Chapter 11 — The Uninsurables

No one is born uninsurable. A person, activity or place becomes so when a contract, market or institution decides that the risk will no longer be carried under ordinary conditions.

Émilie was forty-four when she and her partner applied for a mortgage. The bank approved the application. One step remained. In France, mortgage protection usually requires borrower insurance, which repays the loan if the borrower dies, becomes disabled or, under some policies, is unable to work.

Ten years earlier, Émilie had been treated for breast cancer. Regular checks followed, then the appointments became less frequent. She returned to work and lived without a recurrence. In ordinary life, the illness belonged to the past. On the insurance questionnaire, it returned to the present.

The right to be forgotten does not claim that the illness never existed. It decides that, after a specified period and under certain conditions, the illness may no longer be used to refuse insurance or impose an additional premium. True information need not be allowed to act for a lifetime [1].

Karim was thirty-eight. He employed two people and ran a small renovation company. When he tried to strengthen his professional liability cover, the form asked whether he had ever been convicted of an offence.

The conviction was fifteen years old. After a fight and a short period in prison, Karim had worked, built his company and paid his bills. None of that later life fitted into the box on the form.

One insurer declined the application. Another offered cover at a higher premium with several exclusions. The broker spoke of a specialist market. Karim could still obtain insurance, but he no longer belonged to the ordinary market. Another institution had extended the sentence without a new judgement.

A house can follow a similar path. It remains standing, but a letter announces that the policy will not be renewed. Another company accepts it with a very high deductible, meaning the part of any loss the household must pay itself. A third requires work the owner cannot finance. The bank hesitates, the sale becomes harder and the home loses part of its role as a source of security.

The word uninsurable covers several situations. It may describe a material limit. Loss has become almost certain, too concentrated or too large for a particular contract. It may also mark the end of a chain made up of models, thresholds, underwriting rules, inadequate income and missing assistance. In the first case, an insurer cannot reasonably promise to pay. In the second, a particular arrangement has stopped making protection accessible.

Insurance without boundaries does not exist. Some risks exceed the resources of a private portfolio. Some places will need to be adapted and others left. The difficulty begins when a technical boundary is presented as a complete answer. A claim that something is uninsurable should always specify for which insurer, with what cover, for how long and after which preventive measures.

How a Boundary Is Made

A Technical but Situated Limit

A risk becomes difficult to insure when loss approaches certainty, when many contracts are affected at the same time or when the possible cost requires too much capital. Catastrophes often combine these problems. A flood, wildfire or cyclone can trigger payments across an entire area while also placing pressure on reserves, claims teams and reinsurance [2, 3].

The previous chapter showed how climate change adds uncertainty about the world to come. Insurers combine past observations with scenarios for infrastructure, urban development, forests and adaptation. A climate premium extends more than previous losses into the future. It also contains a view of what that future will be.

Moral hazard1 matters as well. Poorly designed cover can encourage building in an exposed area or reduce the incentive to complete useful work. This does not mean that every exposure follows from personal carelessness. Someone may have inherited a house, bought before the maps were revised or chosen the only home compatible with their work and income.

A technical diagnosis narrows the available responses without selecting one by itself. Highly correlated losses may call for compulsory reinsurance, a public fund or a larger pool. An unaffordable premium may call for targeted assistance, a loan for preventive work or supported relocation [4, 5]. The limit of a private contract is not yet the limit of every collective response.

Moving to a larger scale does not solve everything. A public guarantee uses resources and may sustain development that should not continue. Uniform assistance can benefit owners of valuable property as well as households with no practical alternative. These are questions about how costs should be shared. They do not prove that the whole risk should fall on the person who happens to occupy the property today.

Supply, Membership and the Guarantee

Insurability cannot be measured simply by asking whether a company is present in an area. France’s Observatory of Insurability identifies several separate conditions. The hazard must be identifiable, people must have the financial capacity to buy cover, insurers must be willing to offer it and prevention must be sufficient to keep the system viable [6]. These conditions can change independently.

