Federal PBM Rebate Mandate Rewrites Employer Pharmacy Trend Models

From building employer group health renewals over the past several years, the pharmacy cost component has always been the line item where the actuary has the least visibility into what the plan sponsor is actually paying. PBMs negotiated manufacturer rebates on one side and credited a fraction to the plan on the other, pocketing the spread. Actuaries projected a "net drug cost" trend that bundled this invisible margin into the trend rate, and nobody outside the PBM knew the true magnitude. That era ended on February 3, 2026, when the Consolidated Appropriations Act of 2026 was signed into law. The legislation mandates 100% pass-through of all manufacturer rebates, fees, alternative discounts, and utilization-based remuneration to self-insured ERISA plan sponsors (Mintz, February 2026). PBM compensation must shift to bona fide service fees not calculated on drug price or utilization.

For pricing actuaries, the consequence is structural: the net pharmacy trend factor that anchored renewal rating models for a generation no longer represents the cost structure being priced. This article walks through the four-component decomposition framework that replaces it, uses West Virginia's early experience data as a calibration benchmark, and addresses the credibility challenge of transitioning from a net-trend historical basis to a gross-plus-credit projection.

100%

Manufacturer Rebates Now Required to Pass Through to Plans

40%

Rate Increase Reduction in West Virginia Under State-Level Mandate

$10K/day

Penalty for PBM Nondisclosure Under the CAA 2026

What the CAA 2026 Requires

The PBM reform provisions in the Consolidated Appropriations Act of 2026 impose three structural changes that directly affect actuarial pharmacy trend modeling. First, PBMs must remit 100% of all manufacturer rebates, fees, and utilization-based remuneration to self-insured plan sponsors on a quarterly basis, no later than 90 days after each quarter-end. Upstream, PBMs must also structure contracts requiring rebate aggregators and group purchasing organizations to pass through 100% within 45 days of each quarter (Mintz, February 2026). Second, PBM compensation must consist of bona fide service fees that are not calculated as a function of drug price, volume, or utilization. This eliminates the spread-pricing model where PBMs earned margin on the gap between gross manufacturer payments and net plan credits. Third, the DOL's companion proposed rule on PBM fee disclosure, published January 30, 2026, requires semiannual drug-level reporting of gross spend, net spend, rebate amounts, and spread pricing for self-insured plans with 100 or more participants (Federal Register, January 2026).

Penalties of $10,000 per day for nondisclosure and $100,000 for knowingly false information create strong compliance incentives. The NFP analysis notes that most provisions take effect for plan years beginning on or after January 1, 2029, for calendar-year plans, though the DOL proposed rule targets applicability for plan years beginning on or after July 1, 2026 (NFP, 2026). The PCMA confirmed on April 9, 2026, that the market and federal law have now made 100% rebate pass-through the industry baseline, with PBM CEOs testifying that 98% to 100% of negotiated rebates already flow to employers in recent contracts (PCMA, April 2026).

Why Net Pharmacy Trend No Longer Works

Under the prior framework, actuaries projected pharmacy costs using a net drug cost PMPM that reflected ingredient cost minus whatever rebate credit the PBM chose to share. The rebate retention percentage varied by PBM, by contract, and by therapeutic class. A PBM might pass through 85% of rebates on a traditional formulary but only 60% on a specialty tier. The actuary had no line of sight into the gross-to-net waterfall. The "net trend" factor implicitly bundled three distinct economic forces into a single number: actual ingredient cost escalation, shifts in utilization and therapeutic mix, and the PBM's margin on retained rebates.

When the PBM retains less, the net trend looks lower. When the PBM retains more, the net trend looks higher. Neither movement reflects an actual change in the cost of drugs. From tracking group health renewals through multiple PBM contract transitions, we have seen net pharmacy trend swings of 3 to 6 percentage points driven entirely by changes in rebate pass-through percentages, not by any change in drug costs or utilization. The CAA 2026 eliminates this noise by requiring the full rebate to flow through. But it also means that historical net trend data collected under variable pass-through regimes is no longer a reliable basis for forward projection.

The Four-Component Decomposition Framework

Under the new regime, the actuarial pharmacy trend model must explicitly separate four components. This decomposition replaces the single "net pharmacy trend" factor with a build-up approach that makes each cost driver visible and independently projectable.

