The 100-employee threshold gets cited constantly in conversations about self-funding eligibility — and like most rules of thumb in the benefits industry, it's partially right, often misapplied, and insufficient on its own to answer the actual question. This article lays out the real criteria for self-funding candidacy: what matters, what doesn't matter as much as people think, and what the analysis actually looks like when done correctly.
Why group size matters — and what it's really a proxy for
Group size isn't important because of some arbitrary regulatory threshold. It's important because of statistical credibility. A self-funded health plan is, at its core, a bet that you can predict your claims costs well enough to manage your cash flow and structure appropriate stop-loss protection. Prediction requires data. Data requires volume. Volume comes from headcount.
With 50 employees, a single catastrophic claim can represent 30–40% of your expected annual claims spend. Your experience in any given year is dominated by the random occurrence of serious illness — cancer, premature birth, transplant — rather than the underlying health patterns of your population. Stop-loss helps, but at this size it's expensive relative to premium, and your net savings opportunity shrinks accordingly.
With 300 employees, the law of large numbers starts to work in your favor. Your claims profile becomes more predictable year-over-year. Stop-loss is competitively priced because carriers have more data to price against. The structural savings from eliminating carrier margin and restructuring pharmacy are large enough relative to plan costs to produce a compelling financial result.
The factors that matter more than headcount
Size is a threshold condition, not a sufficient condition. Within each size band, these factors determine whether self-funding makes sense — and how compelling the economics are.
Claims history and loss ratio
Three years of claims data is the standard for a credible actuarial analysis. What we're assessing: your historical loss ratio (claims paid divided by premium), the trend in that ratio over time, and the concentration of claims. A group with a 68% loss ratio trending downward, claims distributed across the population, and no recurring high-cost claimants is a much better self-funding candidate than a group with a 92% loss ratio, two members each running $200,000+ annually, and pharmacy costs dominated by specialty drugs.
Counterintuitively, a group with a recent bad year is sometimes an excellent candidate. If the adverse experience was driven by a one-time event — a difficult pregnancy, an accident — rather than chronic conditions, and the affected members are no longer in the plan, the actuarial picture may be quite favorable. Stop-loss carriers are pricing prospective risk, not sunk cost.
Demographics and workforce composition
Age distribution is the most significant demographic variable. A workforce with an average age of 38 has materially lower expected claims than one with an average age of 52. That doesn't make older workforces bad self-funding candidates — it means the savings opportunity looks different. The structural cost reduction from eliminating carrier margin is the same regardless of demographics. The pharmacy savings opportunity may actually be larger in an older workforce with more chronic conditions, if the PBM relationship is restructured appropriately.
Geographic concentration matters for network design. A single-site employer in a major metro area has straightforward access to competitive network options. A multi-state employer with employees in rural markets needs more careful TPA and network selection to ensure adequate provider access.
Financial capacity
Self-funding requires balance sheet liquidity that fully insured plans do not. Specifically, you need to fund a claims reserve at inception — typically 60 to 90 days of expected claims — and you need to be able to absorb monthly claims variability without that variability creating an operational cash flow problem. For most mid-market employers, the reserve requirement is manageable. For companies operating with very tight working capital or significant seasonal cash flow variation, the timing of the reserve funding needs careful planning.
The level-funding question
Level-funding is a structure that sits between fully insured and fully self-funded. You pay a fixed monthly amount — like a fully insured premium — but the arrangement includes a claims account, and if your claims come in below your funded amount, you receive a refund at year-end. Carriers typically wrap level-funded arrangements with built-in stop-loss, and you get limited claims data visibility.
Level-funding makes sense as a first step for employers in the 50–150 employee range who want data visibility and some economic upside without the full claims variability exposure of self-funding. It is not a permanent solution for groups that are genuinely good self-funding candidates — the carrier is still capturing margin, the stop-loss terms are typically less favorable than you'd get in the open market, and the data access is limited compared to a true self-funded arrangement.
Our general guidance: if you're 100+ employees with a reasonable claims history, the analysis should start with full self-funding as the baseline, with level-funding as a fallback only if the actuarial picture makes full self-funding genuinely risky.
The 100-employee threshold is a reasonable starting point for a conversation, not a definitive answer. We've recommended against self-funding for groups of 400 employees with problematic claims profiles, and we've recommended self-funding for groups of 85 employees with demographics and history that made the economics compelling. The analysis is what matters. Get the data, run the numbers, and let the results guide the decision.