The first boundary concerns supply. A person receives no quotation, or only an offer from a last-resort scheme. The second concerns membership in the ordinary market. A policy exists, but it belongs to a specialist segment in which more costly profiles are concentrated. The third concerns the content of protection. The policy remains in force, but an exclusion, limit or deductible removes most of its value [7].

A home can be insured on paper and poorly protected in practice. The presence of a few policies in a municipality does not show whether new applications are accepted, premiums are affordable or required work can be financed. The same is true for a person. An extremely expensive offer or a policy emptied of its main guarantee preserves the appearance of choice while organising departure from the ordinary market.

Several decisions make this boundary. The contract defines the loss that will be covered. A model estimates its cost. An underwriting rule determines which applications will be accepted. Reinsurers and regulators influence the capacity available. Public authorities decide whether to finance work that could reduce exposure. No actor controls insurability alone, but their decisions combine to produce a concrete line.

An insurer’s withdrawal does not create an empty space. It moves loss towards the family, bank, municipality or state. Public authorities already redistribute risk through infrastructure, liability rules, emergency assistance and financial guarantees [8]. Insurance also distributes risk rather than merely compensating it [9]. Refusing a contract does not remove the cost. It changes the institution that will bear it.

The same information can move the boundary. A map created to guide public works may later support non-renewal. A medical diagnosis recorded to guide treatment may affect access to a mortgage. This change of purpose does not make the information false. It changes the power attached to it [10].

The Prudent Bad Risk

Insurance debates often feature the careless person. They build too close to water, drive too fast, smoke or neglect maintenance. When the bad risk is assumed to be at fault, a high premium can look deserved. Contemporary uninsurability brings a less comfortable figure into view, the prudent bad risk.

This person has not necessarily ignored a known danger. They may find themselves on the wrong side of a map after buying before it was revised. They may work in an exposed occupation, live in an underprotected municipality or carry the trace of an illness they could not avoid. A risk can be costly without the person carrying it having deserved the cost.

The language of risk still distributes blame easily [11]. A higher premium suggests that the policyholder should have been more cautious. Yet an address carries a history of infrastructure, land values, segregation and planning. Health bears the imprint of work, housing and access to care. A model sees the outcome more easily than the chain that produced it.

Collective contradictions are then returned as individual tasks [12]. A resident must move in response to failed planning. A worker must change their conduct to offset occupational exposure. A former patient must prove recovery to the institution financing their home. Yet autonomy requires money, time, credit and real alternatives [13].

Poverty acts several times. It increases exposure, reduces the ability to prevent loss and makes the price harder to bear. The same deductible takes a larger share of a modest income. Work is postponed for lack of credit. Ordinary cover may then be replaced by a more expensive and more limited policy.

David Caplovitz showed how limited cash and restricted access to ordinary markets led poor families to pay more for goods that were often worse [14]. Contemporary classification continues this mechanism. It does not always close the door. It may open a more expensive one, with conditions that accumulate disadvantage [15].

The premium does not call the person guilty. It simply presents the cost as theirs. That attribution is often enough to hide the decisions about work, housing and planning that also contributed to the exposure.

The Household as Silent Insurer

When institutional protection recedes, loss does not always fall on an isolated individual. It enters the household. A relative advances the deductible, provides a place to stay after a flood, completes forms or reduces working hours. The family becomes a small insurer without capital, reinsurance or a meaningful right to refuse.

Households with more resources can buy time, expertise and services. Others absorb loss through debt, additional work or giving something up. This transfer rarely appears in official accounts. It is also unevenly distributed. Care and administrative work still fall disproportionately on women [12].

A delayed payment or exclusion creates unpaid work somewhere. Help given today may postpone a parent’s retirement, reduce a child’s education or become a moral debt between relatives. People without available family reach the boundary sooner. A policy that quietly assumes the presence of a protective household turns the absence of relatives into another disadvantage.