Component 1: Gross ingredient cost trend. This is the unit cost escalation for drugs measured at a standardized benchmark, either Average Wholesale Price (AWP) or the National Average Drug Acquisition Cost (NADAC). AWP is the traditional commercial benchmark, though it is an inflated list price roughly analogous to MSRP. NADAC, published weekly by CMS based on actual pharmacy invoice data, runs approximately 5% below WAC and 20% below AWP for brand-name drugs and 45% to 50% below AWP for generics (JMCP, 2025). The choice of benchmark matters for trend measurement: AWP-based trends capture manufacturer list price increases, while NADAC-based trends capture actual acquisition cost movement. For employer plan pricing, AWP remains the dominant contractual benchmark, but NADAC provides a useful cross-check. Gross ingredient cost trend for brand-name drugs has been running 5% to 8% annually, while generic ingredient costs have been flat to declining.

Component 2: Utilization and therapeutic mix shift. This component captures changes in prescription volume, days supply, therapeutic substitution, and the mix between generic, preferred brand, non-preferred brand, and specialty tiers. The dominant utilization driver in 2026 is GLP-1 receptor agonist adoption. Prescribing volume increased 15% from December 2025 to March 2026, the largest quarter-over-quarter increase since 2019, with first-time semaglutide prescriptions surging more than 50% following the oral formulation approval in December 2025 (medRxiv, March 2026). An estimated 12.4% of U.S. adults were taking a GLP-1 for weight loss by mid-2025, up from 5.8% in February 2024, and projections show 25 million Americans on GLP-1 agents by 2030 (JPMorgan, 2026). Biosimilar substitution rates, specialty drug pipeline launches, and utilization management program effectiveness also load into this component. For a typical employer group, utilization and mix shift adds 2% to 5% to annual pharmacy trend.

Component 3: Manufacturer rebate credit. Under the CAA 2026, this is now a fully transparent offset modeled as a percentage reduction to gross cost. The rebate credit has its own trend rate reflecting formulary negotiation dynamics, manufacturer pricing strategy, and the competitive landscape within therapeutic classes. Historically, aggregate rebate levels for commercial plans have run between 25% and 35% of gross brand-name drug cost, though specialty rebates on certain classes (oncology, autoimmune) can exceed 50%. The rebate credit trend is typically negative (i.e., the credit grows over time as PBMs negotiate larger rebates on maturing brands), running at roughly negative 1% to negative 3% annually for a stable formulary. GLP-1 rebates are a notable exception: early-stage adoption with limited competition has kept rebate levels for semaglutide and tirzepatide below 10%, well under the 30% average for established therapeutic classes.

Component 4: PBM administrative fees. This replaces the implicit spread that PBMs previously earned. Under the new regime, PBM compensation must be structured as a flat PMPM or per-claim fee not tied to drug price or utilization. Early contract data from plan sponsors renegotiating under CAA 2026 terms suggests administrative fees in the range of $3.00 to $8.00 PMPM, depending on plan size and service scope (WTW, February 2026). This component has its own trend, typically 2% to 4% annually, reflecting wage inflation and technology investment in PBM operations. For the actuarial model, this is now a separate, transparent line item rather than an invisible component of the net drug cost trend.

The Gross Trend Optical Shift

When actuaries decompose the historical net pharmacy trend into its gross components, the headline number changes dramatically. A plan that reported 8% net pharmacy trend under the old framework might show 12% to 14% gross ingredient cost trend before the rebate credit offset. This is not a cost increase; it is the same cost viewed without the rebate netting. The rebate credit, now modeled separately, reduces the gross cost by 25% to 35% for a typical commercial formulary. The net effect on total plan cost should be lower under the new regime because there is no PBM retention eating into the rebate.

This optical shift creates a communication challenge. Plan sponsors and CFOs who have been accustomed to seeing an 8% pharmacy trend number will see a 13% gross trend on the first line of the new build-up, and it takes careful framing to explain that the higher gross number minus the now-visible rebate credit produces a lower total cost than the old net number minus the hidden PBM spread. Patterns we have seen in early adopter plan renewals: the initial presentation meeting requires a reconciliation bridge showing the old single-line net trend alongside the new four-component build-up to demonstrate arithmetic equivalence.

