The Level Funded Workforce: Who These Plans Actually Cover and Who They Miss

LFP-06.01 | Sharp Analysis | Series 06: The Populations

The industry describes the level funded population in actuarial terms: age and gender bands, geographic distribution, industry classification codes. Stop loss carriers price against this abstraction. TPAs build products around it. Brokers sell into the segments the abstraction identifies as viable. The abstraction is functional for pricing. It is inadequate for understanding who level funded actually serves and, just as importantly, who it nominally covers while failing.

The most consequential feature of the level funded market in 2025 is not its growth rate. It is the growing gap between the population the product was designed for and the population that actually works for small employers in the industries where level funded has taken root. That gap is structural. It maps to design assumptions that were reasonable for the model’s original target market and are increasingly misaligned with the workforce that now depends on it.

The Assumed Population

#

Level funded plan design and stop loss underwriting are built on a specific picture of the worker. The picture has five features, each embedded in the plan mechanics, and each worth naming explicitly because the rest of this series is about what happens when any of them does not hold.

The first assumption is full-time employment sustained across a plan year. Plan eligibility typically requires 30 or more hours per week, matching the ACA’s definition of full-time status. The cost-sharing architecture assumes 12 months of continuous enrollment: deductibles accumulate toward an annual limit, out-of-pocket maximums reset at year end, and surplus reconciliation occurs after the plan year closes. A worker employed for four months and terminated provides a much shorter window of contribution relative to potential claims.

The second assumption is a single-employer relationship. The employer-sponsored insurance system was architected for one worker tied to one employer. ERISA’s plan sponsor requirements assume that a single employer can warrant to the stop loss carrier who is in the group, what those people earn, and how long they have worked. Group underwriting depends on the employer’s ability to represent the pool.

The third assumption is income adequate to afford cost sharing. A $2,500 deductible is a planning exercise for a household earning $75,000 annually. It is a care-rationing mechanism for a household earning $28,000. Plan design parameters are set without reference to the actual income distribution of the covered population. The product makes no distinction between a plan serving a professional services firm and a plan serving a home health agency, even though the income distributions of those workforces differ by a factor of two or more.

The fourth assumption is health status within the range the stop loss carrier underwrote for. A 25-person group’s expected claims fund is sized for variance, but not for outliers. One member with hemophilia generates annual claims that can approach the entire expected claims fund for the group. Stop loss underwriting responds by lasering that member out of specific stop loss protection, transferring the exposure back to the employer. The plan covers the member. The employer absorbs the catastrophic risk.

The fifth assumption is proximity to network providers. Leased PPO networks produce national access on paper and geographic density only in metro areas. A network with 40 participating primary care physicians and 15 participating cardiologists within 15 miles in Phoenix has two or three of each in a rural county two hours from the nearest specialist. Plan design assumes the member can reach a participating provider.

The KFF 2024 Employer Health Benefits Survey puts concrete numbers to the cost structure these assumptions produce. Among covered workers at firms with fewer than 200 employees, the average annual deductible for single coverage was $2,575, compared to $1,538 at larger firms. Thirty-two percent of covered workers at small firms were enrolled in plans with deductibles of $2,000 or more. Workers at small firms contributed an average of $7,947 toward family coverage annually, compared to $5,697 at large firms. The KFF 2025 survey found that 37% of covered workers at firms with 10 to 199 employees were enrolled in level-funded plans.

These numbers describe a plan design that works when the assumed population holds. They describe something else when it does not.

The Actual Population by Industry and Income

#

Level funded adoption concentrates in specific industries, and those industries concentrate specific types of workers. Construction, home health care, landscaping and groundskeeping, food service, janitorial services, professional services, and small healthcare practices are the sectors where level funded penetration is highest. These industries do not share a workforce profile. They share employer size characteristics that make level funded economically attractive: enough employees to form a risk pool, but not enough to self-fund without stop loss protection.

What these industries do not share with each other, or with the assumed population, is wage structure. The Bureau of Labor Statistics Occupational Employment and Wage Statistics survey for May 2024 establishes the income floor clearly. Home health and personal care aides, the largest single occupation in the United States at 4.0 million workers, earned a median annual wage of $34,900. Landscaping and groundskeeping workers, employing nearly 1 million people nationally, had a mean annual wage of $40,880. Food preparation workers averaged $33,380. Construction laborers, the entry-level category in the industry, earned a median that ranges from approximately $35,000 in lower-wage states to $55,000 in the highest-cost markets.