Collective protection is not meant to abolish family assistance. Its purpose is to prevent access to care, housing or reconstruction from depending entirely on the availability of a partner, daughter or friend. Withdrawal appears cheaper while this time, labour and uncertainty remain outside institutional accounts.

Inclusion without Belonging

The uninsurable person or place does not always disappear from the market. It may be moved into cover that is more expensive, narrower and less stable. Last-resort schemes prevent immediate abandonment, but they concentrate risks that ordinary insurers no longer want. They should be judged by the protection they offer and by whether people can eventually leave them.

California’s FAIR Plan illustrates the tension. It maintains basic fire cover for property owners who cannot find an ordinary policy. That continuity is essential. The cover is often narrower and has to be combined with costly supplementary protection [16, 17]. When the scheme grows for many years, it ceases to be a temporary refuge. It becomes the normal market for some territories.

When Glenn and Lorraine Crawford bought their home in Agoura Hills, north-west of Los Angeles, in 2012, insurance cost about $500 a month. The amount was substantial, but it still belonged to the ordinary expenses of owning the property.

Fourteen years later, the insurer did not formally reject the house. State Farm offered a new premium of more than $44,000 a year. The policy would still not have covered the full cost of rebuilding after a total loss. The Crawfords searched for an alternative. The only reported offer was about $80,000 a year through Lloyd’s [18].

On paper, the market was working. Two firms had named a price. In practice, an offer that almost no household can pay is not a real choice. Withdrawal can take the form of a premium that keeps the contract alive while placing it beyond reach.

Formal availability is too weak a test. The premium, deductible, exclusions, stability of the policy and adequacy of the possible payment all matter. An exchange can remain contractual and still be deeply unequal when one party has no realistic alternative [19].

A household trying to preserve access to credit or remain in a home that cannot be moved does not enter the market as an ordinary consumer. Lack of alternatives can become a source of revenue for a specialist segment. In a society organised by rankings, a poor score can lead to refusal and create markets for more expensive products [20].

Degraded inclusion can make the original problem worse. An extreme premium reduces the savings available for work that might restore access to ordinary cover. A large deductible delays repairs and increases vulnerability to the next event. A last-resort scheme should provide a route back. The required work, its financing and its effect on the premium should be known in advance.

Territorial uninsurability often emerges from many separate decisions. One insurer stops writing new business. Another raises deductibles. Reinsurance becomes more expensive and banks grow more cautious. No one announces that the neighbourhood is being abandoned, but the combined decisions change property values and the possibility of staying [21].

After the 2018 Camp Fire, some Californians whose houses had burned referred to owners whose homes remained standing as the unlucky ones. An insurance payment could allow the first group to rebuild or leave. The second remained tied to a property in a threatened area that had become difficult to insure and sell [22]. The house had survived the fire but was ceasing to function as financial security.

Some places will have to be adapted and others left. These choices affect property, debt, jobs and attachments to place. They become governable only when they are recognised as collective decisions [23]. Insurance can signal urgency. It should not become the quiet authority that decides by itself who may continue to live somewhere.

Tenants also experience this change. Higher insurance costs may return through rent. Repairs may be postponed and some landlords may sell. A municipality can lose residents and revenue at the moment it most needs to finance adaptation. The risk signal then weakens the institution that might have responded to it.

How Long Should the Past Act?

Uninsurability can attach itself to a biography. A cured illness, old debt, completed sentence or database error may continue to act when a person seeks work, housing, credit or insurance. The past becomes a predictive resource for institutions for which the information was not originally collected.

Some institutional memory is legitimate. Recent fraud, continuing exposure or a poorly reported loss may matter when a contract is assessed. The duration and permitted use of that information are still choices. Information relevant to medical care need not become credit information. A conviction may be examined for a particular position without following a person into every professional activity.

