What captive insurance is
A captive is an insurance company that exists to insure its owner. A business — or a group of businesses — forms and capitalizes a licensed insurer, usually in a jurisdiction whose laws are written for captives, and that insurer issues policies to the parent and takes on specified pieces of its risk. Premiums flow from the operating business into the captive, claims are paid out of the captive when covered losses occur, and money the captive does not pay out in claims remains the owner’s — held as the insurer’s reserves and surplus rather than spent into the commercial market. It is best understood not as a discount on insurance but as a change of machinery: instead of renting an insurer’s balance sheet, the business builds a small one of its own and keeps only the risks it is prepared to own. The risks a company is not prepared to own still get transferred — captives commonly buy reinsurance above the layer they retain, and many commercial coverages stay in the commercial market entirely. A captive typically takes a defined slice: a deductible layer, a specific line, or risks the market prices badly or will not write at all.
Who forms one, and why
The honest profile of a captive owner: a business large enough that insurance is a material, recurring cost; stable enough to have a loss history it more or less believes; and disciplined enough to run an insurance company, because that is literally the job it is taking on. The motivations are consistent. Control — the owner chooses the policy language, the claims handling, and which risks the captive writes, instead of accepting a market form. Economics — premiums paid for risks that run better than the market priced stay in the family, and the captive gains access to reinsurance pricing that only insurers can buy. Coverage the market will not sell — a captive can write a narrow, well-understood risk the commercial market declines or prices out of reach. Groups of similar businesses sometimes form group captives to pool what none could efficiently hold alone, spreading fixed costs and volatility across members. What unites every sound reason is that the owner already has scale and predictability; the captive is a way to keep the benefit of both.
The honest scale threshold — who should not form one
Read plainly: a captive is not a small-shop product, and it is not a tax trick with an insurance policy wrapped around it. Forming one means capitalization up front, licensing, an annual audit, actuarial work, a captive manager, regulatory filings, and board-level governance of an actual regulated insurer, every year, whether claims cooperate or not. Those fixed costs do not shrink for a small premium base, which is why the arithmetic only begins to work for businesses whose insurance spend is large and steady. A small business that forms a captive to insure risks it cannot afford to lose has not transferred its risk at all — it has concentrated it, at additional expense, in a company it owns. Captives formed primarily for tax posture, with premiums set without regard to real risk, have drawn sustained regulatory challenge; the structure rewards genuine risk-bearing and punishes costume. If your insurance problem is that premiums feel high for the coverage you get, the market offers ordinary instruments first — higher deductibles, better loss control, cleaner risk presentation. A captive enters the conversation when those are exhausted and the premium volume is large enough to run an insurer on. Not before.
One scenario, composited
A regional logistics company with a large vehicle fleet spends heavily, year after year, on the first layer of its auto liability exposure — the frequent, predictable claims that come with running trucks every day. Its loss record on that layer is steadier than the market’s pricing of it. The company forms a captive to take that layer: premiums for the predictable slice flow into its own insurer, claims on that layer are handled under its own discipline, and coverage above the layer is still bought commercially, where it belongs. In a good year the captive keeps most of the premium; the company has converted a retail purchase into a managed cost. Now the contrast: imagine the same company captiving its rare, severe exposure instead — the one catastrophic claim a decade. A single bad year would meet a thinly capitalized insurer owned by the very people who suffered the loss, and both balance sheets would take the blow at once. Predictable frequency is what captives digest well; severity belongs with someone else’s capital. Choosing the wrong slice is how a sound tool produces an unsound result.
The one question to ask your broker
“Which of my risks are frequent and predictable enough that retaining them through a captive beats transferring them — and is my total premium volume large enough to run a regulated insurer on top of them?”
If the answer names specific layers and specific economics, the conversation is real. If it dwells on deductions and offshore addresses, you are being sold the costume. A captive should be the last step of a mature risk program, arrived at deliberately — not the first step of a sales pitch.
Related reading: Excess-of-loss layers, explained · How January reinsurance renewals set your commercial premiums