On February 9, 2026, Brazil’s Chamber of Deputies voted 337 to 19 to fast-track a bill declaring tirzepatide, the active ingredient in Eli Lilly’s Mounjaro and Zepbound, a matter of public interest.[16][17] Three days earlier, a separate bill had landed in the Brazilian Senate proposing a compulsory license for the same molecule.[15] Neither bill has become law as of this writing. But the vote shows something pricing teams tend to underweight: a country can move from committee obscurity to a floor vote on overriding your patent in a single sitting. The legal mechanism behind that vote, and behind two decades of similar moves from India to Colombia, is a single clause tucked into Article 31 of the TRIPS Agreement.
This piece maps what that clause actually permits, catalogs the documented pharmaceutical compulsory licenses issued since the 2001 Doha Declaration, and builds a repeatable way to read the early warning signs before a market resets your price without asking. DrugPatentWatch’s patent-family and jurisdiction-tracking tools are referenced throughout as one way to operationalize this kind of monitoring, not as a substitute for the primary legal sources cited below.
The Short Answer
Article 31 of the TRIPS Agreement lets any WTO member authorize the use of a patented invention, including a drug, without the patent holder’s consent, provided it follows a defined process.[1] The 2001 Doha Declaration on the TRIPS Agreement and Public Health confirmed that governments decide for themselves what counts as a national emergency or other circumstance justifying that step, and that public health can outweigh patent exclusivity.[1] A 2005 amendment added Article 31bis, letting a country with no manufacturing capacity import generics made under a compulsory license issued somewhere else.[2] As of a 2023 legislative survey covering 187 patent-holding countries, 176 of them (94.1 percent) had some form of compulsory licensing provision on the books, but only 72 (38.5 percent) had implemented the Article 31bis export mechanism.[3] That gap, between what the law allows and what national legislation actually operationalizes, is where most of the pricing risk in this article lives.
What Is the TRIPS Flexibility Clause, Exactly?
“TRIPS flexibility” is not one clause. It is a set of related legal tools built into the 1994 TRIPS Agreement and clarified by three later instruments: the 2001 Doha Declaration, the 2003 waiver decision, and the 2017 Article 31bis amendment.[1][2] Confusing these four instruments is a common mistake in emerging-market patent risk assessments, because each one authorizes a different action with a different trigger.
Article 31: The Original Compulsory License
The term “compulsory licensing” never appears in the TRIPS text. The operative language is “other use without authorization of the right holder,” which is the title of Article 31.[1] Under normal circumstances, a government or a third party must first attempt to negotiate a voluntary license on reasonable commercial terms before a compulsory license can be granted. That requirement is waived for national emergencies, other circumstances of extreme urgency, and public non-commercial use.[1]
Article 31(f): The Domestic-Market Requirement
Products made under a standard Article 31 compulsory license must be predominantly for the supply of the domestic market.[2] This single clause is why India’s 2012 Natco license could only be sold inside India, and it is exactly the limitation Article 31bis was built to remove.
Article 31bis and the Paragraph 6 System
Because Article 31(f) blocked countries with no domestic manufacturing capacity from importing compulsory-licensed generics, the 2001 Doha Declaration flagged the problem in its Paragraph 6.[2] WTO members adopted a waiver in 2003 and converted it into a permanent treaty amendment, Article 31bis, that entered into force in 2017 after two-thirds of members formally accepted it.[2] This is sometimes called the Paragraph 6 system, and it is the only TRIPS mechanism built specifically for cross-border generic export under compulsory license.
Parallel Importation
A fourth, separate flexibility is parallel importation: buying a patented product in one country where it sells cheaply and importing it into another without the patent holder’s consent. TRIPS leaves the exhaustion doctrine that governs this practice to national law, which is why parallel-import exposure varies enormously by jurisdiction and deserves its own risk map rather than being folded into compulsory-licensing analysis.
The Five Findings That Matter
Before the case studies, five numbers set the frame for everything below.
A 2012 database analysis in PLOS Medicine identified 34 potential pharmaceutical compulsory-licensing episodes across 26 countries between 1995 and 2011, of which 24 were verified and 12 resulted in an actual license, concentrated in upper-middle-income countries and weighted heavily toward HIV/AIDS drugs, with activity peaking between 2003 and 2005.[4]
Only 72 of 187 countries with patent legislation, roughly two in five, have implemented the Article 31bis export mechanism into domestic law, even though 176 have some form of compulsory licensing provision.[3]
Canada is the only WTO member to have completed a full Paragraph 6 export transaction in the eighteen years the mechanism has existed, a single 2007 to 2009 shipment of an HIV combination drug to Rwanda.[18][19][20]
The Medicines Patent Pool, the main voluntary-licensing alternative to compulsory licensing, in-licensed 44 products from 22 originator companies and supplied 62.14 billion doses of treatment between 2020 and 2025, but its licenses routinely exclude upper-middle-income countries such as Argentina, Brazil, China, Malaysia, and Thailand.[33][34]
Across six documented pharmaceutical compulsory-licensing cases with disclosed before-and-after pricing, the average price cut was approximately 74 percent, an original calculation explained in full further below.
