If your company runs a captive, you already own an insurance company that retains losses caused by living systems you do not fund.
That is not a criticism of the captive. It is the most rational structure in corporate risk finance, and it is doing exactly what it was built to do. It is a criticism of where nature currently appears in it: in the claims file, after the event, as a residual — and almost nowhere else in the program.
Captive insurance is a licensed insurance company owned by the organization it insures, used to formally retain risk that the commercial market prices poorly, refuses, or cannot structure. Premiums move from operating units to a balance sheet the parent controls, and underwriting profit stays in the family rather than leaving as margin.
That control is the point of this post. When you own the underwriter, you can decide what the premium buys.
the market is large and no longer exotic
Captives stopped being a specialist tool some time ago. That matters here, because a sleeve is a program design question, and program design happens where the sophistication already is.
Marsh's 2026 Captives Benchmarking Report recorded 9% premium growth among Fortune 500 captive owners and described captives writing property, liability, cyber, trade credit, and intellectual property, with parametric and structured programs now standard features rather than experiments. Artex counted captive premium above $80 billion annually as of mid-2026, growing even while the commercial market softened.
Retained premium and the protection gap are both rising. They describe one problem from opposite ends: more risk is being kept on corporate balance sheets, and less of the underlying hazard is being reduced.
retention is the honest part
A captive tells the truth that a commercial policy obscures: you are keeping this risk.
Once that is on the page, the follow-on question is uncomfortable and useful. What are you doing about the physical cause of the losses you have chosen to keep?
For most retained property and business-interruption exposure, that cause has an address:
| the retained loss | the living system upstream of it |
|---|---|
| plant downtime from water restriction | the watershed and its snowpack, aquifer recharge, and upstream forest condition |
| coastal terminal damage and demurrage | the marsh, dune, and reef in front of the berth |
| distribution center flooding | forty miles of channel, floodplain storage, and upstream impervious cover |
| wildfire loss at a facility or resort | the fuel load and wet meadow structure on the ridge above it |
| crop and supply input volatility | soil organic matter, pollinator habitat, and field-scale water holding capacity |
None of these appear as a line in the captive's underwriting file. All of them appear in its loss experience.
the three places nature can sit in a captive program
Only one of them changes the loss.
| where nature sits | what it means | timing |
|---|---|---|
| the claims file | the flood, fire, or drought that caused the retained loss shows up as a paid claim | after |
| the premium allocation model | business units pay more or less into the captive based on how they manage physical risk | before, but it only reprices behavior |
| the condition-funding sleeve | a defined share of retained premium funds the present condition of the named system upstream of the exposure | now, and it changes the hazard |
The middle row is real practice, not theory, and the brokers got there first. WTW has written directly about using captives for "climate hedging" — quantifying retained climate exposure, then blending self-insurance, funded adaptation and mitigation measures, and commercial transfer, with premium allocation models that adjust what each business unit pays based on how it manages risk. Good idea, honestly described, and the nearest cousin to what we do.
Two more cousins, because they show the direction of travel:
- Michigan's NextGen crop insurance pilot stood up a captive specifically to price the reduced production risk associated with regenerative practices — cover crops, reduced tillage, diversified rotations — using field-level soil health data instead of historical yield alone.
- CILRIF, developed under the UN Capital Development Fund, structures long-term municipal climate insurance where the premium falls as the city hits adaptation milestones, with a linked financing facility for the infrastructure itself.
All three do something valuable: they make risk reduction pay inside an insurance structure. What none of them do is hold the upstream living system as an asset with a present condition that somebody funds continuously, whether or not a rating plan is ready to give credit for it. That is the gap the sleeve fills.
The standards world is moving in the same direction. CEN's CWA 18349:2026 is the first European guidance on nature-based insurance and investments, and it sets out when ecosystem protection and restoration are material enough to support underwriting and investment decisions. It is a useful document and it does not fund anything — which is the subject of when nature is material enough to underwrite, post 3 in this series.
what a condition-funding sleeve actually is
A sleeve is a defined allocation inside or beside your captive program that funds the present condition of one named natural system your retained losses depend on.
Mechanically, it looks like this:
- Name one dependency. Not "climate." One system, one geography, one retained line — the watershed above the plant, the marsh in front of the terminal.
- Establish present condition. What the system currently produces in the services you are exposed to: water regulation, flood attenuation, fire behavior modification, erosion control. Condition today — not a baseline year, not a counterfactual. We coordinate this work with technical partners; the accounting engine is ours.
- Size it against retained loss, not against a donation budget. A fraction of the retained expected loss on one line is the honest starting number. If it cannot be justified against retained loss, do not do it.
- Fund the condition. A certificate funds one named place directly. A coin funds protection across the protocol, indirectly, through market activity. Both are ordinary onchain instruments; neither is a policy.
