This post was initially written in French https://freakonometrics.hypotheses.org/87347
Last summer, I published in Le Monde (the French newspaper) a piece that started from an observation that has become commonplace and yet is still not fully absorbed by financial institutions. Extreme climate events are no longer marginal risks, and the banking sector is slow to grasp their implications. That op-ed, reposted on my blog under the (original) title Climate crisis the next time bomb for the financial system, was mainly an attempt to describe a perception gap, almost a difference in reflexes. On the one hand, there is an insurance perspective, reasoning in frequency, intensity, pooling, and a long time horizon. It accepts uncertainty, as long as it can be priced, and as long as the rules of the game remain stable. That is what I know well. On the other hand, there is a banking perspective. It reasons in collateral value, resale possibilities, default risk, and correlations. It rather quickly loses interest in the debate about annual probabilities if, in the end, the collateral becomes hard to liquidate. If I had to sum it up, insurers reason in probability and a long time horizon, while banks reason in collateral and a short time horizon. I admit it took me time to understand this point. As I recall, I really became aware of it in 2022, while reading House prices falling in flood-prone areas, in Quebec, and I rediscovered it this autumn, through Floods, drought… these climate hazards that push house prices down, this time in France. I wrote a post this autumn, Insurance, real estate, and the manufacturing of ignorance, taking advantage of some U.S. news on this topic. The point was that, for many decision makers, and probably also many citizens, the climate crisis remains an abstraction as long as no line in the accounts moves. It becomes urgent when an asset becomes less tradable.
These are precisely the mechanisms that the report Treading Water the impacts of catastrophic flooding on Canada’s housing market brought to light. It does not tell only a story of falling prices. It also tells a story of scarcer supply and slower sales. Looking at past catastrophic floods in Canada, one observes on average a notable drop in the final sale price, a longer time needed to sell, and a marked decline in the number of homes listed in the months that follow. The lesson is simple, and it goes well beyond Canada. The climate shock is a shock to value, but also a shock to market liquidity. And liquidity, for a housing market, is a fragile “public good” (in the sense that it is something everyone benefits from without being able to reserve it for a single actor). When transactions become rarer, it is not only the seller who suffers. It is the entire chain. Real estate agents, appraisers, lawyers, lenders, insurers, everyone depends on a collective ability to keep goods and information circulating. The report Treading Water says so explicitly, noting that the reduction in listings creates stress that spills over to sector participants, including mortgage lenders. One might think that the core problem is payment default, therefore a credit crisis. Yet the report strongly qualifies this intuition. Even in the worst observed case, missed and deferred mortgage payments exist, but they do not look like a devastating wave at the scale of the residential mortgage market, and their effects generally seem limited to a few months. In other words, banks have good reasons to look at housing before looking at credit. Put differently, banks understand something that macrofinance sometimes underestimates: before being a default crisis, a crisis can be a resale crisis. Before being a solvency problem, it can be a market problem.
This framing helps read what is happening elsewhere. In France, when the press talks about price drops linked to floods, drought, or clay shrink-swell, one finds the same dynamic. Risk becomes visible through a very immediate signal, value. Households grasp this instinctively, because housing is for many their main asset. Banks grasp it instinctively, because housing is the main collateral behind many loans. But another actor weighs heavily, and is mentioned less often. It is the real estate intermediation industry, broadly understood. This world lives from the circulation of transactions, trust, and the simplicity of narratives and comparisons. It does not like information that creates friction, especially when that information is hard to verify, debatable on technical grounds, and potentially explosive commercially. I discussed this in my post Insurance, real estate, and the manufacturing of ignorance, where I explained that U.S. real estate platforms such as Zillow had started displaying climate risk scores, produced in particular by First Street Foundation. Those indicators were then removed from the main view after a backlash from some industry players, while remaining accessible in a less visible way. The issue is not only model quality. The issue is the politics of visibility. Who sees what, when, and in what buying context. There is in this episode a general lesson about misinformation. A kind of misinformation that does not rely on outright lies, but on interfaces, on the way information is made to appear or disappear, made central or peripheral, brought into the calculation, or left as a detail most people will not read. This matters, because it reveals a deep tension. The housing market needs trust, but trust can be produced in two opposite ways. Through transparency, at the risk of triggering abrupt markdowns. Or through smoothing information, at the risk of shifting the loss onto those who will learn too late. If one stops at this diagnosis, one might conclude there is no way out. Any serious risk information eventually shows up in value. Therefore any serious risk information eventually meets resistance. And the more systemic the risk, the greater the resistance, because nobody wants to be the one who turns on the light. In a market, the one who makes a risk visible is only describing reality, but they may trigger price declines and end up accused of having caused panic.
