The Architecture's Blind Spots: What a Genuinely Inclusive Small Employer Benefit System Would Require

Fourteen populations. Eight that the architecture was never designed for. Six that the architecture nominally covers but systematically underserves. The pattern across both categories is not random. The structural mismatches share one origin. The inside-the-architecture failures share a different one. And the combination tells the employer something specific about what the benefit system they are funding actually does and does not do.

The Pattern in the Structural Mismatches

#

The eight structural mismatch populations share one characteristic: they exist at the boundary of the employment relationship. The caregiver is in the employment relationship but has restructured it around an obligation the relationship was not designed to accommodate. The disabled adult at 26 is at the age boundary Congress drew for dependent coverage eligibility. The 62-to-64 worker is at the boundary between working-age insurance and Medicare. The multi-1099 worker is at the boundary between employment and self-employment. The veteran is at the boundary between military and civilian coverage systems. The agricultural worker is at the boundary of seasonal employment and the enrollment calendar. The S-corp spouse is at the boundary of owner and employee. The rural independent is at the boundary of market viability.

The architecture defines coverage through employment relationships. Every population that exists at the boundary of an employment relationship finds a gap. The gap is not an accident. It is the boundary of the design.

What all eight gaps share as an opportunity is a set of components that already exist but have not been assembled for these populations. Direct-pay primary care (DPC membership and FQHC access) provides the primary care relationship without requiring group enrollment, an insurance card, or an employment relationship. It begins when someone pays the monthly fee. Contribution-based employer funding (ICHRA and QSEHRA) provides the tax-advantaged mechanism for an employer to fund individual market coverage without maintaining a group plan, with class structures that can accommodate part-time workers, seasonal employees, and variable work arrangements. Catastrophic-only insurance at high attachment points provides the tail-risk protection that no DPC membership or primary care arrangement covers. HSA-qualified HDHPs with aggressive contribution strategies provide tax-advantaged savings for medical expenses during gap periods and transitions. ABLE accounts provide disability-specific savings that do not jeopardize public benefit eligibility.

The components exist. The assembly has not happened because the distribution channel (the broker advising the small employer) is not designed to reach populations outside the employment relationship. The broker sells group coverage to employers. The populations in the first half of this series do not fit the group coverage model. They need something built from the same components but assembled differently, and nobody in the current distribution chain has an economic incentive to build it for them. The opportunity sits in the gap between the components that exist and the product that does not.

The Pattern in the Inside-the-Architecture Failures

#

The six inside-the-architecture populations share a different characteristic: they are inside the employment relationship, nominally covered, legally eligible, and systematically underserved because the default plan design does not account for their specific needs. The failures are not legal. They are informational and operational.

The LGBTQ+ employee does not need a different kind of coverage. They need the plan to cover PrEP at zero cost-sharing as an affirmative design choice rather than a regulatory compliance artifact, to cover gender-affirming care following WPATH standards with prior authorization rather than exclusion, and to provide access to behaviorally competent providers through telehealth network supplements. These are plan document decisions.

The chronically comorbid employee does not need a different kind of coverage. They need cost-sharing that does not prevent them from using the coverage for the conditions that will generate catastrophic costs if unmanaged. Zero-deductible carve-outs for maintenance medications, DPC for the coordination relationship, and structured disease management programs are plan design choices with identifiable costs and measurable returns.

The ASD family does not need a different kind of coverage. They need ABA therapy at coverage levels that allow the therapy to happen. A $50,000 annual maximum for ABA therapy, with stop-loss inclusion verified before the benefit is added, is a plan document change with a quantifiable retention value that exceeds its cost in most workforce compositions.

The union-adjacent worker does not need a different kind of plan. They need the employer to acknowledge the coverage gap relative to what the union hall offers and to design toward closing it through DPC, zero-cost maintenance medications, and dental carve-outs that address the three dimensions the worker actually notices.

The transgender employee does not need a different architecture. They need an employer who engages ERISA counsel, makes the plan design decisions that legal coverage requires, verifies stop-loss inclusion, and communicates the coverage clearly, including the legal constraints that affect access in states with legislative hostility.

The returning citizen does not need charity. They need day-one DPC membership, a waiting period that does not create a 90-day gap at the most vulnerable moment of reintegration, and behavioral health access through telehealth MAT platforms that bridge the coverage void. The WOTC credit the employer receives for making the hire exceeds the cost of every intervention this series identifies.

