Most employers searching for a captive health insurance plan are self-funded or close to it, and want a way to stabilize what they pay for stop-loss. A stop-loss captive is one option. It lets a group of employers, or one large employer, keep part of the large-claims risk that a conventional stop-loss carrier would otherwise price and keep.
This page explains how stop-loss captives are built, how money moves through them, and what to check before joining one.
What is a captive health insurance plan?
A captive health insurance plan is a self-funded health plan whose stop-loss coverage is financed in part through a captive insurance company. The NAIC describes a captive as an insurer owned by the business or businesses it insures. In employee health benefits, captives almost always sit behind the employer’s stop-loss insurance, not in front of the employees’ health plan.
That distinction matters:
- Your health plan stays self-funded. You still pay claims through your TPA up to your specific and aggregate attachment points.
- The captive insures the employer, not the employees. Like any stop-loss coverage, it reimburses the employer or plan for large claims. The DOL notes that stop-loss generally is not treated as health insurance under state law.
- Each employer keeps its own plan. In a group captive, employers share stop-loss risk through the captive, but each sponsors its own plan with its own plan document.
A stop-loss captive is a way to finance large claims. It does not change plan design, networks or employee cost sharing by itself.
How do the layers stack up?
A stop-loss captive splits large-claim risk into three layers. Each employer pays the first layer, the captive pays the middle layer from pooled premium, and a reinsurer or stop-loss carrier pays the top layer.
| Layer | Who pays | Typical range (hypothetical) | What it means for you |
|---|---|---|---|
| Employer retention | Your plan | Claims up to your specific deductible, for example $50,000 per person | The same as conventional self-funding. You carry routine claims and each person’s claims up to the deductible. |
| Captive layer | The captive, from premiums all members paid in | From your deductible up to a set limit, for example $50,000 to $250,000 per claim | The shared layer. Low claims here across the group create surplus; high claims use it up. |
| Reinsurance or excess layer | A reinsurer or the stop-loss carrier | Above the captive layer limit | Catastrophic claims, such as transplants or long NICU stays, are passed to a larger insurer. |
The dollar figures above are only an illustration. Each captive sets its own layers, and many also include an aggregate layer that responds when a member’s total claims pass its aggregate attachment point (see aggregate stop-loss).
In many group medical stop-loss captives, a licensed stop-loss carrier issues a conventional policy to each employer, then reinsures the captive layer back to the captive. That arrangement is often called “fronting.” It means your policy, claims filing and reimbursement look much like ordinary stop-loss, while the captive carries part of the risk behind the scenes.
Because the front policy is still stop-loss, state stop-loss rules apply to it. The NAIC Stop Loss Insurance Model Act bars a specific attachment point below $20,000 and sets minimum aggregate attachment points, though states that adopt it may change those amounts and many states have no minimum at all.
What is the difference between group and single-parent captives?
The two main structures differ in who owns the captive and who shares the risk. A single-parent captive insures one company and its affiliates. A group captive pools the risk of several unrelated employers.
| Single-parent (pure) captive | Group captive | |
|---|---|---|
| Owner | One company | Several member employers, or a sponsor that rents cells to them |
| Who shares risk | Only the parent and its affiliates | All members share the captive layer |
| Typical employer size | Large employers with enough claims to be predictable on their own | Small and mid-sized self-funded employers |
| Control | Full control of layers, vendors and surplus | Shared governance; rules set by the captive’s manager or board |
| Startup and running costs | The parent bears formation, capital and management costs | Spread across members |
| Surplus | All stays with the parent | Shared under the captive’s formula |
| Effect of others’ claims | None | Other members’ claims affect your dividend and capital calls |
Group captives are the common route for employers with a few hundred lives or fewer. Some are homogeneous (members from one industry) and some are heterogeneous (members from many industries). Many are organized as cells inside a sponsored captive, which Vermont and other domiciles allow under their captive laws, so each employer group’s assets can be kept separate from other cells.
Single-parent captives are usually used by large organizations that already insure property or casualty risks through their own captive and want to add medical stop-loss. Formation, capital, management and regulatory reporting fall on that one company.
Captives are licensed in a domicile. The NAIC reports that more than 70 jurisdictions have captive legislation, and that Vermont is the largest U.S. domicile. The domicile’s regulator, such as Vermont’s Department of Financial Regulation, licenses the captive, examines it and monitors its solvency.
How do dividends and surplus work?
A captive returns money only when the captive layer’s premiums, plus investment income, exceed the claims and expenses charged to it after all claims for that policy year are paid. The return is often called a dividend or surplus distribution.
How it typically works:
- Premium is split. Part of each member’s stop-loss premium goes to the fronting carrier and reinsurer, part to captive expenses, and part funds the captive layer.
- Claims are paid during the year and run-out. Large claims in the captive layer draw down the pooled fund.
- The year is closed. The captive waits until run-out claims are paid and reserves are settled, which often takes a year or more after the policy year ends.
