Insuring the uninsurable
Marcus Schmalbach (pictured), of RYSKEX. explains what the Straits of Hormuz crisis can teach captives about public-private risk sharing.
There is a point at which an insurance problem ceases to be an insurance problem. The Strait of Hormuz might be showing us precisely where that point lies.
Marine insurers have priced war for generations. There is nothing unusual in a vessel entering dangerous waters attracting a higher premium, a larger deductible or more restrictive terms. Risk rises and price follows. Capacity contracts, additional information is demanded and underwriters become more selective. That is insurance doing what insurance is supposed to do.
But take that process far enough and the economics change. At some point the shipowner does not buy more expensive insurance. He does not negotiate another ten basis points or search London for another layer of capacity. He simply stops sailing.
The consequences extend rather further than the vessel. In normal conditions, roughly 20.9 million barrels of oil and petroleum liquids pass through Hormuz each day. That is equivalent to around 20 per cent of global petroleum liquids consumption and approximately a quarter of all oil traded by sea. There are alternative pipelines, but nowhere near enough capacity to replace the Strait. Hormuz is not merely a difficult marine exposure. It is infrastructure for the global economy.
That distinction is at the centre of what has happened in 2026. The problem confronting shipowners, charterers and their insurers has not merely been that the price of war risk increased. A strategically important trade route became exposed to a form of risk whose scale, correlation and political character strained the conventional boundary of private insurability. The result has been an unusually revealing experiment involving Washington, Chubb, some of the largest American insurance groups and, subsequently, the Lloyd’s market.
For the captive industry, it deserves considerably more attention than a conventional marine war-risk story would normally attract. Captives are not at the centre of the present Hormuz arrangements. But the architecture being assembled around the Strait raises a question that goes to the future of the sector: when a risk is too concentrated, too correlated or too strategically important for conventional insurance markets to absorb efficiently, could the captive become the institutional layer connecting the corporate balance sheet, the commercial insurance market and, ultimately, the state?
Insurance is exceptionally good at dealing with uncertainty. It is less comfortable with simultaneity. A factory fire in Manchester tells an insurer relatively little about whether another factory will burn in Singapore the following morning. Diversification works because losses are not perfectly correlated. A geopolitical confrontation affecting one of the world’s most important maritime choke points is different. A single political or military decision can alter the risk profile of hundreds of vessels at once. Hull, cargo, liability, business interruption and commodity exposures begin to move together. Historical data becomes less useful because the dominant variable is no longer engineering reliability, weather or individual operational behaviour, but geopolitics.
This is why the distinction between a large risk and a systemic risk matters. A systemic risk is not simply one capable of producing a very large claim. It is one able to undermine the diversification on which the insurance mechanism itself depends. Pandemic business interruption displayed the same characteristic. A sufficiently systemic cyber event could do so as well. Terrorism has done it before.
Hormuz makes the problem unusually visible because the consequence of insufficient risk-bearing capacity appears almost immediately in the real economy: vessels wait, routes change and trade slows. For an individual shipowner, avoiding the Strait might be a perfectly rational decision. For governments dependent on the energy, commodities and supply chains moving through the Gulf, thousands of owners making the same rational decision creates a very different problem.
That is the point at which private loss and social loss begin to diverge and where insurance becomes public policy.
The US response was remarkable both for its scale and for what it did not attempt to do. On 6 March, the US International Development Finance Corporation announced a revolving maritime reinsurance facility of approximately $20 billion intended to support commercial shipping in the Gulf. The DFC initially focused the programme on hull and machinery and cargo and coordinated its implementation with the US Treasury and Central Command.
Five days later, Chubb was named lead insurance partner. Under the arrangement, Chubb would act as lead underwriter and issue policies for eligible vessels, with DFC-backed capacity sitting behind the insurance structure. Chubb would manage the facility, determine pricing and terms, assume risk, issue policies and administer claims. Eligibility, meanwhile, would be determined in conjunction with the US Government.
That separation of responsibilities is more important than the headline number. Washington did not attempt to build a federal marine underwriting department. It placed commercial underwriting where commercial underwriting belongs.
