Judge Brendan Hurson of the US District Court for the District of Maryland enjoined eight provisions of CMS's 2027 Notice of Benefit and Payment Parameters on July 16, 2026, four days before they were due to take effect, and ordered Exchanges to stop removing enrollees for failing to file and reconcile prior-year premium tax credits (AHA News, July 2026).

Insurers had already filed 2027 rates assuming that cutoff would keep thinning the pool.

Key Takeaways

  • The two-year failure-to-file standard now governs both PY2026 and PY2027. CMS's July 22 guidance directed Exchanges to stop denying or removing advance premium tax credits from any enrollee on the one-year basis.
  • Issuers attributed roughly four percentage points of both the 2026 and projected 2027 increases to a sicker residual pool, on a median proposed 2027 increase of 14% across 77 filers in 16 states and Washington, D.C.
  • Effectuated enrollment had already fallen to 19.2 million from 22.1 million, a 13% decline HHS attributes almost entirely to removing roughly 2.9 million enrollees it considers improperly enrolled.
  • The injunction removes the morbidity-improving half of a net load, not the whole load: affordability-driven attrition continues while the improper-enrollee purge stops.
  • The 2027 FFE user fee fell to 1.9% from 2.5% in the same unlitigated rule, so a fee cut calibrated to a smaller pool now sits in the same filing as a morbidity correction built on a larger one.

What the July 16 Order Reopened

The case is City of Columbus v. Kennedy, No. 26-cv-2215 in the District of Maryland, filed June 3, 2026 by a coalition of municipalities, physician groups and small-business advocates against a Payment Notice CMS had finalized six weeks earlier (Groom Law Group, 2026).

Judge Hurson stayed eight provisions on Administrative Procedure Act grounds. The two that reach an actuarial memorandum sit inside 45 C.F.R. section 155.305(f)(4): the one-year failure-to-file-and-reconcile cutoff for Federally-Facilitated Exchange enrollees, and the rollback of the automatic 60-day extension granted when reported income does not match IRS or Social Security data (Norton Rose Fulbright, July 2026). The other six carry network and benefit-design consequences but do not move the enrollment base.

Before the 2027 rule, an enrollee lost eligibility only after two consecutive years of failing to reconcile. The rule cut that to one year for FFE states from plan year 2027, with state-based Exchanges required to adopt it by 2028. CMS framed it as anti-fraud: "American taxpayers deserve to know their dollars are going only to people who truly qualify," Administrator Mehmet Oz said on finalization (CMS, May 2026).

ProvisionAs Finalized, May 2026Status After July 16 Injunction
FTR cutoff, FFE statesOne-year cutoff beginning PY2027Enjoined; two-year cutoff applies through PY2027
FTR cutoff, SBE statesState option through PY2027, mandatory one-year by PY2028Enjoined; SBEs retain full discretion while stay is in force
60-day income-inconsistency extensionEliminatedReinstated per CMS's July 22 guidance
Documentation for sub-100% FPL and no-tax-data applicantsNew verification requirementEnjoined
Bronze/catastrophic OOP band expansion (130%)Effective PY2027 (bronze), PY2028 (catastrophic)Enjoined
FFE and SBE-FP user fees1.9% FFE, 1.5% SBE-FP, both down from 2026Not challenged; in effect for PY2027

The Half of the Pool Load That Just Went Away

Marketplace enrollment had already taken its largest single-year contraction. Effectuated enrollment fell to 19.2 million in February 2026 from 22.1 million a year earlier, a 13% decline HHS attributes almost entirely to removing roughly 2.9 million improperly enrolled people rather than to affordability (ASPE, June 2026).

That framing is the opposite of an adverse-selection story, and it is what the 2027 filings priced. Across 77 filers in 16 states and Washington, D.C., the median proposed increase was 14%, with issuers attributing about four percentage points of it to a sicker residual pool as eligibility tightening continued (Peterson-KFF, July 2026).

A pool-effect load of that kind is a net number. It sets the higher per-member cost of a smaller, sicker pool against the improvement from removing enrollees HHS considers non-representative of genuine risk. The injunction leaves affordability-driven attrition untouched and stops the removals, which takes out the offsetting half the four-point load was built on.

The correction is not a sign flip, and that is the part that costs work. Morbidity assumptions are layered: the first layer is trend applied to the base period and is unaffected here; the second is the pool-composition adjustment for who is expected to be enrolled in the rating period.

Restoring an enrollee who failed to reconcile a prior subsidy does not restore someone with average-pool morbidity. By HHS's own characterization of the 2.9 million, that population skews toward lower income-verification confidence. Re-estimating the second load therefore needs either encounter data on enrollees CMS had begun flagging before July 16, or a proxy built from the demographic and subsidy-tier profile of the 2024 and 2025 sweeps.

A Fee Cut Calibrated to the Rule as Finalized

The same rule cut the administrative fee load, and that piece was never litigated, so it now sits in the rate build alongside a morbidity assumption pointing the other way.

CMS set the 2027 FFE user fee at 1.9% of premium, down from 2.5% in 2026, the SBE-FP fee at 1.5% from 2.0%, and the risk adjustment user fee at $0.18 per member per month from $0.20 (CMS, May 2026). A 0.6-percentage-point fee cut against a benchmark silver premium is a comparable order of magnitude to the pool load actuaries are now rebuilding upward, and it was calibrated against the rule as finalized rather than as enjoined.

Whether an issuer's 2027 rate lands above or below its original filing therefore depends on the ratio between its own morbidity correction and its fee relief, and that ratio is book-specific. An issuer whose base skewed toward subsidized enrollees with thin tax-filing histories carries a larger correction against the same fee benefit than one with a more stable base.

The calendar is the binding constraint on how well either can be done. CMS issued revised Plan Year 2027 QHP data submission and certification timeline bulletins on July 31 and August 4, 2026, and a full re-estimation from restored-enrollee encounter data takes longer to assemble than that window allows in most states.

Issuers under CMS-performed rate review work to those bulletins directly. Issuers in states with independent review face a separate question of whether the regulator will accept a mid-cycle amendment to a memorandum whose assumption was already under active legal challenge when it was filed. Either way the weight falls on the memorandum's narrative section rather than its numeric exhibits, because that is where the enrollment-policy assumption and its confidence level are documented.

Further Reading

Sources