A Brief History: From Article 31 to the Paragraph 6 System
TRIPS entered into force alongside the WTO in 1995, requiring member states to grant twenty-year product patents on pharmaceuticals for the first time in many developing economies. Within six years, enough governments had run into the same wall, patented HIV drugs priced out of reach of national AIDS programs, that the WTO’s Fourth Ministerial Conference in Doha adopted a standalone declaration in November 2001 confirming the agreement should be interpreted in a way that supports public health.[1]
The 2003 Waiver
Paragraph 6 of the Doha Declaration flagged a specific hole: a compulsory license under Article 31(f) can only supply the domestic market, so a country with no factories of its own could not benefit. On August 30, 2003, WTO members adopted a General Council decision waiving Article 31(f) and 31(h) obligations for the specific purpose of exporting compulsory-licensed medicines to countries that need them.[2]
The 2017 Amendment
That waiver became a permanent treaty amendment, Article 31bis, once two-thirds of WTO members formally accepted it, a threshold reached in January 2017.[2] The mechanism requires the importing country to notify the WTO of the product and quantity needed, and the exporting country to notify the terms of its compulsory license, a two-step paper trail that has proven to be the system’s main practical bottleneck.[2]
Timeline: Major Pharmaceutical Compulsory Licenses Since Doha
The table below lists documented, publicly confirmed pharmaceutical compulsory or government-use licenses discussed in this article, in chronological order. It excludes proposals that were withdrawn or never resulted in a formal government act.
Year
Country
Drug
Legal Basis
Documented Price Effect
Source
2006 to 2007
Thailand
Efavirenz, lopinavir/ritonavir, clopidogrel
Government use, Article 31(b)
Efavirenz cut from 1,400 to 650 baht per month, about 54 percent
[8][11]
2007
Brazil
Efavirenz
Compulsory license, Decree 6.108
Cut from about 580 to 166 dollars per patient per year, about 71 percent
[13]
2007 to 2008
Canada (export to Rwanda)
Zidovudine/lamivudine/nevirapine (Apo-TriAvir)
Article 31bis, Paragraph 6 export
Winning tender price of 19.5 cents per tablet
[18][19][20]
2010
Ecuador
Ritonavir
Compulsory license, Decree 118
About 27 percent cut on the first competitive purchase
[27][28]
2012
India
Sorafenib (Nexavar)
Compulsory license, Section 84
Cut from about 5,500 to 175 dollars per month, about 97 percent
[5][6]
2017
Malaysia
Sofosbuvir
Government use, Section 84
Cut from about 11,000 to 300 dollars per twelve-week course, about 97 percent
[29]
2020
Indonesia
Remdesivir, favipiravir
Government use, Presidential Decrees 100 and 101
Not separately disclosed
[31]
2023 to 2024
Colombia
Dolutegravir
Compulsory license, Declaration of Public Interest
Government estimate of about 1,224 to 44 dollars per patient per year, about 96 percent
IQVIA estimates that pharmerging markets added roughly 140 billion dollars in medicine spending on the path to 2025, and that these markets’ share of global pharmaceutical spending rose from about one-fifth in 2007 to roughly one-third by 2023, which is the underlying reason patent holders now have far more emerging-market revenue exposed to this clause than a decade ago.[44]
An Original Calculation: Average Price Reduction Across Six Documented Cases
Using the six episodes above with publicly disclosed before-and-after pricing, India’s Nexavar (about 97 percent), Thailand’s efavirenz (about 54 percent), Brazil’s efavirenz (about 71 percent), Colombia’s dolutegravir (about 96 percent), Malaysia’s sofosbuvir (about 97 percent), and Ecuador’s ritonavir (about 27 percent), the simple average price reduction is approximately 74 percent. This is an original calculation drawn from the six sourced figures above, not an independently reported statistic. It should be read as illustrative rather than predictive: it says nothing about the two cases in the table with no disclosed price effect, and it says nothing about how a manufacturer’s price responds before a license is ever issued, which the Thailand and Brazil cases below show is often the more consequential number for a pricing team.
Case Study: India’s Nexavar Decision Set the Template
Bayer launched Nexavar (sorafenib) in India in 2008, after receiving its Indian patent on March 3 of that year, pricing a month’s treatment for kidney and liver cancer at roughly 280,000 rupees, about 5,500 dollars.[5][6] The World Health Organization estimated around 29,000 Indians were living with the relevant cancers in 2008, yet fewer than 200 were on the drug by 2011.[5] Natco Pharma applied for a compulsory license, and on March 12, 2012, India’s Controller General of Patents granted it under Section 84 of the Indian Patents Act, the first compulsory license the country had ever issued.[6][7]
The Three-Factor Test
The Controller’s decision, and the Intellectual Property Appellate Board ruling that later upheld it, turned on three statutory conditions: whether the reasonable requirements of the public had been satisfied, whether the patented invention was available at a reasonably affordable price, and whether it was worked in India. Bayer had supplied Nexavar to fewer than 200 of an estimated 8,842 eligible patients in 2011, had not manufactured the drug locally, and had only launched a patient-assistance program after the compulsory-license proceeding began.[5][6] The Board treated affordability, availability, and local working as interdependent tests, not three separate hurdles a company could satisfy piecemeal.
What Survived Appeal
Natco’s license set a 6 percent royalty, later raised to 7 percent on appeal, capped the generic price at 8,880 rupees, about 176 dollars, for a month’s supply, and barred Natco from outsourcing manufacture.[6][7] Bayer challenged the license through the Appellate Board, the Bombay High Court, and finally the Supreme Court of India, losing at every stage over roughly three years, with no change to the license’s core terms.[7] For a pricing team, the lesson is not that India is uniquely aggressive. It is that “reasonably affordable” is a legal standard a patent office will apply using the manufacturer’s own supply and access data, which means a company’s own patient-access reporting can become the evidentiary basis for the license that overrides it.