Against these wages, the standard small-employer cost-sharing structure is not merely challenging. For a single worker earning $34,900 annually, a $2,575 deductible represents 7.4% of gross income before any consideration of premium contributions or coinsurance. The Commonwealth Fund’s 2024 Biennial Health Insurance Survey defines underinsurance as a condition in which the deductible alone equals 5% or more of household income. By that threshold, a substantial share of workers in level funded industries who have coverage are underinsured by definition before they file a single claim.

This is not a peripheral observation about edge cases. It is a description of the structural condition of coverage for a large share of the workers these plans actually serve.

The Gap Between Covered and Uncovered

#

Three populations emerge from the intersection of level funded employer adoption and actual workforce characteristics, and the lines between them are neither fixed nor evenly distributed by employer.

The first population consists of workers who are covered and whose characteristics match the plan design assumptions. These are the full-time, moderate-income employees in professional services practices, established healthcare offices, and stable construction companies with experienced workforces. They can absorb the cost sharing. They live and work near network providers. Their health status, while variable, falls within the range the stop loss carrier priced for. Level funded serves this population well. The product was designed for them, and the industry’s growth projections are built on the assumption that this population is expanding.

The second population consists of workers who are covered but whose characteristics diverge from the design assumptions in ways that materially degrade the coverage. The BLS Employee Benefits in the United States survey (March 2024) provides the access and participation data that makes this concrete. Among full-time private industry workers, 87% had access to medical care benefits and the take-up rate was 67%. For establishments with fewer than 50 workers, the take-up rate fell to 60%. The workers who have access and decline it are disproportionately those for whom cost sharing makes coverage unaffordable in practice.

The Commonwealth Fund 2024 Biennial survey found that among adults who are insured, 23% meet the clinical definition of underinsured: enrolled in health plans with cost-sharing requirements so high relative to income that care access is effectively impaired. Among the underinsured, 66% had employer coverage, not individual market coverage. And 57% of underinsured adults reported forgoing needed care due to cost. The people forgoing care despite having coverage are not primarily uninsured individuals who slipped through the system. They are enrolled workers, carrying insurance cards, skipping appointments and medications because the math does not work.

The third population consists of workers employed by level funded employers but not covered at all. BLS March 2024 Employee Benefits data shows that among full-time workers, 89% have access to medical care benefits, but only 26% of part-time workers do. In industries with high part-time employment rates, a substantial fraction of the workforce is ineligible by hours threshold before eligibility is even evaluated. Waiting periods of 30, 60, or 90 days exclude workers whose average tenure is shorter than the waiting period plus enrollment processing time. In restaurants, where the National Restaurant Association reports annual turnover exceeding 70%, a 60-day waiting period structurally excludes a meaningful share of workers who will never accumulate enough continuous employment to become eligible.

Workers who decline coverage because they cannot afford employee premium contributions for themselves, and even more so for their families, are a fourth category the aggregate statistics compress. The KFF 2024 EHBS shows that 26% of covered workers at small firms must contribute more than half of the premium for family coverage. For workers earning $30,000 to $40,000, the family coverage premium contribution this produces often exceeds what the household can sustain.

The Baseline and the Departures

#

The population that level funded serves well is not controversial. It is the implicit baseline from which the rest of this series departs. The employer has 15 to 50 employees, most of whom work full time with stable tenure. Income is sufficient to absorb the deductible without rationing care. The group is located in a metro area with a functional PPO network. No single member has a health condition that would trigger a laser. The broker understands the product and the employer accepts the administrative responsibility of being a plan sponsor.

This baseline describes a specific segment of the small employer market. The Census Bureau’s Statistics of U.S. Businesses shows that 89% of employer firms in the United States have fewer than 20 employees. The KFF 2025 EHBS notes that the average premiums for covered workers at firms with larger shares of older workers are measurably higher than for firms with younger populations, reflecting the actuarial reality that the demographic composition of the small employer universe is not uniform. The professional services firm averaging $90,000 in annual wages per employee and the home health agency averaging $35,000 are both small employers. They are not the same plan design problem.

The pattern of who level funded serves and who it misses is not random. It maps directly to the five design assumptions identified above, and the gap between those assumptions and the actual workforce working for small employers in 2025 is the structural problem the series examines. Each article that follows traces one population through the specific assumptions that fail for them — and what that failure looks like for the employer, the plan, and the worker.