West Virginia as a Calibration Benchmark

West Virginia implemented an analogous state-level PBM rebate pass-through mandate in 2022, making it the first jurisdiction with multi-year experience data under a full pass-through regime. The results reported by the West Virginia Offices of the Insurance Commissioner provide the most credible calibration data available for actuaries modeling the federal transition. For 2026 coverage, average group health plan rate increases in West Virginia fell from 19.5% to 12.6%, a reduction of 6.9 percentage points or approximately 40% (BenefitsPRO, February 2026). For 2025 coverage, small group health plans saw rate increases cut by 52%. One large insurer cut rates by 2.91% instead of the 5.09% increase originally filed, a swing of nearly 8 percentage points. Other plans reported reductions ranging from 13% to 58%.

These results are directionally consistent with the economic logic: when PBMs must pass through 100% of rebates, the plan cost falls by the amount of the previously retained spread. The magnitude of the reduction depends on how much spread the PBM was retaining, which varied by contract. The West Virginia data suggests that the effective PBM spread in that state's group market was on the order of 4 to 7 percentage points of total rate increase. For national employer plans, the spread may differ because of differences in PBM contract structure, formulary composition, and rebate levels. Nevertheless, the West Virginia experience provides an empirical anchor for calibrating the transition effect.

Credibility Weighting the Transition: A Buhlmann Framework

The transition from net-trend history to a gross-plus-credit framework creates a credibility problem that sits at the center of the actuarial modeling challenge. Historical net pharmacy trend data, accumulated over years of plan experience, no longer represents the cost structure being priced. Gross-cost data under the new regime is limited to at most a few quarters for early adopters and zero quarters for plans that have not yet transitioned. The actuary must blend thin plan-specific gross experience with industry-wide gross benchmarks, and Buhlmann credibility provides the optimal framework for doing so.

The Buhlmann credibility-weighted trend estimate takes the form:

T_blended = Z * T_plan + (1 - Z) * T_industry

where T_plan is the plan-specific gross pharmacy trend estimated from available post-transition data, T_industry is the industry-wide gross pharmacy trend benchmark (sourced from PBM transparency reports, Milliman Health Cost Guidelines, or Segal/Mercer trend surveys), and Z is the Buhlmann credibility factor. The credibility factor Z = n / (n + k), where n is the number of exposure periods of gross-cost data available for the specific plan and k is the ratio of the expected process variance to the variance of the hypothetical means across the population of plans.

For a plan with only one or two quarters of post-transition gross data, Z will be low, perhaps 0.10 to 0.20, meaning the blended trend leans heavily on the industry benchmark. As plans accumulate four to eight quarters of gross-cost experience, Z rises toward 0.40 to 0.60, allowing the plan-specific experience to carry more weight. The parameter k, which governs how quickly credibility builds, can be estimated from the cross-sectional dispersion of pharmacy trends across a book of employer groups. From recent renewal data, k values in the range of 6 to 12 quarters produce reasonable credibility gradients for mid-size employer groups (500 to 2,000 employees).

Critically, the industry benchmark T_industry must itself be constructed on a gross-cost basis. Using the old net-trend industry benchmarks would contaminate the blended estimate with the PBM spread that the CAA 2026 eliminates. Early-adopter states like West Virginia, along with PBM transparency reports now required under the DOL proposed rule, provide the raw material for building a defensible gross-cost industry benchmark. The Milliman Health Cost Guidelines have already begun publishing gross-and-net trend decompositions, giving actuaries a starting framework.

DOL Disclosure Rule: Unprecedented Data Granularity

The DOL's proposed PBM fee disclosure rule, if finalized as drafted, will give actuaries drug-level data they have never had access to before. The rule requires covered service providers (PBMs and their affiliates) to provide both initial and semiannual disclosures to plan fiduciaries of self-insured group health plans with 100 or more participants (DOL Fact Sheet, 2026). The disclosures must include gross drug-level spend, net spend after rebates, rebate amounts by therapeutic class, and any spread pricing (the difference between what the PBM charges the plan and what it pays the pharmacy).