- Surplus is allocated. The captive’s documents set the formula. Some return surplus in proportion to premium paid, some weight each member’s own claims experience, and many hold back part of the surplus as a reserve.
A dividend is never guaranteed. In a bad year the captive layer can be fully used, leaving nothing to return, and in a very bad year it can need more money.
If employees pay part of the premium, ask ERISA counsel how any dividend or surplus is treated and used before you book it as company income.
What are collateral and capital calls?
Joining a captive usually means putting up money beyond premium. That money backs the captive layer in case claims exceed the premium set aside for it.
- Collateral or capital contribution. New members often fund a collateral amount at entry, in cash or by letter of credit, sized to the member’s share of the captive layer. It is usually held until the member exits and its policy years are closed.
- Loss fund. Some captives hold part of each year’s premium as a loss fund that pays captive layer claims first.
- Capital calls or assessments. If losses exceed the loss fund and collateral, members can be asked for more money. The documents should say how calls are sized, whether they are capped, and whether your share depends on your own claims or the group’s.
Put the collateral amount, the maximum possible assessment and the release schedule in writing before you commit. Those numbers set your worst case.
How do you evaluate a captive?
Evaluate a captive the way you would evaluate any long-term financial commitment: who runs it, who stands behind it, who else is in it, and how you get out.
| Question | What to look for |
|---|---|
| Who manages it? | The captive manager and program manager, their experience with medical stop-loss, and whether audited financials and actuarial reports are shared with members |
| Who is the fronting carrier and reinsurer? | Licensed carriers with strong financial strength ratings, and how long they have backed this program |
| Where is it domiciled? | The domicile and regulator, and whether the program uses segregated cells |
| How are members selected? | Underwriting standards for new members, minimum group size, and how the captive handles members with poor results |
| What are the layers? | Your specific deductible, the captive layer limit, the aggregate terms, and the contract basis, such as 12/12 or 12/15 |
| What vendors are required? | Required TPAs, PBMs, care management or cost containment programs, and whether you can keep your current vendors |
| How is surplus shared? | The dividend formula, timing, and how much is held in reserve |
| What do you put at risk? | Collateral, loss fund and maximum assessment, all in dollars |
| What are the exit terms? | Notice period, run-out treatment, when collateral and surplus are released, and what happens if the fronting carrier or reinsurer leaves |
| What data do you get? | Your own claims data and the captive’s pooled results, on a regular schedule |
Also compare the captive quote against conventional stop-loss on the same contract basis, with the same deductibles and the same laser terms. A captive may still laser high-cost members, and lasers or new-member underwriting can shift risk back to you. Ethos procures and audits stop-loss, which can include reviewing a captive quote alongside conventional quotes, through its stop-loss service.
How does a captive compare with conventional stop-loss?
The main difference is who keeps the middle layer of risk. With conventional stop-loss the carrier keeps it, along with any profit or loss. With a captive, the members keep it.
| Conventional stop-loss | Stop-loss captive | |
|---|---|---|
| Who keeps the middle layer | The carrier | The captive’s members |
| Upside in a good year | None, though it may help your renewal | Possible surplus distribution after run-out |
| Downside in a bad year | Higher renewal | Lost surplus, possible capital call, and possibly a higher renewal |
| Cash beyond premium | None | Collateral and loss fund |
| Vendor choice | Usually broad | Often limited to approved vendors |
| Exit | At renewal, subject to run-out and run-in terms | At renewal, with notice, and with collateral and surplus held until prior years close |
| Renewal stability | Depends on your claims and the market | Pooling can smooth renewals for the captive layer, but results still vary |
What are the risks of a stop-loss captive?
A stop-loss captive moves risk from a carrier back to employers. That can pay off, but the risks are real.
- Shared losses. Other members’ large claims reduce or erase your dividend and can trigger a capital call.
- Cash tied up. Collateral and loss funds can sit with the captive for several years.
- Slow returns. Surplus is paid only after run-out, often more than a year after the policy year ends.
- Hard exits. Leaving usually requires notice at renewal, and your money stays until your old years close.
- Vendor lock-in. Required TPAs, PBMs or programs may not fit your plan.
- Program changes. If the fronting carrier or reinsurer exits, the program may need to be re-placed on new terms.
- Underlying self-funding risks. You still carry claims up to your deductible, along with terminal liability, lasering and plan compliance duties. See self-funded health insurance pros and cons.
Who is a captive a good fit for?
A stop-loss captive tends to fit an employer that is already self-funded or ready to be, has stable cash flow, can put up collateral without strain, and plans to stay in the arrangement for several years. Employers that need predictable monthly costs above all else, or that may change funding in a year or two, often do better with conventional stop-loss or a level-funded plan.
Captives are not only for large companies, since group captives were designed for smaller employers. But they are not automatically cheaper than conventional stop-loss either. Whether one makes sense depends on your claims history, your tolerance for shared risk and the specific program’s terms.