And yet it feels like something is moving. And not necessarily where one would expect. Not so much on online listings, but on the side of large portfolios. Part of the real estate sector is beginning to organize adaptation to climate change, with commitments, tools, methods, and an explicit willingness to build standards. In France, the article Climate major real estate players are organizing to anticipate physical risks, published by L’Argus de l’assurance last week, describes the signing of a Charter of commitment to adapt the real estate sector to climate change, led by the Observatoire de l’Immobilier Durable and the Plan Bâtiment Durable. It includes signatories with very large holdings (La Poste Immobilier, Caisse des dépôts, Bouygues Immobilier, BNP Paribas Real Estate, Nexity, and also insurance-linked entities such as Covea Immobilier or Swiss Life). The most interesting detail, in this kind of initiative, is not the list of names. It is the work sequence (the “process”) that is mentioned: analyze and understand risks across the portfolio, define objectives, prioritize them, deploy concrete action plans, organize feedback, write solution guides, and test tools that measure exposure and simulate adaptation options. In short, one begins to glimpse a stabilization strategy for a market that anticipates risk. A charter is not only a promise to do better, it is first of all a communication tool, a way of telling markets and regulators, we are moving, so let us avoid a too abrupt transparency shock, or a constraint that would catch the industry off guard.
To be clear, it also helps understand a tension. On the one hand, markets need standards, otherwise everyone tells their own story about risk and adaptation. On the other hand, the industry mainly fears that overly precise standards could become a machine for producing markdowns. In practice, this pushes toward forms of transparency that remain collective and aggregated, rather than fine-grained disclosure property by property, like what had begun to appear on consumer-facing platforms. This tension between micro transparency and macro learning makes it possible to connect three dynamics that seem distinct, but in reality describe the same object. The first dynamic is the capitalization of risk into prices. For several years, the academic literature has documented that some climate risks are already partly reflected in real estate prices, notably in coastal areas exposed to sea level rise. Disaster on the Horizon: The Price Effect of Sea Level Rise, by Asaf Bernstein, Matthew Gustafson and Ryan Lewis, showed markdowns associated with exposure, with more visible effects among better informed buyers. This is not a story of a generalized collapse, it is a story of a progressive adjustment, heterogeneous, sometimes discreet, sometimes rapid after an event. The second dynamic is the central role of insurance. This is often where risk becomes a number one cannot ignore, not an abstract probability, but a premium, a monetary amount. Carolyn Kousky, in Understanding Disaster Insurance New Tools for a More Resilient Future, stresses that insurance is not only a safety net. It is also an incentive mechanism, a pooling architecture, a political compromise, and sometimes a source of fragility when insurability recedes. Rebecca Elliott, in Underwater Loss, Flood Insurance, and the Moral Economy of Climate Change in the United States, goes further into the social and moral dimension. When risk increases, the question quickly becomes who pays, who is protected, who is left facing loss, and what collective expectations have been built by decades of public policy. Insurance does not only compensate, it shapes beliefs, fears, habits, and anger. Finally, the third dynamic is the governance of adaptation. Mary Comerio, in Disaster Hits Home New Policy for Urban Housing Recovery, shows that the post-disaster moment is a time of urban and social bifurcation. Rebuild in place or relocate. Reinforce or patch. Help everyone or target. These choices have lasting effects on future exposure, therefore on future value, therefore on future stability.
These three dynamics interact. And they make it possible to reformulate a question that often remains implicit. Who should bear the cost of adaptation, and through which channels will that cost be distributed? One way to tell that story is to return to what the Canadian report makes very tangible. After an extreme event, the market slows. Some sellers prefer to wait for stigma to fade. Others need time to repair. Some buyers become more cautious. The result is slower sales, fewer listings, and lower prices. For a bank, this is immediate information. Mortgage lending rests on a simple assumption: if the household stops paying, the property can be sold without losing too much. If that assumption weakens, banking risk increases, even if defaults remain rare. And that banking risk can increase before losses become statistically frequent, simply because confidence in resale begins to crack. For an insurer, the reasoning is different. Disaster losses can be pooled, up to a point. But if frequency and intensity increase everywhere, pooling becomes more costly. And if politics forbids reflecting risk in premiums, or if the market refuses to pay, insurance withdraws or becomes fragile. That is where the question of system sustainability becomes political. For real estate actors, finally, the topic is explosive, because it touches the engine of the sector itself. Any information that changes buyers’ preferences can change volumes, as I discussed in Insurance, real estate, and the manufacturing of ignorance. Any information that creates markdowns can turn property wealth into a hard-to-hold asset. And anything that makes transactions more complex can break market liquidity.