Case Study: Thailand’s Triple License and the Retaliation Playbook
Thailand’s Ministry of Public Health issued government-use licenses against three patents in ten weeks: efavirenz (Merck’s Stocrin) in November 2006, and lopinavir/ritonavir (Abbott’s Kaletra) and clopidogrel (Sanofi-Aventis and BMS’s Plavix) on January 26, 2007.[8][10] Thailand used the Article 31(b) national-emergency and government-use route, which does not require prior negotiation with the patent holder, and imported over 66,000 bottles of generic efavirenz from Ranbaxy in India within weeks of the decision, at roughly half of Merck’s price.[10][11]
The Corporate Response
Abbott’s reaction to the Kaletra license shows how a threatened compulsory license can move a price before any government act takes effect. Abbott offered to cut its Thai price from 347 to 167 dollars a month, roughly in line with what it already charged lower-middle-income countries, in a bid to head off the license.[9] Thailand proceeded with the license anyway.
The Political Cost
The US Trade Representative’s office pressed Thailand to reconsider, including a call from the Deputy USTR that Thai officials described publicly as bullying, while twenty-two members of the US Congress wrote to the USTR urging Washington not to interfere with a decision they viewed as legally sound.[10][12] Thailand did not reverse any of the three licenses. The durable lesson for a pricing team assessing political risk is that a compulsory license, once issued under the emergency track, has historically proven resistant to bilateral trade pressure once the domestic public-health rationale is on the record, even when that pressure comes from the country’s largest trading partner.
Case Study: Colombia’s Dolutegravir Timeline Shows the Full Sequence
Colombia’s Ministry of Health issued a Declaration of Public Interest for dolutegravir, ViiV Healthcare’s first-line HIV drug, on October 4, 2023, citing a price of about 1,224 dollars per patient per year against a roughly 44-dollar generic already available through the Pan American Health Organization.[23][24] The declaration was explicitly framed around the country’s Venezuelan migrant population, estimated at 2.9 million people as of October 2022, among whom UNAIDS reported an HIV prevalence roughly double the national adult rate.[25]
Why ViiV’s Own Licensing Model Created the Opening
Generic dolutegravir was already available worldwide through ViiV’s voluntary license with the Medicines Patent Pool, but that license excluded Colombia and a list of other populous middle-income countries, letting ViiV maintain exclusive pricing in exactly the markets that could not reach the generic version already circulating elsewhere.[25] That exclusion is what Colombia’s compulsory-license process was built to override.
From Declaration to License: A Seven-Month Process
Colombia’s patent office, the Superintendency of Industry and Commerce, moved from the October 2023 declaration to inviting generic suppliers to express interest by February 2024, and issued the compulsory license itself on April 25, 2024, roughly seven months after the initial declaration.[23][25] ViiV opposed the license throughout the administrative process and stated afterward that it did not consider compulsory licenses an effective route to sustainable access, while continuing to supply the Colombian market under existing arrangements.[26] The Colombian government projected the license could cut the treatment’s cost by as much as 80 percent.[24]
Case Study: Brazil’s 2026 Mounjaro Bills Show How Fast Overnight Really Is
Brazil already has a compulsory-licensing precedent, its own 2007 efavirenz decree, issued under Article 71 of the Industrial Property Law after Merck rejected a Brazilian counteroffer of 45 cents a pill against Merck’s own price of 1.59 dollars.[13] Nineteen years later, that same statutory mechanism is the vehicle for a very different kind of target.
Three Bills, One Molecule
On February 6, 2026, the Brazilian Senate received Bill 160/2026, proposing a temporary, non-exclusive compulsory license for tirzepatide under Article 71 of the Industrial Property Law, contingent on a technical finding by Brazil’s health regulator, ANVISA, of insufficient supply, excessive pricing, or public-health impact.[15] Three days later, the Chamber of Deputies approved an urgency procedure for a related bill, 68/2026, declaring tirzepatide-based medicines a matter of public interest, by a vote of 337 to 19.[16][17] A third bill, 157/2026, would broaden compulsory licensing authority to oncology drugs generally.[15]
Why Mounjaro Is a Different Kind of Case
Every earlier case in this article involves an infectious disease or cancer drug with a defined, medically necessary patient population. Tirzepatide’s Brazilian approval covers both type 2 diabetes and, since June 9, 2025, chronic weight management, putting a GLP-1 class obesity drug at the center of a compulsory-license fight for the first time.[15] Brazil’s Health Minister, Alexandre Padilha, has stated publicly that the government follows World Health Organization guidance and does not currently support a compulsory license for GLP-1 pens, preferring ANVISA’s priority-review track for competing GLP-1 dossiers and eventual patent expiry as the access mechanism.[15]
Where the Process Actually Stands
As of an April 2026 legal analysis, Bill 160/2026 remained at an early procedural stage, still requiring committee review and a Senate floor vote before moving to the Chamber of Deputies and presidential sanction, and any resulting compulsory license would require a separate Executive Branch act under the safeguards Brazil added to its patent law during the COVID-19 pandemic.[15] Eli Lilly has publicly opposed the effort.[16] The bills are not law. But the single-day, 337-to-19 urgency vote is the fact that should reset how quickly a pricing team assumes legislative risk can move: a floor vote authorizing fast-track consideration happened in less time than most companies take to route an internal pricing memo.
The Paragraph 6 System’s Only Completed Export: Canada and Rwanda
Article 31bis exists specifically so countries without pharmaceutical manufacturing capacity can still benefit from someone else’s compulsory license. In the eighteen years since the mechanism was created, exactly one WTO member has completed an export transaction under it.
