ICHRA, ACA Markets, and Level Funded: Three Models in Search of a Strategy

The three coverage models that dominate the small employer benefits conversation, Individual Coverage Health Reimbursement Arrangements, ACA marketplace plans, and level funded arrangements, are routinely discussed as if they sit on a spectrum from simple to complex, or cheap to expensive, and the employer’s job is to pick their spot on the line. They are not on a spectrum. They are structurally different responses to different problems, with different risk allocations, different information architectures, and different implications for the TPA’s role. Most of the confusion in the market, and most of the bad product strategy decisions at TPAs, comes from treating them as interchangeable options rather than as distinct architectures. The expiration of the ACA’s enhanced premium tax credits on January 1, 2026 has made the structural differences sharper and the strategic stakes higher.

Three Different Architectures, Not Three Price Points

#

Level funded places the employer inside the risk. The employer pays a fixed monthly amount funding three things: expected claims, stop loss premium, and administrative fees. If claims come in below expectations, surplus is returned. If claims spike above the attachment point, stop loss absorbs the excess. The employer has real financial exposure, real data on their population’s health utilization, and a real relationship with a TPA that manages claims, tracks stop loss accumulators, reports on cost drivers, and navigates member service. The TPA’s value proposition is built on claims intelligence, data, and employer engagement.

The ACA marketplace places the individual alone. The employee buys coverage directly from a carrier. Premium is set by age, geography, and plan tier under community rating rules. The employer is absent unless contributing through an ICHRA or similar mechanism. Nobody in the employer relationship has claims data. The carrier has it. The individual has EOBs. Risk is fully transferred to the carrier.

ICHRA places the employer at arm’s length. The employer sets a monthly reimbursement amount. The employee buys a marketplace plan of their choice. The employer reimburses up to the cap, tax-free. The employer has no claims exposure, no stop loss relationship, and no visibility into health utilization. The TPA’s role is reimbursement processing and compliance verification: necessary work, but fundamentally thinner than the claims management, stop loss coordination, employer analytics, and member navigation that constitute the TPA’s relationship in a level funded arrangement.

The structural insight that matters for product strategy: the TPA’s differentiation in level funded is built on capabilities (claims intelligence, stop loss management, employer reporting, member navigation) that are entirely irrelevant in an ICHRA arrangement. A TPA adding ICHRA is not extending its value proposition to a new market. It is offering a different, lower-value service. This is not inherently wrong, but the distinction needs to be understood before it can be managed.

Where Each Model Wins and Where It Fails

#

Level funded wins when the employer has 10 or more relatively stable employees, wants data on their own health utilization and will use it, has a broker capable of managing the complexity, and operates in a market where the ACA individual market is thin or expensive. It wins especially in the current regulatory environment for employers whose workforce includes members above 400 percent of the federal poverty level, a population for whom the marketplace became dramatically more expensive on January 1, 2026.

Level funded fails when the employer’s workforce is highly mobile or seasonal (turnover destroys plan year economics), when the employer has no broker capable of managing the complexity, or when the group is below the size where the economics work without a viable pooling mechanism (FWD.03).

The ACA marketplace wins when employee income distribution makes subsidies meaningful (below 400 percent FPL, the basic premium tax credits still reduce costs materially, though less than the enhanced credits did), when individual market quality is high (metro areas with active carrier competition and meaningful plan choice), and when the employer wants zero administrative burden. The marketplace won more convincingly when the enhanced premium tax credits were in effect. Since their expiration, the marketplace’s value proposition has narrowed.

The marketplace fails for the population this series is most concerned with. Enrollees with income above 400 percent FPL (approximately $63,000 for an individual or $129,000 for a family of four in 2025 poverty level terms) lost all premium tax credit eligibility when the enhanced provisions expired. A 60-year-old couple at 402 percent FPL could face yearly premiums of approximately $22,600 in 2026, roughly a quarter of their annual income, for a benchmark silver plan (Bipartisan Policy Center, “Enhanced Premium Tax Credits”). The Urban Institute projects 4.8 million people becoming uninsured as a result of the expiration, with 7.3 million losing subsidized marketplace coverage overall (Urban Institute, “4.8 Million People Will Lose Coverage in 2026”). KFF estimates that average net premium payments for marketplace enrollees more than doubled, increasing approximately 114 percent (KFF, “ACA Enhanced Premium Tax Credit Calculator”). Individual market premiums for 2026 increased an additional 18 percent on average because insurers priced in the expectation that healthier enrollees would drop coverage, worsening the remaining risk pool (Peterson-KFF Health System Tracker).

