Coverage as Retention: The Case for Variable Employer Contribution

The prevailing norm in employer-sponsored health benefits holds that coverage should be uniform across the workforce. The same plan, offered on the same terms, available to all eligible employees. Non-discrimination rules, ACA provisions, and industry convention all reinforce this posture. Varying the employer’s health benefit contribution based on an employee’s value, tenure, role, or retention priority is treated as legally suspect, ethically questionable, and operationally complicated.

This article argues that the uniformity norm serves the insurance product architecture, not the employer or the employee. Every other component of employee compensation, salary, bonus, equity, paid leave, parking benefits, and professional development budgets, varies by employee value. Health coverage is the one exception where the benefits industry insists on uniformity, and that insistence rests on a legal framework that is far narrower than commonly understood and on a cultural norm that the employer-as-plan-sponsor model has never been required to maintain.

The Legal Framework: What It Actually Prohibits

#

Three overlapping legal frameworks govern employer health benefit discrimination. Getting them right is the precondition to the rest of this analysis.

Internal Revenue Code Section 105(h) governs self-insured health plans. It prohibits discrimination in favor of highly compensated individuals in both plan eligibility and benefits. For 105(h) purposes, a highly compensated individual is defined as one of the five highest-paid officers, a shareholder owning more than 10 percent of the employer, or a member of the highest-paid 25 percent of all employees. The rule is written to prevent employers from running health benefits that disproportionately favor executives over rank-and-file workers. It does not prohibit differentiating benefit levels between non-HCI employee classes in general. An employer who offers richer benefits to experienced project managers than to entry-level staff has not necessarily violated 105(h) unless the class receiving richer benefits disproportionately maps onto the highest-paid quartile.

Internal Revenue Code Section 125 governs cafeteria plans, the pre-tax premium payment vehicle that most employers use to allow employees to pay their share of health premiums with pre-tax dollars. Section 125 has its own non-discrimination tests: an eligibility test (the plan may not discriminate in eligibility in favor of highly compensated participants), a contributions and benefits test (key employees may not receive a disproportionate share of benefits), and a key employee concentration test (key employees may not receive more than 25 percent of all nontaxable benefits). Like 105(h), the Section 125 framework is oriented toward preventing disproportionate tax benefit accrual to executives and owners, not toward requiring identical contributions across all employee classifications.

ACA Section 2716 would have extended nondiscrimination rules similar to 105(h) to non-grandfathered fully insured group health plans. This provision became effective, on paper, for plan years beginning on or after September 23, 2010. In practice, the IRS issued Notice 2011-1 in December 2010 acknowledging that regulatory guidance was essential to implementation and that compliance would not be required until regulations were issued. No implementing regulations have been issued as of 2026. No sanctions have been imposed. The provision exists in statute. It has never been enforced. Fully insured employers operating in 2026 face no active legal obligation under Section 2716.

The practical upshot of this legal framework is that the non-discrimination constraints on employer health benefit variation are substantially narrower than the industry generally treats them. The operative prohibitions target employer enrichment of executives and owners at the expense of ordinary workers. They do not prohibit an employer from offering richer health benefits to its senior project managers, lead technicians, or experienced department heads than to its entry-level staff, provided the plan structure passes 105(h) and 125 tests, which means the favored class cannot be predominantly composed of HCIs or key employees.

The ICHRA Class Structure as Proof of Concept

#

Individual Coverage Health Reimbursement Arrangements, finalized in June 2019 and available beginning January 1, 2020 under the final rules codified at 26 CFR 54.9802-4, already implement class-based variable contribution as a standard product feature. The ICHRA regulation explicitly permits employers to divide their workforce into distinct classes and offer different reimbursement amounts to each class. The permissible classes include full-time employees, part-time employees, seasonal employees, salaried employees, non-salaried employees, employees whose primary work site falls in the same geographic rating area, and employees covered under a collective bargaining agreement, along with combinations of these categories.

