GLOBE Final Rule: What Changed and Why It Matters for Global Pricing

SyMAP product update: The GLOBE final rule is now reflected in SyMAP’s financial modelling, with tools to assess rebate exposure and test a hypothetical extension. Explore the updated GLOBE model.

How we got to the GLOBE final rule

CMS announced the final rule for the Global Benchmark for Efficient Drug Pricing (GLOBE) Model on 30 September 2026. It was published in the Federal Register on 2 October. GLOBE is the Part B component of Medicare’s Most Favored Nation drug pricing models, testing an alternative way to calculate inflation rebates using prices from 19 reference countries. We covered the proposed design in Decoding GLOBE & GUARD. Its Part D counterpart, GUARD, is still under review at the Office of Management and Budget.

The final rule excludes more products, and CMS intends to waive GLOBE for manufacturers taking part in the GENEROUS Medicaid model. CMS’s estimate of total federal Medicare savings fell from $11.9 billion to $440 million. These changes give manufacturers a firmer basis for assessing potential Part B exposure, including scenarios in which a GENEROUS waiver no longer applies.

What changed in the GLOBE final rule

The basic design is unchanged. The main revisions concern timing, product exclusions, the source of the GDP adjusters, missing benchmarks and participation in GENEROUS.

The rule retains the 19 reference countries, the seven drug classes and the random selection of ZIP-code areas covering about a quarter of Original Medicare Part B beneficiaries. The benchmark remains the greater of the Method I and Method II calculations, with thresholds of 102% and 105%. The final rule takes effect on 30 November 2026.

Pricing teams now have two connected jobs: put a value on GLOBE exposure, and decide what that value means for their global pricing strategy. The first depends on getting the final rule’s dates, exclusions and benchmarks right. The second uses that analysis to judge voluntary MFN commitments and international pricing decisions, including the exposure that could emerge when agreements end. We start with the strategic choices, then turn to the modelling needed to assess them.

Why GLOBE matters for MFN strategy

Back to section links

GLOBE’s narrower scope should not obscure the significance of its finalisation. CMS proceeded with the rule despite calls to withdraw it, retaining international benchmarking within the regulatory framework for Medicare Part B rebates. In our view, that gives MFN a firmer regulatory footing and strengthens the case for planning around a continuing link between US revenues and international prices. The model can still be challenged or changed, but manufacturers should be cautious about assuming that MFN will disappear when today’s voluntary agreements end.

Effects on prices outside the US

CMS’s own impact analysis assumes that manufacturers will respond to GLOBE by reporting higher international net prices and raising international prices over time, an effect it phases in from 10% of 2027 price data to 60% by 2030, reducing GLOBE rebates by 60% by the end of the model. In other words, the regulator expects GLOBE’s economics to play out in reference-country prices.

Changes in publicly referenced prices can also affect prices in other countries through international reference pricing. We therefore see a case for modelling GLOBE alongside IRP to capture those wider effects, as we argued in The End of Two Silos.

What the participation estimate leaves out

In CMS’s low-participation scenario, assumed GENEROUS uptake reduces the illustrative GLOBE manufacturer count from 19 to four. Manufacturers receiving waivers still have a reason to model GLOBE: it helps assess the value of GENEROUS participation and the financial implications of any permitted change to their commitments.

CMS declined to write an exemption into the rule. Instead, it intends to waive mandatory participation for each manufacturer, quarter by quarter, while that manufacturer participates in GENEROUS. It would do so under the authority used to run GENEROUS as a voluntary model. The waiver is not codified; its terms sit in GENEROUS participation agreements.

Codified versus voluntary

GLOBE is codified in 42 CFR Part 513. Renegotiation and withdrawal from GENEROUS and manufacturer agreements are governed by the terms of those agreements.