The right to be forgotten for some cancers places a limit on this memory. It does not change the medical past. After a specified period and under certain conditions, that past can no longer be treated as an admissible reason for different terms in French borrower insurance [1]. Recovery must allow a return to ordinary conditions of protection.

The same reasoning can apply to criminal records. When employment, housing, credit and insurance keep renewing the effect of a completed sentence, punishment continues without a new decision. Traces produced by one institution feed tools that allow others to keep the person at a distance [24].

A deletion date is not enough. Purpose must also be limited. Information may remain stored without being available for every decision. Correction matters just as much. A conviction attached to the wrong person, a medical code interpreted incorrectly or a debt already repaid can circulate for years. The person receiving the consequence must know which trace was used and be able to reach an institution capable of changing it.

The Right to a Fresh Start

The right to be forgotten belongs to a broader idea, the right to a fresh start. A conception of freedom extending across a whole life cannot require every illness, error or past difficulty to retain the same force forever. A just institution must allow changed circumstances to open new possibilities [25].

This does not erase responsibility. When a person has harmed someone, the repair owed to the victim remains a separate question. Repair does not require exclusion from employment, credit or insurance without an end date. The duration of institutional memory, the area in which it may be used and the severity of its consequences must each be justified.

A fresh start also has an insurance logic. Before knowing who will experience illness, bankruptcy, error or regret, people may have a common interest in living in a society where return remains possible. Collective protection then finances more than payment after loss. It also protects the possibility of regaining a place within ordinary institutions.

Return may require time, obligations or a new assessment. Its essential condition is reversibility. A person should know what they can do, how long the restriction will last and under what conditions the past will stop deciding on their behalf.

The bad risk is not simply the person or place that costs more. The category often appears at the point where collective explanation stops. A health cost is attached to a body rather than to work, housing or access to care. A flood cost is attached to an address rather than to a century of urban development and sealed ground. Pricing assigns loss to the actor the contract can reach. Politics must travel back through the chain that produced the exposure.

The threshold of minimum security helps identify the point at which this attribution becomes untenable [26]. A premium may reflect risk and still make the home impossible to keep. A deductible may be technically defensible and still empty the policy of value. Accurate information may close access to credit for years.

When housing, work and the body become fully priceable, they risk being treated as mere supports for a price [27]. They are also the settings in which a life is lived. Calculation can make their fragility visible. It does not acquire the right to decide alone who may continue to inhabit them.

A Politics of Insurability

A politics of insurability begins by naming the boundary. A refusal, extreme increase or transfer to a specialist market should identify the peril involved, the information used, the duration of the decision and the cover that can no longer be offered. The phrase “risk too high” is not enough when the consequence closes access to housing, credit or work.

It must then make return possible. A last-resort scheme should not become a permanent destination. A person or territory needs to know which work, change in circumstances or passage of time could reopen access to the ordinary market. When the necessary measures are beyond reach, policy should finance them rather than turn an inability to act into fault.

It must also limit institutional memory. A cured illness, repaid debt or completed sentence should not retain the same force indefinitely. An appeal must be able to challenge inaccurate information, the legitimacy of its use and the scale of the consequence.

These requirements do not remove financial limits or difficult choices. They prevent a boundary made by several institutions from presenting itself as a natural fact. An isolated person cannot see other refusals, compare reasons or challenge the rule. Their dispute remains confined to one file. The next chapter examines how people classified separately can turn their experiences into a common problem.

