Executive Summary
In the span of two days in late July 2026, The Wall Street Journal ran two pieces on the No Surprises Act’s Independent Dispute Resolution process: a July 23 news analysis of new CMS data putting 2025 provider payouts at $14.85 billion, more than triple 2024, and a July 24 editorial board piece titled “The ‘Surprise Billing’ Racket.”1,2 A CMS spokesperson supplied the soundbite, telling reporters “the system is being gamed to get higher prices.”3 The insurance lobby is now spending six figures converting that framing into momentum against reform it opposes.
This brief tests the “racket” narrative against the federal record: CMS public use files, the Congressional Budget Office, the Congressional Research Service, the Government Accountability Office, and, notably, the insurance industry’s own published survey data. The narrative does not survive contact with the evidence.
Five findings anchor this brief. First, IDR is a rarely-used backstop, not a runaway machine: by the insurers’ own numbers, fewer than one in two thousand commercial claims reaches arbitration. Second, dispute volume exceeded projections because the federal projection was built on a single-state dataset that missed millions of annual surprise-billing events, not because providers invented claims. Third, the widely repeated 40% ineligibility figure conflates objections filed with objections sustained; the adjudicated rate is well under half that. Fourth, provider win rates reflect the benchmark Congress deliberately rejected, insurer lowballing, and payer procedural defaults, not arbitrator capture. Fifth, the patient-protection core of the law is working by both sides’ account, and network participation has risen, not collapsed.
Outlier abuse exists, and this brief names it. But policy built on outliers, marketed with a manufactured statistic, is not reform. It is a rate-setting campaign wearing reform’s clothes.
01 The Denominator Problem
Start with the number neither Journal piece prints. In June 2026, the Congressional Budget Office observed that insurer surveys indicate less than 0.05 percent of all claims go to arbitration, citing AHIP’s 2024 survey.4 That is the insurance trade association’s own data, restated by Congress’s nonpartisan scorekeeper.
AHIP’s underlying report is worth reading closely. Surveying insurers covering roughly two-thirds of the commercial market, AHIP estimated 10.2 million NSA-eligible claims in the first three quarters of 2023, about 0.7% of all commercial claims processed. Of those, approximately 8.0 million, or 79%, were resolved when the provider simply accepted the insurer’s initial payment. Fewer than a quarter entered open negotiation at all, and 73% of those resolved without an IDR filing. Roughly 668,000 claims, about 6.6% of the eligible pool, were ever submitted to arbitration.5
The entire $15 billion controversy lives inside a fraction of one percent of American commercial claims, in a process most eligible claims never touch. That is not a system devouring healthcare. That is a backstop functioning as designed.
Payouts concentrate where insurers underpay most aggressively. When the initial payment is adequate, providers take it; the AHIP data prove they overwhelmingly do. Arbitration volume maps onto the residue of disputes where the first offer failed, which is precisely the population the mechanism was built to resolve.
02 Why Volume Exploded: A Flawed Forecast, Not a Gold Rush
The editorial leans hard on the gap between the government’s original projection of roughly 17,000 to 22,000 disputes per year and the 2.56 million disputes initiated in 2025.2,9 The implication is that the excess is fabricated demand. The projection’s own provenance says otherwise.
The federal estimate was extrapolated from a 2018 New York State dataset that bore no relationship to national out-of-network exposure.10 Pre-NSA research put surprise billing at roughly 18% of emergency department visits and 16% of in-network inpatient stays for commercially insured patients; applied to national utilization, that implies on the order of 7.5 to 8 million surprise-billing instances per year.10,11 GAO data show in-network claim rates for the four most-affected specialties had already slid from about 98% to 90% between 2019 and 2021, before the first dispute was ever filed.12 The eligible pool was always millions deep. The forecast missed it by two orders of magnitude, and the miss is now being reclassified as provider misconduct.
2.56 million disputes were initiated in 2025 against an estimated 7.5 to 8 million annual surprise-billing events. Even at peak volume, and even before accounting for disputes rejected, withdrawn, or settled, the majority of billing events the NSA governs never generate an arbitration filing.9,10
03 The Ineligibility Myth
The payer lobby’s most effective talking point is that roughly 40% of provider disputes are “ineligible,” implying industrial-scale improper filing. The figure comes from an AHIP/BCBSA survey of insurers describing their own objections.13 It measures accusations, not adjudications.
The adjudicated record is public. The Congressional Research Service reports that approximately 19% of disputes closed in 2024 were found ineligible, down from roughly 22% in 2023.7 CMS public use data covering the life of the program show ineligibility determinations near 17%, trending downward.6 CMS’s own H2 2025 release shows the non-initiating party challenged eligibility on 42% of disputes, a number that quietly became “42% ineligible” in advocacy material.9 Filing an objection is free of consequence; sustaining one requires evidence. The gap between those two numbers is the tell.
The grounds that do produce ineligibility findings are dominantly procedural and jurisdictional: state-law claims filed federally, timing errors, improper batching. CMS’s May 2026 Operations Final Rule addresses exactly these failure points with standardized remittance codes identifying the correct venue, clearer batching rules, and defined documentation windows.14 Regulators treated ineligibility as a plumbing problem. The lobby markets it as fraud.
04 Why Providers Win
Providers prevailed in roughly 80% of determinations in 2023 and 85% in 2024.7 Critics treat the win rate itself as proof of corruption. The record supports three duller explanations.