The Timeline
Apotex, a Canadian generic manufacturer, developed a fixed-dose combination of zidovudine, lamivudine, and nevirapine, branded Apo-TriAvir, and received Health Canada approval in August 2006, but had no importing country lined up to trigger the process.[20] Rwanda notified the WTO on July 17, 2007 of its intent to import 260,000 packs of the combination drug.[18][19] Apotex then sought voluntary licenses from the three patent holders, GlaxoSmithKline, Boehringer Ingelheim, and Shire BioChem, and when those negotiations stalled, Canada’s patent commissioner granted a compulsory license covering nine patents on September 19, 2007.[18][19] Rwanda then ran its own international tender, which Apotex won in May 2008 at a bid price of 19.5 cents per tablet, priced at cost.[20] The first shipment reached Rwanda in September 2008, roughly fourteen months after Rwanda’s initial WTO notification and nearly two years after Health Canada’s original approval.[20]
Why the System Has Not Been Used Again
Apotex’s own president at the time said the company invested millions in research, development, and legal costs, made no profit, and did it because it was the right thing to do, while calling the process excruciating and painful in its current form.[21] A researcher who reviewed the case for Canada’s own access-to-medicines regime concluded the mechanism was too far removed from developing-country procurement realities to function as a general-purpose tool, noting Rwanda could likely have obtained a similar drug faster and cheaper directly from an Indian manufacturer.[22] The barrier is not legal uncertainty. It is transaction cost: nine patents, three separate patent holders, roughly two months of negotiation before the compulsory license was even granted, and eight more months of tendering after that.
The Voluntary Licensing Alternative, and Its Blind Spot
The Medicines Patent Pool, founded in 2010 with United Nations backing, is the industry’s preferred alternative to compulsory licensing: originator companies license a patent to the Pool on a voluntary basis, and the Pool grants non-exclusive, geographically limited sublicenses to generic manufacturers.[33] Over 2020 to 2025, the Pool in-licensed 44 products from 22 originator companies and supplied 62.14 billion doses of treatment.[33] A 2017 cost-benefit study of the Pool’s HIV licenses estimated 2.3 billion dollars in direct savings through 2028, a benefit-to-cost ratio of roughly 43 to 1.[36]
The Geographic Exclusion Pattern
Voluntary licenses routinely exclude upper-middle-income countries with real disease burdens and existing generic manufacturing capacity. ViiV’s dolutegravir license excluded Colombia until the country’s own compulsory license forced the issue.[25] Pfizer’s 2021 voluntary license with the Pool for its COVID-19 antiviral excluded Argentina, Brazil, China, Malaysia, and Thailand from the covered territory even though all five have established generic industries.[34] Chinese generic manufacturers have signed sublicense agreements with the Pool to export tenofovir alafenamide, dolutegravir, lopinavir/ritonavir, and atazanavir to other countries, while China itself remains outside the license territory for all four products.[35]
Why This Matters for Pricing Strategy
This exclusion pattern is not incidental, and it is the direct cause of several episodes in the timeline above. Colombia moved to a compulsory license precisely because ViiV’s voluntary license did not cover it.[25] The countries most likely to invoke Article 31 are frequently the same countries a tiered-pricing or voluntary-license strategy deliberately carved out, because they were judged too large, too commercially valuable, or too capable of generic production to include on preferential terms. A voluntary-licensing footprint that excludes a country on commercial grounds should be read internally as a compulsory-licensing risk flag on that same country, not as a separate, unrelated decision.
TRIPS-Plus Provisions: The Contractual Layer on Top of the Treaty Layer
TRIPS sets a floor, not a ceiling. Many bilateral and regional free trade agreements, most involving the United States or the European Union, contain TRIPS-plus provisions that go beyond the treaty’s minimum standards and can narrow the practical value of a compulsory license even where one is legally available.[37][38]
Data Exclusivity
Data exclusivity gives an originator company a fixed period during which a drug regulator cannot rely on the originator’s clinical trial data to approve a generic, independent of whether a patent exists at all.[37] The United States failed to get data exclusivity written into TRIPS itself in 1994 and instead pursued it bilaterally, embedding the requirement in a long list of subsequent free trade agreements.[37] The effect on this article’s subject is direct: if a country grants a compulsory license but a data-exclusivity term is still running, a generic manufacturer holding a valid license may still be unable to secure marketing authorization until the exclusivity period lapses.[38]
Patent Linkage and Enforcement Provisions
A related TRIPS-plus category requires a drug regulator to check a patent registry before approving a generic, or grants customs officials authority to seize shipments on suspicion of infringement, both of which raise the practical cost of relying on a compulsory license even after it has been legally granted.[38] A 2022 systematic review of the empirical literature on intellectual property and medicine access concluded that stronger, TRIPS-plus intellectual property protection is generally associated with higher drug prices, delayed generic availability, and higher costs to governments and patients, while TRIPS flexibilities can facilitate access, though their use to date remains limited relative to their legal availability.[39]
What Determines Whether a Country Actually Uses These Flexibilities
The cases above sort into four recognizable patterns, an original classification built from the documented episodes in this article rather than an established legal taxonomy.
Type One: Reactive HIV-Era Licensing
Thailand, Brazil, Ecuador, and Canada-Rwanda all fall into a single wave concentrated between 2006 and 2010, driven almost entirely by antiretroviral pricing during the peak of the global AIDS response.[4][8][13][20][27] This wave has largely subsided as first-line HIV regimens moved into the Medicines Patent Pool’s voluntary-license structure.