In every case, the lever exists. The employer controls plan design in a self-funded arrangement in ways the fully insured employer does not. The broker who designed the standard level funded plan did not surface the lever because the broker was not thinking about these populations. This series surfaces the levers.

What a Genuinely Inclusive Small Employer Benefit System Would Require

#

It would not look like the current architecture extended to cover more populations. It would look like a different arrangement built from the same components.

The primary care layer would be DPC membership for every covered individual and, where the Section 152 dependency test is met or where the employer chooses to fund it directly, for designated care recipients outside the employment relationship. DPC provides the 30- to 60-minute appointments, proactive medication management, and care coordination that the claims-based primary care model cannot deliver at scale for complex populations. At $75 to $150 per member per month, DPC is the highest-value, lowest-complexity component available to the small employer. Starting in 2026, HSA funds can pay for DPC membership, creating a tax-efficient bridge between DPC and HDHP coverage.

The catastrophic protection layer would be stop-loss insurance or an HDHP marketplace plan, depending on the population. For employees in a group arrangement, level funded coverage with specific and aggregate stop loss provides the tail-risk protection. For individuals outside the group (the caregiver’s parent, the 62-year-old independent, the multi-1099 professional), a marketplace HDHP or catastrophic plan provides the protection at an individual level. The catastrophic layer is the only component that is genuinely insurance in the risk-transfer sense.

The pharmacy layer would separate from the insurance product. Transparent PBM arrangements, manufacturer assistance programs, discount card access (GoodRx and comparable platforms), 340B program access where eligible, and international pharmacy purchasing where legal provide medication access at costs below what the insurance formulary produces. Zero-deductible carve-outs for chronic disease maintenance medications would be standard plan design, not an innovation.

The contribution mechanism would be ICHRA or QSEHRA for populations that do not fit the group plan model, with class structures designed around the actual workforce composition (part-time workers, seasonal employees, variable-hour workers, multi-location workers) rather than the conventional full-time-employee assumption. Employer contributions would flow to the individual rather than to the group, with the individual choosing the coverage components that fit their situation.

The plan design layer would reflect the populations the employer actually employs. Explicit PrEP coverage. ABA therapy coverage with defined annual maximums. Gender-affirming care following WPATH standards. Zero-cost maintenance medications for chronic conditions. Behavioral health telehealth supplements. These are not aspirational additions. They are plan document decisions with identifiable per-member costs that the employer can evaluate against the retention value and the cost trajectory each decision affects.

The regulatory changes that would help are specific and bounded. ABLE account access for populations beyond the current disability-onset definition (partially addressed by the 2026 expansion to age 46). ICHRA class structures that allow contribution variation by caregiving status. Clarification of ERISA preemption for gender-affirming care in hostile-state environments. Medicaid suspension rather than termination upon incarceration to enable immediate reactivation upon release. HSA eligibility for DPC membership (addressed in 2026). None of these requires inventing a new coverage mechanism. They require applying existing mechanisms to populations that were not in the original design conversation.

The Employer’s Honest Accounting

#

The employer who reads this series and makes none of the plan design changes it identifies has learned that the gaps exist and chosen not to address them. That is a choice. The employer who reads it and makes specific changes (eliminates the ABA therapy exclusion, adds DPC membership from day one for returning citizens, covers PrEP at zero cost-sharing, reduces the waiting period, verifies stop-loss inclusion for gender-affirming care) has made a different choice. The difference is not values. Both employers may value their workforce. The difference is whether the values produce specific operational decisions or remain general intentions.

The third employer objective from TOS.PRE is keep it honest. Tell the employee what the company can do for them, what it cannot do, and what it asks in return. The populations in this series have been told implicitly what the company cannot do: cover their parent’s care coordination, bridge the 26-year-old coverage cliff, provide ABA therapy at a meaningful level, offer gender-affirming care in a legally complex environment. This series identifies how many of those answers are actually “have not done” rather than “cannot do,” and what it would take to change them.

The gaps are real. The populations are identifiable. The levers exist. The costs are knowable. What has been missing is someone who names the gap, identifies the lever, and tells the employer: this is yours to decide.