneth and Emmanuel Bocquet, CFA. They kindly left me space to talk about something a little different – how resilience issues could challenge our investment portfolios in non-explicitly financial ways. As I pondered the topic and did my research, what became abundantly clear is that physical climate risk is finally being taken seriously by financial markets. But my central message was that we may still be looking in the wrong place for another source of instability.
Let me start the journey with something that Günther Thallinger of Allianz wrote last year that I found profound and stayed with me. His argument was fairly simple and straightforward - as physical climate risks increase, insurance becomes progressively more expensive. Eventually, in some places, it becomes unavailable at any sensible price. Once a house cannot be insured, it becomes difficult to mortgage, and if it cannot be mortgaged, the pool of possible buyers shrinks. Then property prices fall, collateral values deteriorate and lenders begin to take losses.
Repeat that across enough properties, in enough places, and what began as a climate problem becomes an insurance problem, then a property problem, then a credit problem and, potentially, a financial stability problem. For the panel session, this struck a chord because the sequence is so easy to understand and obviously financial. And recent US evidence makes it considerably less theoretical. The Federal Insurance Office’s analysis of 246 million homeowners’ policies found markedly higher premiums and non-renewal rates in the areas most exposed to climate-related losses. People living in the highest-risk fifth of ZIP (post) codes were already paying average premiums 82% higher than those in the lowest-risk fifth. Insurers were carrying higher loss ratios despite charging those higher premiums.
Pilita Clark in the Financial Times has recently explored much the same possibility. Patrick Bolton, one of the authors of The Green Swan, told the paper that physical risk was materialising considerably faster than many had expected. Adair Turner went further: if he were looking for something capable of producing a financial crisis, he said, his attention would be on lenders exposed to properties that become uninsurable and consequently fall in value, all of which seemed highly persuasive to me and could sensibly be shared with my LPPI panel audience. But I began to wonder whether it still leaves us looking at only one part of the problem, because what if the house doesn’t actually burn down? Or, let’s take the analogy further, what if the factory doesn’t flood? What if the physical asset remains almost completely intact, but the natural systems upon which their economic value depends begin to fail? I think that gives us a rather different route from environmental degradation into financial instability, one that doesn’t necessarily begin with an insurance claim at all.
Consider a food-processing plant. The buildings may be fine, the machinery operational and the debt perfectly serviceable according to every conventional measure of physical risk. But if water becomes scarce, agricultural yields decline or the cost of its raw materials rises sharply, the economics of that asset can change without anything happening to the actual asset itself. Alternatively a farm doesn’t have to disappear beneath floodwater to become less productive. Soil can lose organic matter and its ability to retain moisture, or pollination can deteriorate. Water availability can become less reliable, while the cost of fertiliser, irrigation or pest control can rise as some of the ecological functions previously provided without charge weaken. If you prefer, a reservoir can still exist while the catchment feeding it becomes less capable of storing water, slowing runoff or maintaining water quality. In each case, the financial problem arises not principally because we have damaged a stock of manufactured capital. Rather it arises because we have impaired the flow of services that makes that capital productive.
All of this drew me to conclude that we have become much better at thinking about physical climate risk as an asset problem. This can include flood depth against buildings, wildfire exposure, coastal erosion, or heat stress on infrastructure. What we can observe is expected loss, insurance premium, collateral value. However nature degradation behaves differently. It can work its way into financial statements through declining productivity, higher operating costs, interrupted production, volatile input prices and eventually weaker cash flows. What this gives us is a transmission mechanism that might look something like this:
nature degradation → ecosystem-service failure → lower productive capacity → food, water and resource scarcity → higher prices and interrupted supply → falling corporate margins and household purchasing power → defaults and asset repricing → financial instability.
So there won’t always be an insurance claim in there. This must surely broaden what we mean when we talk about physical risk?
There is a hint of this in The Green Swan itself. The paper is remembered, understandably, for bringing climate risk into the language of central banking and systemic financial stability, but tucked away towards the end is something rather more expansive. Climate change, the authors wrote, may be only the “tip of the iceberg”. They pointed to biodiversity decline, freshwater availability and soil erosion as threats to the Earth’s natural life-support systems. Elsewhere they described the sort of chain reaction that becomes possible when water scarcity leads to agricultural losses and food insecurity, while ecological, economic and regulatory systems begin interacting with one another.
Oddly, much of the financial conversation that followed this passage narrowed again. It envisaged that we had the capacity to become increasingly sophisticated about carbon, transition risk and, latterly, climate-related physical damage. But the wider systems argument remained much less developed. I believe that thinking about food and water expose why this matters, in the sense that neither behaves like an ordinary discretionary input. People cannot choose to stop consuming food when supply falls. Cities cannot decide they no longer require water. Agriculture, semiconductor manufacturing, power generation, chemicals, pharmaceuticals and food production can’t simply substitute away from water indefinitely. So when ecological degradation constrains supply, the first financial manifestation may be a price shock rather than an asset write-down.
I knew that this was what I wanted to cover on the panel – the idea that as food gets more expensive, water becomes more costly to treat, transport or secure, or companies compete for scarce resources, margins are squeezed. Households spend more of their income meeting essential needs and less elsewhere. That brings us to a second transmission mechanism, which I suspect also deserves considerably more attention:
food and water insecurity → inflation + social pressure + emergency intervention → fiscal deterioration/sovereign risk → tighter financial conditions → financial instability.
