Employers looking past a fully insured renewal are usually shown two alternatives that sound similar and are not: a level-funded plan, and membership in a medical stop-loss captive. Both self-fund claims and cap the downside with stop-loss. What differs is whether you are buying a product or joining an insurance company. This is the head-to-head for groups in the 25 to 250 employee range.
Both level funding and a medical stop-loss captive are ways of self-funding your health plan while capping the downside with stop-loss insurance. Employers frequently hear them presented as two points on the same spectrum, which understates how different they are to own.
The cleanest way to hold the distinction: level funding is a product you buy. A captive is a structure you join and partly own. Everything else follows from that.
What Level Funding Actually Is
Under a level-funded arrangement you self-fund claims, but you pay a fixed monthly amount that bundles three things: an amount for expected claims, the administrative fee, and the stop-loss premium. The month-to-month experience feels like a fully insured premium, which is the point — it is self-funding packaged to behave predictably.
Stop-loss sits inside the package in two layers. Specific (or individual) stop-loss caps what the plan pays on any one person. Aggregate stop-loss caps total claims across the group for the year. Because both are embedded, your maximum exposure for the year is known before it starts.
If claims come in below the funded amount, the contract may return some of the surplus after the run-out period closes. This is the provision to read carefully rather than assume: whether a refund exists at all, what share of the surplus it represents, how long after year-end it arrives, and whether it survives your leaving the arrangement all vary by contract. A level-funded plan sold on the promise of a surplus return, under a contract that returns surplus only if you renew, is a different product from the one described in the pitch.
Level funding is commonly available to groups well under 100 employees, sometimes from around 10 to 25 lives, usually with medical questionnaires or other underwriting at entry. The commitment is typically one year.
What a Medical Stop-Loss Captive Actually Is
In a group medical stop-loss captive you self-fund your plan and buy specific stop-loss as usual, but instead of transferring the whole excess layer to a commercial carrier, you and the other member employers jointly take a slice of it through a captive insurance company you collectively own.
The risk is sliced into three:
- The layer you retain. Claims below your specific deductible, funded by you, exactly as in any self-funded plan.
- The captive layer. A band of risk above your retention that the captive underwrites. Some programs have each member bear their own experience in this layer first and pool only above that; others pool from the bottom. Which design you are in determines how much another member’s bad year can cost you.
- The reinsurance layer. Catastrophic claims above the captive’s limit, transferred to a commercial reinsurer.
Because the captive is an insurance company, joining means capitalizing it. You post collateral, usually a letter of credit or cash contribution, and that capital is genuinely at risk. In a good year the captive layer produces underwriting profit and investment income that can be distributed to members. In a bad one, members can face assessments above their premium, and that possibility is the part most worth understanding before you sign.
Programs commonly want at least 50 to 100 enrolled employees, though this varies, and they generally expect a multi-year commitment — three years is typical. Members usually participate in governance, which means real meetings and real decisions about the program’s underwriting and vendors.
Side by Side
| Level-Funded | Medical Stop-Loss Captive | |
|---|---|---|
| What you are buying | A packaged self-funded plan from a carrier or TPA | Membership and an equity stake in a jointly owned insurance company |
| Typical minimum size | Often available from roughly 10–25 employees | Commonly 50–100 or more enrolled employees, varying by program |
| Commitment | Usually one year | Usually three years, sometimes longer |
| Capital required | None beyond the monthly funding | Collateral or a capital contribution, genuinely at risk |
| Monthly cash flow | Fixed and predictable | Variable — claims are paid as incurred, plus fixed costs and captive premium |
| Upside if claims are low | Possible surplus return, governed entirely by the contract terms | Distribution of underwriting profit and investment income from the captive layer |
| Downside if claims are high | Capped by embedded aggregate stop-loss | Capped at your retention, but assessments are possible in the pooled layer |
| Claims transparency | Reporting is provided, though often limited | Full claims data, which is the basis of the whole model |
| Vendor control | Generally the carrier’s bundled network, TPA and PBM | Ability to select TPA, network and PBM, including pass-through pharmacy terms |
| Administrative burden | Low — close to fully insured | Meaningful — governance participation and active plan management |
| Exiting | Straightforward at renewal, though watch surplus provisions | Harder — collateral release and run-off of your share of the risk take time |
Read the first and last rows together, because they are the same fact stated twice. The features that make a captive attractive — owning the underwriting margin, controlling the vendors, seeing all the data — exist because you took on the obligations of an insurer. The features that make level funding attractive exist because you did not.
Which One Fits Your Size
Group size does not settle the question on its own, but it narrows it considerably.
- Under about 50 employees. Level funding is usually the only one of the two available, and often a sensible step up from fully insured if your claims experience is favorable. Most captive programs will not take a group this size, and those that will are asking you to post capital against a very small claims base, where one catastrophic claim dominates your own experience.
- Roughly 50 to 100 employees. Both become possible, and the decision turns on cash position and temperament more than on arithmetic. A captive asks for a three-year view and capital you cannot easily recall. If the business could not comfortably absorb a bad claims year without that capital, level funding is the more honest fit whatever the projected return says.