- Hold the evidence beside the claims data. Present condition, funding routed, and loss experience on that line over three to five years — the horizon captive owners already use for structured reinsurance.
What the sleeve is not, stated flatly because ambiguity here would be dishonest:
- Not risk transfer. No limit, no attachment point, no trigger, no indemnity, no claims process.
- Not a replacement captive. We do not form, manage, license, or domicile captives, and we do not sell a captive product.
- Not a policy. Certificates are not policies in the insurance sense and should never be described to a regulator or an auditor as if they were.
- Not a credit. It is not a carbon or biodiversity offset, and it does not price a counterfactual.
- Not an admitted asset. Do not expect a domicile to count a certificate toward captive capital or reserves. Treat the sleeve as loss-prevention expense, wherever it sits.
It is a loss-prevention expenditure with an onchain record of where the money went and what condition it funded. Captive programs already carry loss-prevention spend — sprinklers, roof upgrades, flood barriers, monitoring. This is the same line pointed at the part of the hazard no single property can buy.
the numbers your CFO will ask for
The benefit-cost evidence for hazard reduction is decent, and worth quoting accurately rather than generously:
- $11 saved per $1 invested in adopting current model building codes; $6 per $1 for federal mitigation grants; $4 per $1 for above-code design — National Institute of Building Sciences, Natural Hazard Mitigation Saves, 2019.
- Median projected benefit-cost ratio of 1.86 across catastrophe adaptation projects — Swiss Re Institute, 2026.
Those are ratios for hazard mitigation at building, parcel, and program scale. They are not a return promise on any instrument, including ours. Read them as the reason the loss-prevention line exists at all, not as underwriting for a trade.
And our own position, plainly: the agents, coins, and certificates are live on Base with real but small volumes. This is a mechanism you can fund one dependency through, not a market you can allocate a program into. The unvarnished version is in is nature an asset class yet.
this is not tax or regulatory advice
We are not tax counsel, actuaries, captive managers, or your domicile regulator, and nothing here is advice on any of those.
Captive taxation and reporting are contested ground. The IRS issued final regulations in January 2025 identifying certain micro-captive arrangements as listed transactions and others as transactions of interest; in April 2026 a federal district court in the Southern District of Texas (Drake Plastics v. IRS) vacated the listed-transaction rule while upholding the transactions-of-interest rule, with appeals filed. Deductibility, risk distribution, domicile requirements, premium allocation methodology, and how any loss-prevention expenditure is treated for your entity are questions for your captive manager, actuary, auditor, and tax counsel.
Ask them before you fund a sleeve. If the answer is that it belongs outside the captive as ordinary loss-prevention spend by the parent, that is a fine answer and the mechanism works the same way.
frequently asked questions
what is a captive insurance company?
A captive insurance company is a licensed insurer owned by the organization it insures, formed to retain risk formally rather than buy it from the commercial market. Premiums move from operating units to a balance sheet the parent controls, underwriting profit is retained, and coverage can be structured for risks the market prices poorly or declines. There are more than 7,000 captives worldwide across 78 tracked domiciles.
can a captive fund nature-based solutions?
Yes, and some already fund adaptation and mitigation work. Brokers including WTW describe captives being used to fund risk mitigation directly — climate-resilient infrastructure and adaptation measures — alongside self-insurance and commercial transfer. What the market has not standardized is funding the present condition of a named living system upstream of the retained loss, continuously, rather than paying for a discrete project. Whether that spend sits inside the captive or at the parent is a structuring question for your advisors, not a philosophical one. The general investment case, outside the captive context, is in how to invest in nature-based solutions.
how do captives pay for loss prevention?
Three common routes. First, direct loss-prevention expenditure from captive surplus or the parent's risk budget — sprinklers, hardening, monitoring, hazard reduction. Second, premium allocation models that charge business units by how well they manage risk, funding prevention indirectly by making it pay. Third, structured programs where reduced retained loss over a multi-year window releases capital. A condition-funding sleeve uses the first route, pointed at landscape-scale hazard rather than parcel-scale hardening.
is ensurance insurance?
No. Insurance is ex post — it pays after a loss, and the business model requires the event. Credits are ex ante — they price a counterfactual. Ensurance is ex nunc, from now: it funds a system that is presently producing the services you are exposed to. No trigger, no limit, no claim. The definitional argument is in what is ensurance and the end of insurance as we know it.
one sleeve, one dependency
Pick the single retained line where your loss experience is most obviously a function of a living system with an address. Bring the geography, the retained expected loss on that line, and whatever condition data you already have. We will map the dependency, show what present condition looks like against it, and tell you honestly if the sleeve is too small to matter yet.
Start there, or look at what a named-place instrument actually is first — specific ensurance certificates, the natural assets they attach to, and the stocks and flows the condition is measured against.
Your captive is already financing the consequences. The sleeve is the first line in the program that is aimed at the cause.