The structure expanded quickly. In early April, DFC and Chubb announced that Travelers, Liberty Mutual, Berkshire Hathaway, AIG, Starr and CNA would join the programme. Their participation brought the stated capacity to $40 billion: $20 billion of rolling DFC support and a further $20 billion from Chubb and its private insurance and reinsurance partners.
For an industry accustomed to arguing about whether governments crowd out private insurance, Hormuz offers a more nuanced example. Chubb, Berkshire Hathaway, AIG and their peers were not being replaced by the sovereign balance sheet. The sovereign balance sheet was being used in an attempt to move the frontier at which their underwriting remained commercially viable.
Evan Greenberg’s Chubb is a significant choice in that respect. This was not a distressed-market rescue handed to a peripheral insurer. It placed one of the world’s largest property and casualty groups in the underwriting seat, with the state acting further down the risk chain. The public sector could supply exceptional capacity, but the market would still decide what constituted an acceptable risk and at what price.
That principle might prove more durable than the programme itself. Government should not replace underwriting where functioning underwriting expertise exists. Its role, where intervention is economically justified at all, should be to expand the boundary inside which private underwriting remains possible.
London’s response adds another dimension. In June, Lloyd’s welcomed a separate marine war-risk consortium led by Chubb, supported by participating Lloyd’s syndicates and specialist market partners. The consortium provides up to $200 million of capacity for hull and P&I risks and a further $200 million for cargo. Cover is accessed through brokers in the usual way and remains subject to individual risk assessment, underwriting criteria, sanctions screening, policy terms and exclusions.
That matters because Lloyd’s has spent more than three centuries doing something governments generally do badly: turning extraordinary and unfamiliar risks into risks that can, eventually, be priced. The market’s relevance to Hormuz is therefore not simply another pot of capital. It is the institutional machinery around that capital – brokers bringing granular risks to market, syndicates selecting exposures, specialist underwriters distinguishing one voyage from another and capital being allocated on differentiated terms rather than through a generic promise of indemnification.
When discussing protection gaps, there is a tendency to assume the principal problem is insufficient money. If $5 billion of capacity does not solve a problem, perhaps $20 billion will; if $20 billion proves insufficient, maybe $40 billion will. Hormuz exposes the weakness of that reasoning. A shipowner contemplating passage through an active conflict zone is not simply calculating whether the steel value of a vessel will be reimbursed after a missile strike, rather the safety of the crew, detention risk, sanctions, route availability, military developments, operational continuity and the possibility that the risk environment will change between departure and arrival.
Capital cannot solve all of those problems. Nor can a policy wording.
This is where risk governance becomes more useful than the narrower concept of risk transfer. Extreme risks become more insurable when capital is accompanied by selection, information, prevention, monitoring and intervention. Lloyd’s underwriting criteria matter. Chubb’s individual risk assessment matters. Government eligibility criteria matter. Military protection of navigation matters. Sanctions intelligence matters. So does the behaviour of the insured itself.
A missile cannot be diversified away.
For captive practitioners, this is where the story becomes particularly interesting. The traditional description of a captive – an insurance subsidiary used to retain risk, improve control over insurance economics and access reinsurance markets – remains correct. But it increasingly understates what a sophisticated captive can become. Properly constructed, it can serve as a risk-governance institution sitting between operating companies and external risk capital.
Consider a hypothetical shipping group, energy trader or consortium with substantial exposure to Hormuz. Rather than asking a commercial insurer or the state to absorb the exposure from dollar one, the risk could be deliberately layered. The corporate might retain an initial deductible on its own balance sheet. Its captive could then assume a defined first-loss layer – say the next $10 million or $25 million across an agreed portfolio of voyages – giving the group meaningful skin in the game.
Commercial marine insurers and Lloyd’s syndicates could attach above that retention, with conventional reinsurance providing additional severity capacity. Only at a much higher attachment point, where losses become genuinely systemic rather than merely severe, would sovereign or multilateral capacity enter the tower.
The numbers are illustrative; the architecture is the point. The captive is not there simply to make the insurance tower taller. Its retention changes the incentives below it and the information available above it.