ICHRA wins when the employer wants to contribute toward coverage without managing a plan, the employee population is geographically dispersed (each employee buys the best local option), and the reimbursement amount is generous enough to purchase real coverage in the local market. It wins as an on-ramp for employers who have never offered benefits: 83 percent of employers offering ICHRA or QSEHRA in 2025 had not previously offered any coverage (HRA Council, “Growth Trends Vol. 4”).

ICHRA fails when it fails to purchase adequate coverage. A $400 monthly ICHRA reimbursement in a high-cost market, post-PTC-expiration, buys a bronze plan with a deductible exceeding $7,000 for a 55-year-old. The employer has fulfilled their obligation. The employee has nominal coverage. The average benchmark silver plan premium in 2026 is $625 per month; the average lowest-cost bronze plan is $456 (Peterson-KFF Health System Tracker). For above-subsidy earners, the gap between the ICHRA reimbursement amount and the actual premium cost is now wider than it was at any point since ICHRA’s inception. ICHRA also fails when the employee population is not equipped to navigate individual market plan selection, a problem that is worse in markets with limited carrier participation and plan options. And it fails when the employer cares about workforce health as a business input: with zero claims data visibility, the employer cannot identify cost drivers, benchmark utilization, or intervene on population health. ICHRA is a financial arrangement, not a benefits management strategy.

How Brokers Actually Select Between Models

#

The outline treated model selection as an analytical exercise. In practice, brokers mediate almost all small employer coverage decisions, and brokers have their own economics.

Compensation structure matters. Level funded and fully insured products typically pay the broker a percentage of premium or a per-employee-per-month fee. ICHRA compensation is less standardized. Some ICHRA administration platforms pay brokers a PEPM fee; others pay less or require the broker to build the fee into a consulting arrangement. A broker whose ICHRA compensation is lower than their level funded compensation will, consciously or not, recommend level funded more often. This is not corruption. It is the predictable result of incentive structures.

Complexity and service burden drive defaults. Level funded requires the broker to understand stop loss, manage renewals, interpret claims data, and advise on plan design. ICHRA requires the broker to help employees select marketplace plans, a different skill set (individual market knowledge rather than group benefit design). Many brokers default to whichever model they are more comfortable servicing. The path of least explanation is a real force in model selection, particularly in a broker practice where the small employer account is not the highest-revenue relationship.

E&O exposure enters the calculation. A broker who recommends level funded to a 5-person group that has a catastrophic claims year faces different professional liability exposure than a broker who recommends ICHRA. The risk calculus is not identical, and brokers who have been burned on a bad level funded year may shift toward ICHRA for smaller groups regardless of whether that serves the employer.

What this means for the TPA: a TPA that offers both level funded and ICHRA without a clear broker education strategy, with honest criteria for which employer profiles belong in which model, is letting the broker’s economics and comfort level determine the product mix. The broker’s incentives and the employer’s needs are not always aligned. The TPA that trains its broker partners on model selection builds a more durable broker relationship than one that simply offers both products and hopes for the best.

The Regulatory Trajectory

#

The regulatory environment for all three models shifted materially in 2025 and early 2026, and the direction of the shifts is not uniform across models.

ICHRA’s legislative moment arrived and partially stalled. ICHRA adoption grew over 1,000 percent from 2020 to 2025, with large employer adoption (50-plus FTEs) increasing 34 percent from 2024 to 2025 and small employer adoption increasing 52 percent. The HRA Council estimates that over one million Americans now have access to coverage through ICHRA or QSEHRA (HRA Council, “Growth Trends Vol. 4”). The CHOICE Arrangement provisions in the 2025 reconciliation bill would have codified ICHRA into statute, introduced small employer tax credits, and enabled pre-tax cafeteria plan elections, but those provisions were removed from the final version of the One Big Beautiful Bill Act before enactment (Healthcare Dive; Alegeus). ICHRA remains a regulatory construct, not a statutory fixture. Its long-term stability depends on administrative rulemaking that a future administration could revise. HRA Council members are reporting 400 to 800 percent increases in employer requests for quotes for 2026 and 2027, a leading indicator that suggests the PTC expiration is driving new ICHRA demand even as the marketplace it reimburses into has become more expensive (HRA Council via Becker’s).