The benchmark argument assumes a rule Congress rejected. Award multiples of the Qualifying Payment Amount look scandalous only if the QPA is the correct price for out-of-network care. The QPA is the payer-calculated median of 2019 in-network contracted rates, terms out-of-network providers never signed, reflecting discounts exchanged for patient volume they never received. Congress considered pegging OON payment to in-network or Medicare benchmarks and declined, choosing multi-factor arbitration precisely to avoid administrative price-setting; federal courts in the Texas Medical Association litigation then struck down regulators’ attempts to restore QPA primacy through rulemaking.15 The out-of-network premium long predates this law: pre-NSA research documented OON emergency physician charges averaging 637% of Medicare against 266% for identical in-network services.16 Measuring IDR against QPA convergence is not analysis of the statute. It is advocacy for the statute the insurers wanted and lost.
Baseball-style arbitration punishes lowballing. The arbitrator must select one offer. A payer that anchors to a depressed QPA has submitted a number built to lose against any credible offer reflecting the actual OON market. Insurer offers are accepted roughly a fifth of the time or less, and when payers do prevail, outcomes land near 110% of QPA, evidence that reasonable payer offers win.1,17 The win rate is partly a mirror held up to payer bidding strategy.
Payers forfeit a startling share of disputes. In Q4 2024, 26% of disputes were resolved by default because one side failed to submit complete information, and roughly 90% of those defaults went against the payer side.17 A party that does not fully participate in a proceeding should not be surprised by its results, and should not be permitted to cite those results as evidence the tribunal is rigged.
05 The Market Context the Editorials Skip
None of this happened in a vacuum. Independent physicians did not drift out-of-network as a lifestyle choice; for many specialties, in-network rates stopped covering the cost of running a practice. The result is visible in every market in America: independent groups absorbed into hospital systems, staffing consolidated under national platforms, private practice in structural retreat. Reimbursement pressure from consolidated payers is a primary driver, and the same carriers applying that pressure have vertically integrated into PBMs, surgery centers, physician employment, and, in at least one case, a chartered bank. The Journal’s editorial board has yet to find a racket in any of that.
The squeeze is measurable. Adjusted for practice-cost inflation, Medicare physician payment fell 33 percent between 2001 and 2025 while the cost of running a practice rose 59 percent.19 The procedure-level record is starker: inflation-adjusted Medicare payment for hip and knee replacement down 31 to 41 percent from 2000 to 2019 and as much as 56 percent through 2024, with revision arthroplasty down 37 to 39 percent even as case volumes climbed.20 Anesthesiology, emergency medicine, and spine surgery run the same slope: down 21 to 34 percent in real terms.21 Commercial fee schedules key off these anchors; when the anchor falls by a third in real terms, the in-network offer built on it stops covering the cost of care; physicians sell, consolidate, or leave the network. Awards reflecting genuine out-of-network market rates look outsized only against that depressed baseline.
The predicted counterfactual also failed. Critics warned IDR would pay providers to flee networks. GAO found the opposite: post-NSA, in-network claim shares rose, or reversed prior declines, across the affected specialties.12 Patients are protected, out-of-pocket spending on covered services fell 16.5%, and both sides told RAND the surprise-billing problem is largely solved for covered care.8,18 By its statutory purpose, the law is succeeding.
06 The Bad Actors, Named Honestly
Credibility requires conceding what is true. Some award patterns on elective procedures are indefensible on their face: six-figure awards on surgeries reimbursed in four figures in-network.2,22
But look inside even the ugliest awards, because the mechanics matter. IDR is final-offer arbitration: each side submits one number, the arbitrator must select one, no compromise, no appeal. In the most notorious case in the Times investigation, the two numbers in front of the arbitrator for a surgical assist on a 4.5-hour breast reconstruction were the provider’s $100,000 and the insurer’s $105. In another, a plan sent $30 for a three-hour robotic surgery.22 A six-figure outlier award cannot exist without a three-figure counteroffer beneath it. Final-offer arbitration is engineered to punish the least credible number in the room, and it does not care which side submits it. Insurers that bid real market rates do not lose to absurd asks. The outliers are co-authored.
But note what the lobby does with those outliers. It generalizes from the tail of the distribution to condemn a process that, by its own survey, touches a twentieth of one percent of claims, and it aims the resulting outrage at the one bill it fears most.
07 What the Enforcement Act Actually Does
The No Surprises Act Enforcement Act (H.R. 4710 / S. 2420) is a bipartisan, physician-authored bill with one operative idea: when an insurer loses a binding arbitration, it must actually pay, within the 30 days existing law already requires, with penalties and interest for noncompliance.23 The bill exists because nonpayment and late payment of final IDR awards became routine, a pattern documented by the AMA and dozens of state and specialty societies.24 Courts have compounded the problem by holding that providers generally lack a private right of action to enforce unpaid awards, leaving winners with judgments no one is obligated to honor.25
Hold both facts at once: the carriers refusing to pay binding arbitration awards are the carriers telling Congress the arbitration process lacks integrity. The enforcement gap is the scandal. The Enforcement Act is the correction.
08 Conclusion
Strip the adjectives and the record reads simply. A protection law eliminated surprise bills and cut patient costs. Its dispute mechanism handles a sliver of claims, resolves most of them without arbitration, and pays above a benchmark Congress explicitly refused to impose, in a market where the out-of-network premium predates the law by decades. Its real defects, procedural ineligibility churn and a small population of abusive filers, are being addressed by rule and are addressable by enforcement. What is being marketed as a racket is, in the data, a functioning backstop that insurers are losing on the merits and litigating in the press.
Reform the outliers. Enforce the awards. And retire the manufactured statistics.