Each class must receive its benefit on the same terms, meaning uniform treatment within a class. But the contribution level can differ across classes. A 25-person construction firm can offer its salaried project managers an ICHRA contribution of $1,200 per month and its hourly laborers a contribution of $500 per month. The differential is lawful, explicit in the regulation, and documented in the eCFR. The regulatory structure that the industry treats as prohibiting variable contribution has built class-based variable contribution into its most recent major product innovation.

The ICHRA minimum class size rules add a constraint worth understanding precisely. Minimum class sizes apply when the employer offers both a traditional group health plan to one class and an ICHRA to another class. For employers with fewer than 100 employees, the minimum class size is 10 employees. For employers with 100 to 200 employees, it is 10 percent of total employees. The minimum class size does not apply if the employer offers only an ICHRA to all eligible employees, with no group plan offered to any class. For a small employer designing a pure ICHRA-based benefit, the class size constraint disappears entirely.

Variable Contribution as Compensation Strategy

#

The structural argument for variable health benefit contribution does not rest on regulatory permissiveness. It rests on the same logic that governs every other component of compensation.

A 30-person software development firm pays its senior engineers $180,000 and its junior developers $90,000. That differential reflects scarcity, skill, and retention priority. The same firm spends $15,000 annually on conference attendance and professional development for its senior engineers and $3,000 for junior staff. No one characterizes that differential as discriminatory. The firm’s equity pool is allocated almost entirely to the people it most needs to retain. The differential in total compensation is the mechanism by which the firm communicates investment priority.

That same firm’s health benefits, by industry convention, are offered uniformly: the same plan, the same premium split, the same deductibles. The senior engineer whose total compensation package includes $8,951 in single-coverage equivalent health premium (the 2024 KFF average) receives the same benefit as the junior developer. The employer treats the most retention-critical employees identically to the most replaceable ones in the one area that many workers, particularly those with families or chronic conditions, value most highly.

Variable contribution corrects this misalignment. An employer who offers its senior project managers full employer-paid premium equivalent and its entry-level staff a 70 percent employer contribution has communicated something honest about the employment relationship: the company invests more in the people it most needs to keep. This is not discrimination in the legally prohibited sense. It is transparent compensation strategy. The average annual employer contribution to single-coverage premiums reached approximately $7,583 in 2024, according to KFF, representing a material component of total compensation that most employers currently deploy as if it were a fixed operating expense rather than a variable retention instrument.

The industry norm of uniform contribution exists partly because bundled insurance products are administratively easier to apply uniformly, partly because brokers who manage the renewal relationship for the whole book find uniform plans simpler to service, and partly because the cultural conflation of uniformity with fairness runs deep in the benefits profession. None of those rationales serves the employer or the employee. They serve the product and the intermediary.

What the Employer Actually Wants to Say

#

The preface to this collection establishes the employer’s third objective: make it simple and honest. Tell the employee what the company can do for them. Tell them what it cannot. Tell them what it asks in return.

Variable contribution makes the benefit honest in a way that uniform contribution does not. When a company invests $14,000 annually in a senior employee’s family health coverage and $5,500 in a junior employee’s single coverage, both amounts are explicit components of total compensation. The employee can evaluate the offer, compare it to alternatives, and understand what the company values. When both employees receive the same nominal health plan with the same employer contribution, the senior employee does not see the retention investment that she might weigh against a competitor’s offer. The benefit is invisible as a differentiation tool precisely because it is uniform.

The honesty argument extends to the employee compact the preface describes. If the employer’s health investment is proportional to the employee’s value to the company, the implicit message is clear: the company invests more where it expects more in return, and more experienced employees who use the healthcare system responsibly and preventively protect both their own health and the plan’s financial stability. That compact is harder to articulate when the benefit is identical regardless of the employment relationship.

The Reciprocity Dimension

#

The TOS collection’s organizing framework includes a reciprocity dimension: the employer invests in the employee’s health access, and the employee engages with their own health. Variable contribution can make this reciprocity explicit and graduated. An entry-level employee who joins the firm receives a baseline benefit. As that employee’s tenure increases, their contribution level rises, their cost-sharing decreases, and their benefit structure improves. The escalation is not a secret: the employer communicates it during onboarding. The employee understands that building a track record at this company has tangible benefit consequences.