Eligibility for the captive layer could depend on vessel standards, security protocols, routing rules, intelligence requirements, crew procedures, sanctions compliance and real-time monitoring. Loss information accumulates within a defined institutional structure. Better risks can be distinguished from poorer ones before they reach Chubb, a Lloyd’s syndicate or a reinsurer. The corporate has capital at risk if its own risk controls fail, while public capacity is pushed away from expected loss and towards the exceptional tail for which intervention can actually be justified.
In that architecture, the captive performs a function neither the state nor the conventional commercial market manages particularly well on its own. It translates corporate behaviour into an insurable risk profile.
There is precedent for this closer to home than the Hormuz debate might suggest. Following September 11, the withdrawal of terrorism capacity threatened activity far beyond the insurance industry. The US response through the Terrorism Risk Insurance Act created a public-private system in which private insurers continued to write the underlying risk while the federal government provided a defined backstop for qualifying catastrophic losses.
Captives are not a theoretical addition to that architecture. They are already inside it. The US Treasury operates a dedicated captive-insurer reporting framework under the Terrorism Risk Insurance Program (TRIP), and its programme data show hundreds of domestic US captives writing terrorism insurance subject to TRIP. In 2023 alone, 615 domestic US captive insurers did so.
That precedent matters. It demonstrates that the proposition at the heart of the Hormuz thought experiment – corporate risk retention, private insurance and a sovereign tail-risk mechanism existing within the same architecture – is not alien to modern insurance. We already do it.
The question is where e lse that logic might become useful.
Cyber is an obvious candidate. A ransomware attack against one company is an insurance event. A simultaneous failure or compromise of digital infrastructure relied upon by tens of thousands of companies is something rather different. Pandemic business interruption offered an even-clearer demonstration: diversification becomes severely constrained if a shutdown or pathogen produces losses across almost every insured at approximately the same time.
Natural catastrophe pools already employ public-private mechanisms in several jurisdictions. Emerging technologies and strategic infrastructure – hydrogen, carbon capture, advanced nuclear technologies and critical supply chains among them – raise another version of the problem, where the constraint might be the absence of sufficiently mature loss histories rather than certainty of extreme correlation.
The risks differ, but the allocation problem does not. Which losses should remain with the corporate because it can control them? Which belong in a captive because they are retainable and can be governed as a portfolio? Which can the commercial market efficiently price and diversify? Where does conventional reinsurance belong? And, in the rare cases where public intervention is justified, how remote should the sovereign attachment point be?
Those questions also expose the danger in the argument. Not every protection gap warrants a government backstop. Public insurance can create moral hazard, suppress proper pricing and leave taxpayers underwriting exposures they neither understand nor control. A temporary intervention can become a permanent subsidy. The fact that an economic activity is strategically fashionable does not create an automatic claim on the sovereign balance sheet.
A properly designed public-private architecture should therefore make government intervention harder, not easier. Before public capital attaches, the corporate should be able to explain what it retains, the captive what it governs, the commercial market what it will write and the reinsurer what it can diversify. Only then does the remaining question become legitimate: is there a genuinely systemic layer whose economic consequences justify putting sovereign capital at risk?
Hormuz is interesting not because Washington has discovered how to make dangerous shipping safe, nor because Chubb or Lloyd’s can somehow price away geopolitics. It is so because some of the world’s most sophisticated risk-bearing institutions are being forced to confront, in real time, the boundary between an expensive risk and one conventional markets can no longer support on their own.
For those of us who spend time around Lime Street, there is something reassuringly familiar in the market’s response. Lloyd’s did not answer the problem by declaring the Strait either insurable or uninsurable. It brought together underwriting expertise, specialist capacity, broker distribution and individual risk selection and asked the much more useful question: under what conditions can we write it?
The captive industry should ask the same question.
For much of its history, the captive has been a different way of financing risks that were already understood. Its next chapter might be rather more ambitious. At the intersection of corporate retention, commercial underwriting, reinsurance and, in exceptional circumstances, sovereign capacity, it can become the institution in which difficult risks are selected, governed and prepared for external capital.
The future of captives might therefore not be about insuring more risk, but about making the uninsurable insurable.
Marcus Schmalbach is the co-founder and chief executive of RYSKEX [ryskex.ai]. He can be contacted at schmalbach@ryskex.com.
Did you get value from this story? Sign up to our free newsletters and get stories like this sent straight to your inbox.