Type Two: Targeted Disease-Burden Licensing
India’s Nexavar case and Malaysia’s 2017 sofosbuvir license target a single high-cost drug for a specific, well-documented disease burden, cancer and hepatitis C respectively, using the ordinary compulsory-licensing statute rather than an emergency declaration.[5][29] These cases tend to move more slowly and survive legal challenge because they rest on a documented record of undersupply.
Type Three: Access-Gap Licensing Tied to a Voluntary-License Exclusion
Colombia’s dolutegravir case is a distinct pattern: the trigger was not the drug’s price in isolation, but a documented population, Venezuelan migrants, excluded from an existing voluntary-license framework that covered the same drug everywhere else.[25] Watch for this pattern wherever a voluntary license’s territorial map does not match a country’s actual patient population, including migration flows and regional treatment-access programs.
Type Four: High-Revenue, Non-Communicable-Disease Licensing
Brazil’s 2026 Mounjaro bills are the first documented attempt to apply this mechanism to a blockbuster metabolic-disease drug rather than an infectious disease or a lower-volume cancer treatment.[15] If this pattern spreads, the relevant risk signal shifts from disease burden and mortality data toward simple revenue exposure and out-of-pocket cost as a share of national income, a different set of inputs than the ones that predicted the first three waves.
South Africa’s Pending Reform Shows Where This Is Headed Next
South Africa currently grants patents through a pure deposit system: if the paperwork is filed and the fees paid, the patent is granted, with no substantive examination of novelty or inventive step.[41] The country has historically granted more pharmaceutical patents than the United States or the European Patent Office, a gap long cited by South Africa’s Fix the Patent Laws campaign as the mechanism behind extended evergreening of patents that would not survive examination elsewhere.[41]
The Coming Patents Bill
A new Patents Bill, previewed at a Department of Trade, Industry and Competition stakeholder workshop in September 2025 and expected to reach parliament in early 2026, would introduce a phased substantive search and examination system, a pre-grant third-party observation process, post-grant opposition, and a new administrative tribunal to handle compulsory licenses under Article 31bis.[40] As of July 2026, patient advocacy groups were still publicly pressing the responsible minister to formally publish the bill, indicating the reform remains in preparation rather than in force.[41]
Why This Belongs in a Pricing Risk Framework
A dedicated compulsory-licensing tribunal is a structural change, not an isolated policy event. Once a country builds standing administrative infrastructure for compulsory licensing, the marginal cost of issuing the next license drops sharply compared with the ad hoc ministerial and presidential decrees that characterized the 2006 to 2012 wave. South Africa’s reform is worth tracking specifically because it would convert compulsory licensing from an exceptional political act into a routine administrative procedure.
Building an Early-Warning Framework for Emerging-Market Pricing Risk
The cases above share four observable precursors that, taken together, form an original heuristic for flagging pricing risk before a formal declaration. This is a screening tool built from the documented pattern in this article, not a validated predictive model, and it should be paired with local legal counsel and primary-source patent-family monitoring, including tools like DrugPatentWatch’s jurisdiction-level patent tracking, rather than used alone.
Signal One: A Public Declaration of Interest or Disease-Burden Statement
Colombia’s Ministry of Health issued a formal Declaration of Public Interest seven months before the compulsory license itself.[23] Brazil’s 2007 decree followed a public statement of intent from the Ministry of Health weeks earlier.[13] A government health ministry publicly citing a specific drug’s price against a specific patient population is historically the first visible step, not a side comment.
Signal Two: Legislative Bills Citing the National Compulsory-Licensing Statute by Number
Brazil’s three 2026 bills each cite Article 71 of the Industrial Property Law explicitly.[15] A bill that names the specific statutory compulsory-licensing provision, rather than proposing new legislation from scratch, signals its drafters intend to use the fastest available legal path.
Signal Three: A Documented Gap Between a Voluntary License’s Territory and a Country’s Patient Population
This is the Colombia pattern. Any country excluded from a company’s own voluntary-license map, particularly one with documented demand for the same molecule, carries elevated structural risk regardless of anything else happening in that market.[25]
Signal Four: The Country’s TRIPS-Plus Exposure Under Its Own Trade Agreements
A country with a US or EU free trade agreement carrying data-exclusivity or patent-linkage obligations faces a materially different post-license outcome than a country without one, because a compulsory license alone may not clear the regulatory approval pathway for the generic competitor.[37][38] Mapping which emerging markets carry TRIPS-plus data-exclusivity terms is a prerequisite for judging how much a compulsory license would actually cost a brand in a given market, since the answer is not the same in every jurisdiction.
What This Means for Brand Manufacturers
The documented record points to three practical takeaways. First, price and supply data submitted to local regulators or health ministries becomes evidentiary material in exactly the kind of proceeding described in the India and Colombia cases, so patient-access reporting should be treated as a document that could be read back in an administrative hearing, not just an internal metric.[5][25] Second, a voluntary-license territory map that excludes a country on purely commercial grounds should be flagged as compulsory-licensing risk on that country specifically, given how directly the Colombia case traces to exactly that gap.[25] Third, the transaction-cost lesson from Canada-Rwanda cuts both ways: the same friction that has kept Article 31bis exports rare for two decades also means that once a country builds standing administrative infrastructure, as South Africa is doing, that friction disappears for future cases in that jurisdiction.[20][41]
What This Means for Generic and Biosimilar Manufacturers
For a generic or biosimilar manufacturer evaluating a compulsory license as an entry strategy, the documented cases suggest the legal grant is rarely the binding constraint; negotiation timelines, patent-holder litigation, and downstream regulatory approval are. Natco’s Nexavar license survived three years of appeals before becoming final.[7] Apotex’s Rwanda export took fourteen months from WTO notification to first shipment despite Health Canada approval already in hand.[20] Any generic manufacturer building a compulsory-license strategy into a market-entry plan should model the full sequence, negotiation, license grant, appeal, and regulatory approval, rather than treating the license grant date as the effective market-entry date.