GLOBE’s scheduled performance period extends 15 months beyond GENEROUS’s, which is scheduled to end on 31 December 2030. The final rule ties the intended waiver to quarterly GENEROUS participation and provides no express continuing exemption afterwards. Unless that protection is extended, manufacturers with eligible Part B products could therefore face GLOBE exposure during its remaining quarters. The precise timing will depend on their participation agreement terms and any subsequent CMS changes.

The zero tariff rate for manufacturers with both an MFN agreement and an approved onshoring plan ends earlier, on 20 January 2029, when the next presidential term begins.

For qualifying Part B products, GLOBE can therefore act as a backstop if a manufacturer leaves GENEROUS, the model ends or a future administration unwinds the voluntary framework. That exposure depends on GLOBE remaining in force.

What GLOBE means for the value of GENEROUS participation

GENEROUS delivers supplemental Medicaid rebates against an eight-country net-price benchmark; GLOBE adjusts Part B rebates against a 19-country benchmark. The rebate calculations are product-specific; the participation decision must reflect the portfolio and obligations covered by the manufacturer’s agreements. GLOBE’s narrower reach could make a permitted withdrawal financially attractive for some portfolios. Establishing that requires comparing the GENEROUS commitments that would actually cease with additional GLOBE exposure, rebates that would continue and any access or other benefits that would be lost.

Manufacturers should not assume they can withdraw individual drugs independently. The available options and their timing depend on the participation agreement, negotiated product terms and state rebate contracts, as outlined in the CMS GENEROUS manufacturer application. MFN scenario modelling can then assess those options across the relevant portfolio.

What to change in your GLOBE model

Back to section links

Models built on the proposal need updated dates, reference prices, adjusters and product lists. Those changes can alter exposure even where the underlying formula stays the same.

Later performance years and a new reference quarter

The five-year performance period now runs from April 2027 to March 2032, six months later than the proposed October 2026 to September 2031 window. There is no GLOBE rebate exposure before April 2027. At the other end, a drug entering late could face up to two additional quarters before the model finishes.

For pricing teams, the shift in the reference period may matter more. GLOBE benchmarks use prices from the applicable ASP calendar quarter, two quarters before the quarter being rebated. The first performance quarter now looks back to Q4 2026 rather than Q2. That puts the initial reference period in the current quarter. The first window for manufacturer-submitted international net prices opens on 1 January 2027, covering the quarter that began on 1 October 2026.

International and US price movements in the second half of 2026 therefore feed the first benchmarks. Drug selection also shifts, because it uses Part B spending over a 12-month period ending six months before each quarter. Simulations that fixed reference prices or drug lists at mid-2026 will need updating.

Same reference countries, new adjusters

The GDP (PPP) adjuster still scales each country’s prices by the gap between US and local GDP per person. The data source changes to the World Bank. CMS’s new illustrative values are lower for 13 of the 19 countries, including the United Kingdom (1.438 to 1.383), South Korea (1.498 to 1.406), Spain (1.560 to 1.480) and Israel (1.596 to 1.500).

A lower adjuster reduces that country’s PPP-adjusted reference price and can lower the benchmark. CMS will use the most recent data available, so the values will need refreshing. The badges below show the 19 GLOBE reference countries, their final illustrative adjusters and the percentage change from the proposed values, rounded to one decimal place. Ireland, the Netherlands, Norway and Switzerland are at the 1.000 floor.

You can compare every country’s figures in our MFN basket countries explorer.

Fewer products, same geography

The additional exclusions remove some products from GLOBE altogether. The geographic design still covers about 25% of Original Medicare Part B beneficiaries, but CMS has yet to identify the areas. It must select them by 1 January 2027 and publish the list by 16 January. Until then, models need an assumption for the share of each product’s Part B use that will fall within those areas.

Benchmark re-checks and launch sequencing

Under the proposal, a drug would have had no Method I benchmark for the whole model if CMS could not find international pricing in its first quarter. The final rule requires CMS to re-check existing data sources each quarter while a benchmark is unavailable. A drug can therefore receive a Method I benchmark later if suitable pricing data become available.