For pharmacy experience rating and trend calibration, this data is transformative. Actuaries currently rely on aggregate PBM reports that show net cost by broad therapeutic category. The new disclosures will enable NDC-level (National Drug Code) analysis of gross-to-net waterfalls, allowing actuaries to build trend models at the drug level rather than the category level. The initial comment period closed March 31, 2026 (extended to April 15 for an additional 15 days), and a final rule is expected in 2026, with effectiveness for plan years beginning on or after July 1, 2026, for calendar-year plans (WTW, February 2026). Plan fiduciaries will also gain audit rights to verify the accuracy of PBM disclosures, a provision that the Epstein Becker Green employer guide emphasizes as a critical compliance tool (EBG, 2026).

Worked Example: Transitioning a 1,500-Life Group

Consider a self-insured employer group with 1,500 employees and dependents. Under the prior framework, the actuary projected pharmacy cost using a single net trend of 9.0%, reflecting a blend of ingredient cost escalation, utilization growth, and an implicit PBM spread of approximately 3 percentage points. The plan's pharmacy PMPM was $145 on a net basis.

Under the four-component decomposition:

Gross ingredient cost trend: 6.5% (AWP-based unit cost escalation, blending brand at 8% and generic at flat)
Utilization and mix shift: 4.0% (driven primarily by GLP-1 adoption and specialty pipeline)
Combined gross trend: 10.5% (multiplicative: 1.065 * 1.040 = 1.1076, or approximately 10.8% before rounding)
Rebate credit: -28% of gross brand cost (approximately -7.5% of total gross pharmacy cost)
Rebate credit trend: -1.5% annually (rebates growing slightly as a share of gross cost)
PBM administrative fee: $5.50 PMPM, trending at 3.0%

The net effective pharmacy trend under the new framework: gross trend of 10.8% minus the rebate credit improvement of 1.5%, producing a net cost trend of approximately 9.3% before the administrative fee layer. Adding the fee component at $5.50 PMPM (roughly 3.4% of the $160 gross pharmacy PMPM) brings the all-in trend to approximately 9.0%: arithmetically similar to the old net trend, but now fully decomposed and auditable.

The critical difference: the plan sponsor can now see that 10.8% of gross cost growth is partially offset by a 28% rebate credit, rather than receiving a single opaque 9.0% number. If the PBM negotiates better rebates next year, the credit percentage rises and the net trend falls visibly. Under the old regime, that improvement would have been partially captured by the PBM as retained spread.

For credibility weighting, this plan has zero quarters of gross-cost data. Using the Buhlmann framework with k = 8 quarters: Z = 0 / (0 + 8) = 0. The blended trend is 100% industry benchmark for the first renewal cycle. After one year (four quarters of post-transition data): Z = 4 / (4 + 8) = 0.33. The plan's own gross experience begins to influence the trend selection at roughly one-third weight.

Why This Matters for Pricing Actuaries

The CAA 2026 PBM reform is not merely a compliance event. It restructures the information architecture of pharmacy pricing in employer health plans. Five implications demand attention in the current renewal cycle.

First, every existing pharmacy trend assumption in the renewal rating model needs to be restated on a gross-plus-credit basis. Carrying forward net trend factors developed under the old regime will systematically misstate the cost structure, even if the arithmetic coincidentally produces a similar bottom-line number. The decomposition is required for defensibility, not just accuracy.

Second, the transition creates a one-time step-down in total pharmacy cost for plans moving from partial pass-through to full pass-through contracts. West Virginia's experience suggests this step-down is on the order of 4 to 7 percentage points of the total rate increase. Actuaries must model this as a level adjustment in the projection year, separate from the ongoing trend.

Third, the DOL disclosure data will rapidly improve the quality of pharmacy experience rating. Within two to three years, actuaries will have drug-level gross-to-net data for every large self-insured plan, enabling segmented trend analysis by therapeutic class, brand/generic split, and specialty tier. This is a step change in analytical capability.

Fourth, stop-loss pricing must be recalibrated. The elimination of PBM spread changes the distribution of large pharmacy claims because the plan now bears the full gross cost offset by the full rebate credit, rather than the partially netted amount. For specific stop-loss, the expected claim size at the deductible threshold shifts, and excess loss factors need restatement.

Fifth, the GLP-1 utilization surge is compounding the transition challenge. GLP-1 drugs carry low rebate levels relative to established therapeutic classes, meaning the rebate credit offset is smaller for the fastest-growing cost driver. Actuaries modeling the four-component decomposition must apply class-specific rebate percentages rather than a blended average, or the projection will understate the net cost of the GLP-1 component.