What Happens Next: Reading the WTO’s Stalled Waiver as a Signal
The clearest evidence that multilateral consensus on this topic has stalled, even as national-level activity continues, is the fate of the COVID-19 TRIPS waiver’s planned extension. The June 2022 Ministerial Decision on the TRIPS Agreement gave members six months, until December 17, 2022, to decide whether to extend the vaccine waiver to diagnostics and therapeutics.[2] That deadline was postponed repeatedly through 2023, and at the Thirteenth Ministerial Conference in Abu Dhabi in February 2024, members formally acknowledged that consensus could not be reached, closing the question without an extension.[42][43] The World Health Organization had already ended the COVID-19 public health emergency in May 2023, which several observers pointed to as removing the urgency that might otherwise have forced a decision.[43]
The pattern this leaves behind matters for a pricing team: multilateral TRIPS reform, even for an emergency with global political will behind it in 2022, moved slower than any single national compulsory-licensing case in this article’s timeline. National-level action, as Colombia, Brazil, and South Africa each show in their own way, has become the more active front, not the WTO Ministerial Conference.[15][23][40]
Methodology
This article draws on WTO primary legal texts and fact sheets, peer-reviewed database analyses, national government decrees and legislative bills, and contemporaneous reporting from public-health organizations, generic-industry sources, and the affected patent holders themselves, covering pharmaceutical compulsory-licensing and government-use episodes from 1995 through August 2026. A case was included in the comparative table only where a formal government act, decree, or legislative vote is publicly documented; proposals withdrawn before any formal act, or that remain purely rhetorical, are described in the narrative sections but excluded from the table and from the average price-reduction calculation. The average price-reduction figure is a simple, unweighted mean of six cases with disclosed before-and-after pricing and should not be read as representative of undisclosed or unresolved cases. Bills still pending as of publication, including Brazil’s 2026 tirzepatide legislation and South Africa’s Patents Bill, are described as pending, not enacted law, and their outcomes may change after this article’s publication date.
Key Takeaways
Article 31 of TRIPS lets any WTO member override a pharmaceutical patent without the holder’s consent, and the 2001 Doha Declaration confirmed governments define for themselves what counts as an emergency justifying it.[1]
Only 72 of 187 patent-holding countries have implemented the Article 31bis cross-border export mechanism, even though 176 have some form of compulsory licensing provision.[3]
A 2012 academic database found 24 verified compulsory-licensing episodes in 17 countries between 1995 and 2011, with 12 resulting in an actual license.[4]
Canada’s 2007 to 2008 export of an HIV combination drug to Rwanda remains the only completed Article 31bis transaction in the mechanism’s eighteen-year history.[18][19][20]
Across six documented cases with disclosed pricing, from India’s 2012 Nexavar license to Colombia’s 2024 dolutegravir license, the average documented price reduction was approximately 74 percent.
Colombia’s 2023 to 2024 dolutegravir case traces directly to a country excluded from the originator’s own voluntary-license territory, a pattern worth screening for in any emerging-market license map.[25]
Brazil’s 2026 tirzepatide bills mark the first documented attempt to apply compulsory licensing to a blockbuster metabolic-disease drug rather than an infectious disease or lower-volume cancer treatment, and remain unresolved as of this article’s publication.[15]
The WTO’s own COVID-19 waiver extension to therapeutics and diagnostics failed to reach consensus at the 2024 Ministerial Conference, showing multilateral reform has moved slower than individual national cases.[42][43]
FAQ
What is the difference between a compulsory license and a government use license? Both authorize use of a patent without the holder’s consent under Article 31 of TRIPS. Government use typically refers to use by or for the state itself, such as Thailand’s 2006 to 2007 licenses issued directly through its Ministry of Public Health, while a compulsory license more often authorizes a third-party manufacturer, as in India’s 2012 grant to Natco Pharma.[6][8]
Does the WTO have to approve a country’s compulsory license? No. A WTO member can issue a compulsory license unilaterally under its own national law. The WTO’s role is limited to the notification requirements under Article 31bis when the license involves cross-border export, as in the Canada-Rwanda case, where Canada notified the TRIPS Council after issuing its license rather than seeking prior approval.[2][18]
Can a company challenge a compulsory license in court? Yes, through the issuing country’s own domestic legal system. Bayer challenged its Nexavar compulsory license through India’s Intellectual Property Appellate Board, the Bombay High Court, and the Supreme Court of India, losing at each stage over roughly three years.[7]
Why did the WTO’s COVID-19 waiver never get extended to therapeutics and diagnostics? Members disagreed on scope and evidence for over a year, commissioned a US International Trade Commission study, and ultimately reported at the February 2024 Ministerial Conference that consensus could not be reached, closing the extension question without action.[42][43]
Does a compulsory license eliminate a patent? No. The patent stays valid and the holder is entitled to remuneration, typically a royalty. India’s Nexavar license set royalties at 6 percent, later 7 percent on appeal, and Brazil’s efavirenz decree set a 1.5 percent royalty on the price paid by its Ministry of Health.[7][14]
What royalty rate do patent holders typically receive under a compulsory license? Rates vary by country and case. Documented examples range from 1.5 percent in Brazil’s 2007 efavirenz decree to 6 to 7 percent in India’s Nexavar license, with the Indian rate described at the time as sitting near the high end of the UNDP’s 2001 royalty guidelines for pharmaceutical compulsory licenses.[6][14]
Can a voluntary license prevent a country from issuing a compulsory license? Not automatically, and the Colombia case shows the opposite dynamic: ViiV’s existing voluntary license with the Medicines Patent Pool excluded Colombia from its territory, which was itself the documented rationale Colombia’s government cited for pursuing a compulsory license instead.[25]
Is Brazil’s 2026 Mounjaro bill a compulsory license? Not yet. As of this article’s publication, Brazil’s tirzepatide bills are legislative proposals that would authorize Brazil’s Executive Branch to issue a compulsory license if further statutory conditions are met. No license has been granted.[15]
Which countries have never implemented Article 31bis, and why does it matter? A 2023 legislative survey found 115 of 187 patent-holding countries had not implemented the Article 31bis export mechanism, even though most have some other form of compulsory licensing provision.[3] That gap matters most for countries with no domestic manufacturing capacity, since Article 31bis is specifically the mechanism that lets them import a compulsory-licensed generic rather than manufacture one themselves.