- Roughly 100 to 250 employees. This is where captives earn their keep for employers who want them. The claims base is large enough that your own experience means something, you can support the governance work, and vendor and pharmacy control starts to be worth real money — particularly where specialty drug spend is concentrated.
- Above 250 employees. A captive remains viable, but standalone self-funding with conventional stop-loss is also fully practical, so the captive has to justify itself against that rather than against being fully insured.
What to Ask Before Signing Either
- For level funding: what exactly happens to a surplus? Get the provision in writing — what share is returned, how long after the run-out period it arrives, and whether you receive it if you do not renew.
- For level funding: how is the claims funding amount set at renewal, and what happens to your fixed monthly payment if claims exceed the funded amount mid-year?
- For a captive: is the captive layer pooled from the first dollar, or does each member absorb their own experience first? This single answer determines how exposed you are to the other members.
- For a captive: what is the maximum assessment in a bad year, stated as a number, and has the program ever levied one?
- For a captive: what does exiting cost, and how long does collateral release take? Ask for the actual history of members who have left.
- For both: is stop-loss written on a paid or an incurred basis, and are there laser provisions excluding named individuals? A lasered high-cost claimant quietly moves that risk back onto you.
These are the questions that decide the outcome, and none of them are answered by the projected-savings page at the front of the proposal.
Where This Sits Alongside Your Other Options
If you are still deciding between fully insured, level-funded, self-funded and defined contribution in the first place, start with the broader comparison in our guide to health plan funding options compared. For more detail on the captive structure itself, including feasibility and the joining process, see group health captives for Michigan employers.
We model these against your own census, claims experience where the carrier will release it, and current renewal, because the comparison is entirely specific to those inputs. We do not quote an expected saving for either structure in advance — with a captive in particular, the honest answer depends on claims that have not happened yet.
Frequently Asked Questions
Is a captive better than level funding?
Neither is better in general. Level funding gives you predictable monthly cost, no capital at risk and an easy exit, in exchange for limited transparency and limited control over vendors. A captive gives you full claims data, control of the TPA, network and pharmacy contract, and a share of the underwriting margin, in exchange for capital at risk, a multi-year commitment, possible assessments in a bad year and real administrative work. The question is not which is superior but which set of tradeoffs suits your size, your cash position and how much of this you want to manage.
How many employees do you need for a medical stop-loss captive?
Most group medical captive programs look for at least 50 to 100 enrolled employees, though the threshold varies by program and some will consider smaller groups. Below that, your own claims experience is too small a sample for the model to work as intended — a single catastrophic claim dominates it — and the capital you post is large relative to the risk being managed. Level funding is generally available considerably lower, sometimes from around 10 to 25 lives.
Can you lose money in a captive?
Yes, and this is the part that deserves emphasis. The capital or collateral you post is at risk, and if the captive layer performs badly members can face assessments above the premium they budgeted. Your exposure on your own claims is still capped by your specific and aggregate stop-loss, so the catastrophic risk is transferred, but the pooled layer is genuine insurance risk that you now partly carry. Before joining, ask for the maximum assessment as a number and whether the program has ever levied one.
Do level-funded plans really refund surplus?
Some do and some effectively do not, and the difference is in the contract rather than the brochure. Ask what share of any surplus is returned, how long after the plan year and its run-out period the payment arrives — often twelve to eighteen months — and critically whether you still receive it if you do not renew. A surplus provision contingent on renewal is a retention device, not a refund, and it changes what the arrangement is worth.
Is level funding the same as self-funding?
Technically yes: in both, the plan pays claims out of employer funds rather than transferring them to an insurer for a premium. The difference is packaging. Level funding bundles expected claims, administration and stop-loss into one fixed monthly amount, so the employer experiences it like a premium. Conventional self-funding has the employer pay claims as they are incurred, which means variable monthly cash flow and a need for reserves. Level funding is self-funding arranged to remove the cash-flow volatility.
What is the difference between specific and aggregate stop-loss?
Specific stop-loss, sometimes called individual stop-loss, caps what your plan pays on any one covered person in a plan year — claims above that deductible are reimbursed by the stop-loss carrier. Aggregate stop-loss caps the total claims your plan pays across everyone for the year. You need both to have a genuinely capped worst case: specific protects against one catastrophic claimant, aggregate protects against many moderate claims arriving at once.
Can you move from level-funded into a captive later?
Often yes, and it is a common sequence. A few years of level funding gives you the two things a captive entry needs: claims experience the underwriters can price, and internal familiarity with how a self-funded plan behaves. Two cautions. Check what happens to any outstanding surplus when you leave the level-funded arrangement, and be sure your claims data belongs to you in a form you can hand to a new TPA — employers sometimes discover at this point that the detailed data they assumed was theirs is not readily portable.
Unhappy with your current broker? Switch to CFH. Your employees won’t notice. You will.