This structure already exists in a different form. Most employers impose waiting periods before health coverage activates, typically 30 to 90 days. Many employers match 401(k) contributions at rates that increase with tenure, with vesting schedules that make long-term employment financially significant. Variable health contribution is the same principle applied to the most tangible health benefit: the employer’s investment in access increases with the employer’s investment in the relationship.

The objection that variable contribution introduces administrative complexity is real but overstated. ICHRA administration platforms already manage class-based contribution differentials as routine functionality. The TPA operational stack for a level funded plan can be configured with different employer contribution percentages by employment class with no more complexity than a standard tiered benefit structure. The complication is not administrative. It is cultural: the benefits industry has built its consulting, compliance, and product infrastructure around uniformity, and variable contribution requires advisors who can articulate the legal analysis rather than retreat to the safer position of recommending the uniform default.

The Fiduciary Accountability Question

#

The employer-as-plan-sponsor fiduciary obligation under ERISA adds a dimension that the variable contribution argument must address honestly. An employer who administers a self-funded health plan has fiduciary responsibilities to plan participants and beneficiaries. The duty of loyalty requires the employer to act in the interest of participants, not solely in the interest of the business. Variable contribution could be read as using the plan to serve the business’s retention objectives rather than the participants’ health interests.

This reading misapplies the fiduciary standard to the contribution decision. ERISA’s fiduciary obligations apply to the administration of the plan once established: claim adjudication, investment of plan assets, selection of service providers, and operational decisions. The employer’s decision about what benefit to offer, including the contribution level, is a settlor function, not a fiduciary function. The DOL has long held that the decision to establish, amend, or terminate a benefit plan is a settlor decision not subject to ERISA’s fiduciary standards (Curtiss-Wright Corp. v. Schoonejongen, 514 U.S. 73, 1995). An employer who sets different contribution levels for different employee classes is exercising a settlor function, not a fiduciary one.

The fiduciary obligation becomes relevant if the employer uses the plan administration process, claim adjudication decisions, or appeals handling in ways that disadvantage participants based on their employment class or retention value. An employer who routes its senior employees’ claims through a preferred adjudication track while letting junior employees’ claims languish has a fiduciary problem. An employer who sets different employer contribution levels at plan design time, documented in the plan document, communicated transparently to employees, and applied uniformly within each class, does not.

Where the Legal Exposure Actually Lives

#

This article is not arguing that any differential contribution structure is automatically lawful. The legal constraints that do apply are real and must be handled correctly.

A self-insured plan that offers substantially richer benefits to the employer’s five highest-paid officers than to rank-and-file workers will fail the IRC 105(h) benefits test. A cafeteria plan that allows key employees to divert a disproportionate share of pre-tax contributions to health premiums while lower-paid workers are functionally excluded will fail the Section 125 key employee concentration test. An employer who explicitly ties health benefit eligibility or richness to whether an employee filed a workers’ compensation claim, exercised FMLA rights, or holds a protected characteristic has violated other laws entirely: ADA, FMLA, Title VII.

The space between these constraints and full uniformity is large. An employer who offers its salaried employees a richer level funded plan and its hourly employees an ICHRA, with minimum class sizes satisfied, is within a structure that the 2019 final ICHRA regulations explicitly contemplate. An employer who offers its senior field technicians a direct primary care membership as an additional benefit beyond the core plan has not created a discriminatory arrangement unless those technicians map overwhelmingly onto the HCI definition. An employer who increases the employer premium contribution percentage with employee tenure, applied uniformly within job classification tiers, has built a retention tool that no legal framework currently prohibits.

The industry’s reflexive caution about any variable contribution structure reflects the compliance profession’s preference for safe harbors over analyzed positions. That preference is rational for the compliance professional. It is not necessarily rational for the employer who wants to use health benefits as a genuine retention and compensation tool rather than a commodity offer that differentiates nothing.