References

1.
Service-Public.fr. Crédit immobilier: À quoi sert la convention AERAS? [Internet]. Official public-service guidance; 2026 [cited 2026 July 16]. Available from: https://www.service-public.fr/particuliers/vosdroits/F20520
2.
Kousky C. Understanding disaster insurance: New tools for a more resilient future. Washington, DC: Island Press; 2022.
3.
Kunreuther HC, Michel-Kerjan EO, Doherty NA, Grace MF, Klein RW, Pauly MV. At war with the weather: Managing large-scale risks in a new era of catastrophes. Cambridge, MA: MIT Press; 2009.
4.
Kousky C, Kunreuther H. Addressing affordability in the national flood insurance program. Journal of Extreme Events. 2014;1(1):1450001.
5.
Federal Emergency Management Agency. An affordability framework for the national flood insurance program. U.S. Department of Homeland Security; 2018 Apr.
6.
Caisse Centrale de Réassurance. Observatoire de l’assurabilité des risques naturels : Premier rapport 2025. Paris: Caisse Centrale de Réassurance; 2026.
7.
Ferrera M. The boundaries of welfare: European integration and the new spatial politics of social protection. Oxford: Oxford University Press; 2005.
8.
Moss DA. When all else fails: Government as the ultimate risk manager. Cambridge, MA: Harvard University Press; 2002.
9.
Abraham KS. Distributing risk: Insurance, legal theory, and public policy. New Haven: Yale University Press; 1986.
10.
Koops BJ. The concept of function creep. Law, Innovation and Technology. 2021;13(1):29–56.
11.
Douglas M. Risk and blame: Essays in cultural theory. London; New York: Routledge; 1992.
12.
Beck U, Beck-Gernsheim E. Individualization: Institutionalized individualism and its social and political consequences. London: SAGE; 2002.
13.
Castel R, Haroche C. Propriété privée, propriété sociale, propriété de soi : Entretiens sur la construction de l’individu moderne. Paris: Fayard; 2001.
14.
Caplovitz D. The poor pay more: Consumer practices of low-income families. New York: The Free Press of Glencoe; 1963.
15.
Fourcade M, Healy K. Classification situations: Life-chances in the neoliberal era. Accounting, Organizations and Society. 2013;38(8):559–72.
16.
California FAIR Plan Association. About the FAIR Plan [Internet]. 2026 [cited 2026 July 6]. Available from: https://www.cfpnet.com/about-fair-plan/
17.
California Department of Insurance. Commissioner Lara adopts judge’s ruling on State Farm emergency rates, balancing consumer protections and financial solvency [Internet]. 2025 [cited 2026 July 6]. Available from: https://www.insurance.ca.gov/0400-news/0100-press-releases/2025/release038-2025.cfm
18.
Eaglesham J, Nelson LJ. A $44,000 bill shows the dysfunction in California’s home-insurance market. The Wall Street Journal [Internet]. 2026 [cited 2026 July 25]; Available from: https://www.wsj.com/finance/a-44-000-bill-shows-the-dysfunction-in-californias-home-insurance-market-bbe89a42
19.
Satz D. Why some things should not be for sale: The moral limits of markets. New York: Oxford University Press; 2010.
20.
Fourcade M, Healy K. The ordinal society. Cambridge, MA: Harvard University Press; 2024.
21.
Jones D. The uninsurable future: The climate threat to property insurance, and how to stop it. The Yale Law Journal Forum. 2025 Dec 3;181–236.
22.
Boutié É. La maison brûle : Cultiver le déni du changement climatique après un mégafeu en californie du nord [Internet] [Thèse de doctorat en anthropologie sociale et ethnologie]. [Paris]: École des hautes études en sciences sociales; 2024 [cited 2026 July 28]. Available from: https://theses.fr/2024EHES0039
23.
Koslov L. The case for retreat. Public Culture. 2016;28(2):359–87.
24.
Harcourt BE. Against prediction: Profiling, policing, and punishing in an actuarial age. Chicago: University of Chicago Press; 2007.
25.
Fleurbaey M. Fairness, responsibility, and welfare. Oxford: Oxford University Press; 2008.
26.
Scott JC. The moral economy of the peasant: Rebellion and subsistence in Southeast Asia. New Haven: Yale University Press; 1976.
27.
Polanyi K. The great transformation: The political and economic origins of our time. New York: Farrar and Rinehart; 1944.

  1. In insurance, moral hazard refers to the possibility that cover changes behaviour or precautions. The term does not necessarily imply moral wrongdoing.↩︎