How does data exclusivity affect a compulsory license? Data exclusivity operates independently of patent status. A generic manufacturer holding a valid compulsory license may still be unable to obtain regulatory marketing approval if the country’s data-exclusivity term on the originator’s clinical trial data has not yet expired, a gap that exists in countries whose free trade agreements include TRIPS-plus data-exclusivity obligations.[37][38]
References
World Trade Organization. (n.d.). TRIPS and public health: Compulsory licensing of pharmaceuticals and TRIPS. https://www.wto.org/english/tratop_e/trips_e/public_health_faq_e.htm
World Trade Organization. (n.d.). TRIPS and public health. https://www.wto.org/english/tratop_e/trips_e/pharmpatent_e.htm
Trade-Related Aspects of Intellectual Property Rights Flexibilities and Public Health: Implementation of Compulsory Licensing Provisions into National Patent Legislation. (2023). PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC10726804/
Beall, R., & Kuhn, R. (2012). Trends in Compulsory Licensing of Pharmaceuticals Since the Doha Declaration: A Database Analysis. PLOS Medicine. https://journals.plos.org/plosmedicine/article?id=10.1371%2Fjournal.pmed.1001154
Cancer Today. (2012). India Overrides Patent on Cancer Drug. https://www.cancertodaymag.org/summer2012/india-overrides-patent-for-kidney-and-liver-cancer-drug/
MSF Access Campaign. (2024). Background Information on India’s First Compulsory Licence. https://msfaccess.org/background-information-indias-first-compulsory-licence
PharmaTimes. (2014). Bayer loses bid to overturn India’s first compulsory licence. http://www.pharmatimes.com/news/bayer_loses_bid_to_overturn_indias_first_compulsory_licence_1002934
Compulsory Licences: Law and Practice in Thailand. (n.d.). PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC7122632/
aidsmap. (2007). Abbott offers price cut to thwart Thai compulsory license on Kaletra. https://www.aidsmap.com/news/feb-2007/abbott-offers-price-cut-thwart-thai-compulsory-license-kaletra
Make Medicines Affordable. (n.d.). Chronology of access to medicine campaign in Thailand. https://makemedicinesaffordable.org/chronology-of-access-to-medicine-campaign-in-thailand/
Make Medicines Affordable. (n.d.). The campaign for use of compulsory licensing in Thailand. https://makemedicinesaffordable.org/the-campaign-for-use-of-compulsory-licensing-in-thailand/
infojustice. (2009). Timeline for US-Thailand Compulsory License Dispute, Version 3. https://infojustice.org/wp-content/uploads/2012/11/pijip-thailand-timeline.pdf
aidsmap. (2007). Brazil issues compulsory license on efavirenz. https://www.aidsmap.com/news/may-2007/brazil-issues-compulsory-license-efavirenz
Nationalizing Efavirenz: Compulsory Licence, Collective Invention and Neo-Developmentalism in Brazil. (n.d.). SpringerLink. https://link.springer.com/chapter/10.1007/978-3-319-76834-2_3
Lexology. (2026). Compulsory Licensing in Brazil (2026): The Mounjaro Case. https://www.lexology.com/library/detail.aspx?g=fb9c0cfa-f32f-410d-93aa-fab6408cbedb
Citeline / Pink Sheet. (2026). Lilly Slams Brazil’s Pursuit To Make Off-Patent Mounjaro And Zepbound. https://insights.citeline.com/pink-sheet/legalandip/patents/lilly-slams-brazils-pursuit-to-make-off-patent-mounjaro-zepbound-QBNPU3NF4ZFI7GDD4C6DLSCKKU/
Tavares. (2026). Brazil’s Congress fast-tracks bill on compulsory licensing of Mounjaro patents. https://www.tavaresoffice.com.br/brazils-congress-fast-tracks-bill-on-compulsory-licensing-of-mounjaro-patents/
Knowledge Ecology International. (n.d.). Canada’s Intervention to TRIPS Council: Experience using the System (Apotex-Rwanda Case). https://www.keionline.org/21413
American Society of International Law. (2007). Canadian-made Drugs for Rwanda: The First Application of the WTO Waiver on Patents and Medicines. https://www.asil.org/insights/volume/11/issue/28/canadian-made-drugs-rwanda-first-application-wto-waiver-patents-and
Canada’s Access to Medicines Regime: Promise or Failure of Humanitarian Effort? (n.d.). PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC2831732/
Asia IP. (2018). Let’s Talk Compulsory Licensing, Not Murder. https://asiaiplaw.com/index.php/sector/licensing-franchising/letrsquos-talk-compulsory-licensing-not-murder