The marketplace is contracting. Marketplace enrollment peaked at 24.3 million in 2025 under enhanced premium tax credits. The expiration of those credits is projected to reduce subsidized enrollment by 7.3 million and increase the uninsured population by 4.8 million (Urban Institute). A bipartisan Senate group has been negotiating the Consumer Affordability and Responsibility Enhancement (CARE) Act to reestablish enhanced PTCs for two years, but as of March 2026, no legislation has been enacted (ASTHO). Individual market premiums increased approximately 18 percent for 2026 in anticipation of adverse selection as healthier enrollees exit. The marketplace that ICHRA reimburses into is more expensive and less stable than it was twelve months ago.

Level funded regulation is tightening at the state level. At the NAIC’s August 2025 national meeting, regulators heard recommendations for increased regulation of level funded plans, including common contract definitions, clearer disclosure requirements, compensation transparency rules for all level funded service providers, and detailed claims disclosure mandates (InsuranceNewsNet). The NAIC Stop-Loss Insurance Model Act sets minimum attachment points: no lower than $20,000 per individual, and for groups of 50 or fewer, no lower than the greater of $4,000 times the number of members, 120 percent of expected claims, or $20,000. Some states go further: Delaware and New York prohibit stop-loss insurance for small groups entirely, effectively barring self-funding at small group sizes. Others, like North Carolina, regulate stop-loss as if it were health insurance. The regulatory patchwork is growing, and a national TPA serving small employers across states faces increasing compliance complexity. ERISA preemption remains the countervailing force, and the boundary of that preemption continues to be tested in litigation.

The KFF 2025 Employer Health Benefits Survey found that 37 percent of covered workers at firms with 10 to 199 employees were covered by a level funded plan, a share that has held steady over the past two years. Sixty-seven percent of all covered workers are in self-funded plans (KFF, “2025 Employer Health Benefits Survey”). Level funded’s share of the small employer market is stable and substantial. The regulatory question is whether state-level tightening will erode the flexibility and cost advantages that sustain that share.

The Model Confusion Error

#

The analytical payoff of the preceding sections is a specific strategic error that can now be named precisely. The error is treating ICHRA as a complement to level funded when it is in some cases a substitute, and building toward both without a clear theory of which employers belong in which model, which brokers should recommend which product under what circumstances, and why.

The error manifests predictably. The TPA adds ICHRA administration because the market is growing (it is) without analyzing whether ICHRA growth is coming at the expense of its own level funded pipeline. The broker channel receives two products from the same TPA with no clear selection criteria. The broker defaults to whichever is easier to sell, pays better, or generates less E&O exposure. The result is a product portfolio that competes with itself, confuses brokers, and serves no segment with distinction.

The sharper position: for a TPA with genuine claims management capability, the value proposition of level funded is real and defensible. The TPA’s differentiation, claims intelligence, stop loss management, employer analytics, member navigation, exists in level funded. It does not exist in ICHRA. Adding ICHRA because the market is growing is rational. Building the business around ICHRA because level funded is hard is a strategic retreat that is not always recognized as one. The PTC expiration sharpens this: ICHRA’s value depends entirely on the quality and affordability of the marketplace it reimburses into, and that marketplace just became significantly more expensive for the population most relevant to TPA strategy.

The honest question: is the TPA adding ICHRA to serve employers who genuinely belong there (small employers who have never offered benefits and for whom ICHRA is an on-ramp), or to avoid the harder work of making level funded viable at smaller group sizes (FWD.03)? Both answers are legitimate. Only one is honest about what is happening. FWD.05 frames the full set of strategic choices and their tradeoffs.