This mechanism, I imagine, is particularly tricky for central banks. A supply shock caused by ecosystem degradation isn’t readily fixed by changing interest rates. Higher rates don’t produce rainfall, restore soils or rebuild an aquifer, but the inflationary consequences may still demand a monetary response, especially if repeated shocks begin to feed into wages and expectations. Worse still, households then experience the problem twice: first through higher prices for essential goods and services, and then through higher borrowing costs.
Companies experience something similar: input costs rise as revenues weaken, followed potentially by an increase in the cost of capital. And somewhere in the mix sits the government. They aren’t passive observers when food or water systems become stressed. They subsidise consumers, compensate producers, invest in emergency infrastructure, support stricken regions, protect strategic industries and, in extremis, intervene in prices or allocation. We are familiar with governments becoming insurers of last resort after catastrophe. But nature degradation creates the possibility of something broader. The state may increasingly become the provider of last resort when ecosystem services fail. And that turns ecological degradation into a contingent fiscal liability.
For countries with plenty of fiscal capacity, that may mean higher borrowing and difficult choices over taxation and expenditure. For more vulnerable sovereigns, food and water insecurity can interact with currency weakness, import dependence, debt servicing and political instability much more quickly. None of this requires an apocalyptic ecological collapse. Small reductions in resilience can accumulate, with several modest harvest failures possibly arriving together. Water restrictions will constrain production during a heatwave, and food prices can rise at the same time energy demand increases. This forces hands, so that governments need to intervene just as borrowing costs are already elevated. Suddenly individual shocks that appear manageable in isolation begin interacting.
The UK’s experience with wildfire this summer offers a glimpse of the same problem from another direction. More than 1,000 wildfires had been attended by fire and rescue services by the end of August, and nearly one in five large fires in England and Wales reportedly came within 500 metres of built-up areas. The resulting debate has understandably focused on firefighter numbers, specialist teams, equipment and preparedness, and obviously those things matter enormously. But wildfire is in large part a landscape question – it’s about how water is held in the landscape matters, the condition of peatland, the state of vegetation and land management. Now we are at the heart of the issue - the resilience of ecosystems during extended dry periods matters greatly.
Response capacity is one form of adaptation. Reducing the probability that the landscape becomes so vulnerable in the first place is another. The same logic applies to flooding, and it shouldn’t come as a surprise that the government announced the setting up of a taskforce late last week to deal with expected flooding this winter. We can continually increase the amount of concrete protection downstream, or we can also ask whether soil, woodland, peatland, wetlands and floodplains are performing some of the infrastructure function we need from them. Flip it, it applies to drought as well. Building reservoirs and transfers matters, but so too does the condition of the catchments that collect, store, clean and release water. This is why I increasingly think the phrase nature as infrastructure is more than just a useful metaphor.
Infrastructure is something upon which other economic activity depends. We invest in it because the economy becomes less productive and less resilient when it fails. My audience would be used to this concept, I hoped. But had they taken the extra leap? By my definition, soils, rivers, aquifers, wetlands, forests and functioning ecosystems are infrastructure already. The difficulty is that most of the services they provide have never appeared properly on either side of the financial ledger their contribution tends to become visible only when it disappears.
This also suggests a slight problem with Thallinger’s otherwise compelling argument. He concludes that we ultimately have two levers: stop emitting carbon or capture it. For climate change, those levers are fundamental, but they won’t restore exhausted soil. They won’t replenish an overdrawn aquifer, recreate lost wetlands, restore pollinator abundance or repair a degraded catchment. Nor does decarbonisation automatically restore the natural systems that help us cope with the warming already locked in.
Mitigation remains essential, but preserving financial stability may increasingly require something else alongside it: namely maintaining and rebuilding the ecological infrastructure upon which productive capacity depends (can we refer to that as adaptation?). That should alter the way we think about nature-related financial risk. So perhaps the question isn’t primarily, “what is nature actually worth?” It may be much more useful to ask what happens to company cash flows, household finances, public balance sheets and ultimately the financial system when a particular ecosystem function deteriorates. Where does the loss appear and who absorbs it? Can it be substituted or can it be insured? When everything else fails, does the state eventually inherit it? And, crucially, are many apparently diversified companies or financial assets actually dependent upon the same underlying natural system? Those feel like financial-stability questions to me.
The next environmental financial crisis may still begin in the housing market. Thallinger may be exactly right. Insurance retreat could expose the vulnerability of property markets and transmit physical climate losses into credit markets remarkably quickly, but it isn’t the only route. The first warning might instead be a drought, a failed harvest, or a restriction on industrial water use. Maybe a sharp increase in food prices, a river whose water requires progressively more treatment, or a government forced to step in because the private economy can’t absorb another shock. By the time these begin appearing as defaults, falling asset values or bank losses, the important event has already happened somewhere else.
The productive capacity of nature has deteriorated, and our financial system, which has spent decades treating that capacity as effectively unlimited and largely free, has discovered that it wasn’t. Oops
References
Bolton, P., Després, M., Pereira da Silva, L.A., Samama, F. & Svartzman, R. (2020), The Green Swan: Central banking and financial stability in the age of climate change, BIS and Banque de France.
Clark, P. (2025), How the next financial crisis starts, Financial Times.
Federal Insurance Office, US Department of the Treasury (2025), Analyses of U.S. Homeowners Insurance Markets, 2018–2022: Climate-Related Risks and Other Factors.
Thallinger, G. (2025), Climate, Risk, Insurance: The Future of Capitalism, LinkedIn, 25 March 2025.
Webber, C. & Noll, L. (2026), Wildfires are now an escalating risk: here’s how the UK needs to prepare ahead of next summer, The Conversation, 1 October 2026.