Executive Summary: Business Choices for TPAs at the Inflection Point
FWD.05 — The Changing Market#
The level funded market is substantial and growing. The KFF 2025 Employer Health Benefits Survey found 37 percent of covered workers at firms with 10 to 199 employees in level funded plans and 67 percent of all covered workers in self-funded plans of some kind. Fully insured medical market enrollment has dropped nearly 17 percent since 2019. The stop loss market reached $39 billion in premium in 2024, up from $31.6 billion in 2022. The direction is toward employer-borne risk with stop loss protection. Loss economics within that growth are deteriorating: loss ratios worsened from 81.6 percent in 2019 to 86.0 percent in 2024, and million-dollar-plus claims are rising steeply.
The competitive pressure that is not coming from other TPAs: large carriers have entered the small group self-funded space. Nationwide’s $1.25 billion acquisition of Allstate’s stop loss segment in January 2025 and Prudential’s 2024 market entry signal intent. Insurtechs are building from scratch. Angle Health raised $134 million in December 2025, reports 26-fold revenue growth since 2022, serves over 3,000 employers across 44 states, and achieves an 80 percent renewal rate with a 36 percent reduction in median rate increases compared to the broader small group market. HR platforms (Rippling, Gusto, Justworks) already have the micro-employer relationship and are building or acquiring the benefits depth they lack.
Five strategic alternatives. Each with what it requires, what it assumes, and where it breaks down.
Choice A is deepening the core: level funded administration for 10 to 499 employees with better technology, stronger broker relationships, and deeper stop loss carrier partnerships. It requires capital for the Tier 1 AI deployments in FWD.07 and an expandable broker network. It breaks down when an insurtech builds comparable capabilities with a purpose-built technology stack, or when broker channel consolidation erodes relationship-based distribution.
Choice B is extending into ICHRA alongside level funded with a clear theory of which employer profiles belong in which model. It requires new compliance infrastructure and honest broker education on model selection criteria. It breaks down when ICHRA administration commoditizes, which Take Command, Remodel Health, and comparable platforms are driving with over $100 million in combined investor funding in 2025 alone, or when running two models simultaneously creates service quality problems in the level funded core.
Choice C is doubling down on micro-employers: investing in the infrastructure to serve the 1 to 10 life segment profitably at scale rather than absorbing it as a relationship cost. It requires the FWD.07 Tier 1 AI deployments that change unit economics (quoting cost from $120 to $480 per group to near-zero), a viable pooling mechanism with pool-level reinsurance per FWD.03, and regulatory navigation for association or captive structures. It breaks down when adverse selection makes the product unprofitable even with pooling and technology, or when Rippling or Gusto enters with lower administrative costs because their infrastructure was built for this scale.
Choice D is investing in the fractional worker market: participating in association-based pooled coverage for incorporated independent professionals using existing legal mechanisms. The strategic bet is that the fractional workforce’s doubling rate and the ACA marketplace’s deterioration for above-subsidy earners create sufficient demand. It breaks down when portable benefits legislation stalls and the pool cannot achieve sufficient scale without regulatory tailwind.
Choice E is platform evolution: transforming from claims administrator to data platform providing employer analytics, member navigation, and broker intelligence. It requires significant technology investment across FWD.06’s component architecture and Tier 2 AI capabilities from FWD.07. It breaks down when the investment exceeds available capital before the platform generates revenue that justifies continuation.
The operational test across all five choices: can you name which choice you are making, what it requires that you do not currently have, and what breaks the thesis? If not, the choice has not been made. The market will reward deliberate choice and punish drift.