One can then see why information about risk is a battle. Not only a scientific battle, but an economic battle. And sometimes a cultural battle. Risa Palm and Toby Bolsen, in Housing Market Response to Sea Level Rise in Florida, describe this ambivalence well through the eyes of real estate agents. Florida is an extreme case, because coastal amenities are highly valued there, and yet the risk is obvious. People were already talking about it 20 years ago. The market can continue to function well, even when everyone knows, because people think they can resell before it becomes acute, because they downplay it, because they adapt to it, because insurance premiums are temporarily bearable, because risk perception is filtered by habits and identities. This matters, because it prevents a conclusion that is too simple, of the kind “just inform people”. Information can be denied. It can be relativized. It can be made invisible. It can be experienced as an attack. It can even produce counterintuitive reactions, where maps do not increase caution but instead reinforce denial, depending on who is looking, and depending on the narrative that comes with them.
This does not mean information is useless. It means information does not circulate in a vacuum. It circulates through social structures, economic interests, and political conflicts. That is why disaster studies are also useful, to step back a little in a debate that quickly becomes technical. Chester Hartman and Gregory Squires, in There Is No Such Thing As a Natural Disaster Race, Class, and Hurricane Katrina, remind us that disaster is not only a hazard. It is the meeting of a hazard and a constructed vulnerability, through housing quality, infrastructure, inequalities, and public choices. The housing market does not only record risks. It produces them, because it organizes who lives where, in what built environment, with what protections, and with what ability to recover. The charter led by the Observatoire de l’Immobilier Durable is a sign that something is happening. When a sector begins to produce its own adaptation standards, it is not only a virtue, it is also a way to control how risk will be said, measured, compared, and therefore capitalized. And this is where the real estate question becomes a systemic finance question. Because if adaptation becomes a standard, it can become a value factor, as one might say in a management class. An adapted asset, better protected, better ventilated, less exposed, could defend its liquidity, preserve demand, reduce insurance costs, and secure financing. Conversely, a non-adaptable asset, or one too costly to adapt, risks becoming a discounted asset, or even an asset nobody wants to carry. Jesse Keenan, in work on adaptation in North America and in more recent books (in particular Climate Change Adaptation in North America: Fostering Resilience and the Regional Capacity to Adapt, with Walter Leal Filho, and more recently North: The Future of Post-Climate America), emphasizes the idea that climate change does not only produce damages, it reconfigures the geography of values and flows. This prepares a natural continuation toward the question of displacement. Jake Bittle, in The Great Displacement Climate Change and the Next American Migration, shows how repeated disasters, public policy, insurance, and housing intertwine to produce departures, often constrained and often unequal. In that narrative, real estate is not a simple economic sector, it is a sorting mechanism. Those with slack can anticipate. Those without it bear the brunt, lose capital, or remain in areas where life becomes more expensive and riskier. In our article Insurance against natural catastrophes balancing actuarial fairness and social solidarity, we highlighted this clearly using French data.
Climate change does not enter finance only through risk models. It enters through housing. Because housing is the core of household wealth, and the core of banking collateral. When climate risk moves real estate values, it affects at the same time household confidence, the robustness of bank balance sheets, and the collective capacity to pool losses. What makes the topic so delicate is that transparency can be both necessary and dangerous. Necessary to avoid manufacturing hidden losses that will fall back on the most vulnerable. Dangerous because it can trigger concentrated markdowns, and therefore political conflict. Between these two poles, actors look for intermediate paths. Making adaptation practicable before transparency becomes punitive. Building standards before constraints land. Learning collectively without exposing individual properties too early. The risk, however, is to believe that organization alone is enough. A charter, a tool, a guide does not solve the central question: who pays, and when. If adaptation raises construction costs, who absorbs them? If adaptation raises operating costs, who absorbs them? If adaptation becomes a financing condition, who is excluded? If insurance withdrawal becomes a reality, who is stuck? If markdowns hit, who loses capital? Behind maps, behind scores, behind charters, there is a political economy of housing taking shape. A political economy where climate risk becomes a factor of market segmentation. And where misinformation, in the broad sense, does not flow only through fake news, but through invisible arrangements, interfaces, standards, silences, aggregates, details pushed into the background. I will stay modest, it is just a blog post, and I do not pretend to decide between transparency and caution. I only wanted to make the structure of the dilemma visible. And that is perhaps, in the end, the “time bomb” (which I mentioned in the title of the article I sent to Le Monde, but which was not retained). Not only the rise in losses, but the way risk enters prices, credit, and stability. That path depends on tiny decisions: whether information is displayed or not, whether disclosure is required or not, whether premiums rise or not, whether a loan is granted or not, whether renovation is done or not. And it is probably because these decisions are tiny that they are so hard to govern collectively.
OpenEdition suggests that you cite this post as follows:
Arthur Charpentier (January 7, 2026). Real Estate Wealth and Climate Disasters. Freakonometrics. Retrieved January 13, 2026 from https://doi.org/10.58079/15gq3