What the final rule keeps is the lock-in we described in The Fine Print. Once a Method I benchmark is identified, it stays in place until the end of the model. The first reference-country prices CMS can observe therefore set the benchmark for every remaining quarter. For launch planning, the sequence and level of reference-country prices now matter more than their absence. CMS has also said it will monitor manufacturers’ strategic responses, so this is a consideration for planning, not a workaround.

The Method II decision arrives in January

Manufacturers can choose to submit their international net prices to CMS for a Method II benchmark, starting with the window that opens on 1 January 2027. Because the benchmark is the greater of the two methods, a Method II submission can raise the benchmark but cannot lower it, and when Method II is the higher of the two, the threshold rises from 102% to 105%. CMS expects manufacturers to submit where their data show higher prices than Method I. Submission comes with a data agreement and reporting requirements, so the choice is not free, but it is a calculation each manufacturer can make product by product before the first window opens.

Entry and exit points across the lifecycle

GLOBE exposure needs to be assessed product by product, quarter by quarter. A drug must fall within one of the seven classes and meet the spending test. The threshold starts at $100 million for Q2 2027 and is adjusted quarterly for inflation. Once met, the spending criterion remains satisfied even if spending subsequently falls. The orphan-only exclusion lasts while every approved indication is for a rare disease, so a non-orphan indication can bring a drug into the model.

According to CMS, biosimilars and their reference products are excluded once a biosimilar is marketed. A negotiated Maximum Fair Price also takes a drug out, with the earliest Part B prices applying from 1 January 2028. These entry and exit points make indication sequencing, loss-of-exclusivity timing and selection for negotiation part of the exposure calculation.

What to watch next

• 1 January 2027: GLOBE begins, and the first Method II reporting window opens for fourth-quarter 2026 prices.
• By 16 January 2027: CMS publishes the list of GLOBE geographic areas.
• 1 April 2027: the first performance year begins.
• GUARD: the Part D final rule has been under OMB review since 15 June 2026. GLOBE is its likely template, and Part D is where more of the spending sits.
• GENEROUS: the final state count after the 30 September deadline, and any extension of the model.
• Challenges: any litigation or rulemaking that changes the final rule.

GLOBE final rule: frequently asked questions

When does the GLOBE Model start?

The GLOBE Model begins on 1 January 2027, when manufacturers can start submitting international net pricing data. Its five performance years run from 1 April 2027 to 31 March 2032, and the final rule takes effect on 30 November 2026.

Which countries are in the GLOBE reference basket?

Nineteen: Australia, Austria, Belgium, Canada, the Czech Republic, Denmark, France, Germany, Ireland, Israel, Italy, Japan, the Netherlands, Norway, South Korea, Spain, Sweden, Switzerland and the United Kingdom. The final rule kept the proposed list.

Which drugs does GLOBE cover?

Separately payable Part B single source drugs and biological products in seven drug classes that meet the Original Medicare Part B spending test, measured over 12 months. The threshold starts at $100 million for Q2 2027 and is adjusted quarterly for inflation; once met, the spending criterion remains satisfied even if spending subsequently falls. The model covers beneficiaries in randomly selected areas representing about a quarter of Original Medicare Part B. Orphan-only drugs, FDA-listed cell and gene therapies, plasma-derived products and drugs with a negotiated Maximum Fair Price are excluded.

Are GENEROUS participants exempt from GLOBE?

Not by rule. CMS declined to codify an exemption, but it intends to waive mandatory GLOBE participation for a manufacturer in each quarter in which that manufacturer participates in GENEROUS. The final rule provides no express continuing exemption after participation ends. The precise duration of protection depends on the participation agreement terms and any subsequent CMS changes.

Is the GUARD Model final?

No. The GUARD final rule for Medicare Part D has been under OMB review since 15 June 2026 and had not been published by 3 October 2026.