A one-time-only combination: Emergency medicine exports under Canada’s access to medicines regime. (2013). HHR Journal. https://www.hhrjournal.org/2013/08/27/a-one-time-only-combination-emergency-medicine-exports-under-canadas-access-to-medicines-regime/
Public Citizen. (2024). Colombia Moves Forward with Compulsory License for HIV Medicine Dolutegravir. https://www.citizen.org/news/colombia-takes-significant-next-step-to-expand-peoples-access-to-affordable-hiv-treatment-moves-forward-with-compulsory-license-for-hiv-medicine-dolutegravir/
UNAIDS. (2023). UNAIDS welcomes new decision in Colombia allowing more affordable access to quality HIV medicines. https://www.unaids.org/en/resources/presscentre/pressreleaseandstatementarchive/2023/october/20231004_colombia
Health Policy Watch. (2024). Colombia Issues Compulsory License To Enable It To Access Generic HIV Drug, Dolutegravir. https://healthpolicy-watch.news/colombia-issues-compulsory-license-to-get-key-generic-hiv-medicine-dolutegravir/
ViiV Healthcare. (2024). ViiV Healthcare statement on compulsory licence for dolutegravir in Colombia. https://viivhealthcare.com/hiv-news-and-media/news/company-statements/viiv-healthcare-statement-on-compulsory-licence-for-dolutegravir-in-colombia/
Intellectual Property Watch. (2010). Ecuador Grants First Compulsory Licence, For HIV/AIDS Drug. https://www.ip-watch.org/2010/04/22/ecuador-grants-first-compulsory-licence-for-hivaids-drug/
Public Citizen. (2010). By Authorizing Generic Competition, Ecuador Cuts Cost of Key HIV/AIDS Drug. https://www.citizen.org/news/by-authorizing-generic-competition-ecuador-cuts-cost-of-key-hiv-aids-drug/
CodeBlue. (2019). Study: Voluntary Licences For Hepatitis C Medicines Boost Treatment Access. https://codeblue.galencentre.org/2019/09/study-voluntary-licences-for-hepatitis-c-medicines-boost-treatment-access/
FB Rice. (2021). Indonesia updates patent compulsory licensing provisions. https://www.fbrice.com.au/ip-news-insights/indonesia-updates-patent-compulsory-licensing-provisions/
Make Medicines Affordable. (n.d.). Indonesia Issues Government Use Licenses for Remdesivir and Favipiravir. https://makemedicinesaffordable.org/indonesia-issues-government-use-licenses-for-remdesivir-and-favipiravir/
Public Citizen. (n.d.). Indonesia Licenses Patents for Seven HIV & Hepatitis B Medicines. https://www.citizen.org/wp-content/uploads/indonesia_licenses_patents_for_seven_hiv.pdf
The Conversation. (2026). The Medicines Patent Pool is improving access to essential medication where it’s most needed, but does it make economic sense? https://theconversation.com/the-medicines-patent-pool-is-improving-access-to-essential-medication-where-its-most-needed-but-does-it-make-economic-sense-286914
Health Policy Watch. (2021). Pfizer And Medicines Patent Pool Reach ‘Ground-breaking’ Voluntary Licensing Deal For New COVID-19 Treatment Pill. https://healthpolicy-watch.news/pfizer-and-medicines-patent-pool/
MSF Access Campaign. (2020). Voluntary Licenses and Access to Medicines. https://msfaccess.org/sites/default/files/2020-10/IP_VoluntaryLicenses_full-brief_Oct2020_ENG.pdf
Juneja, S., Gupta, A., Moon, S., & Resch, S. (2017). Projected savings through public health voluntary licences of HIV drugs negotiated by the Medicines Patent Pool (MPP). PLOS ONE. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5444652/
South Centre. (2020). Evolution of Data Exclusivity for Pharmaceuticals in Free Trade Agreements. Policy Brief No. 76. https://www.southcentre.int/wp-content/uploads/2020/04/PB-76.pdf
MSF Access Campaign. (n.d.). TRIPS Agreement of Intellectual Property. https://msfaccess.org/spotlight-trips-trips-plus-and-doha
What is the impact of intellectual property rules on access to medicines? A systematic review. (n.d.). PMC. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9013034/
Adams & Adams. (2025). South Africa Prepares Major Patent and Design Law Reforms. https://www.adams.africa/adams-news/south-africa-prepares-major-patent-and-design-law-reforms/
Treatment Action Campaign. (2026). No more delays, Minister Tau publish the patent law reform bill now! https://www.tac.org.za/news/no-more-delays-minister-tau-publish-the-patent-law-reform-bill-now/
World Trade Organization. (2024). TRIPS Council finalizes preparations for MC13. https://www.wto.org/english/news_e/news24_e/trip_13feb24_e.htm
From proposal to compromise: the TRIPS waiver debate and the crisis of WTO decision-making. (n.d.). PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC12821828/
