Can Patents Deter Innovation? The Anticommons in Biomedical Research (Heller & Eisenberg, 1998)

Summary

Michael A. Heller (then University of Michigan Law, later Columbia Law School) and Rebecca S. Eisenberg (University of Michigan Law) argue that the privatization of upstream biomedical research — genes, gene fragments, receptors, research tools — can produce the opposite failure from the one usually feared. Where Garrett Hardin's "tragedy of the commons" describes overuse when too many people have a privilege to use a resource and no one can exclude, Heller and Eisenberg describe underuse when too many people have a right to exclude and no one has an effective privilege of use — a "tragedy of the anticommons" (p. 698). Applied to biomedical patents, a proliferation of upstream rights held by separate owners means a firm needing to combine several patented inputs to build one product may face a bargaining challenge so costly that the product is never brought to market, or is developed on the basis of incomplete information. The paper's own conclusion states the paradox directly: "more upstream rights may lead paradoxically to fewer useful products for improving human health" (p. 701).

The Core Argument and Findings

The definition (p. 698). The authors set the anticommons explicitly against its mirror image: "A resource is prone to overuse in a tragedy of the commons when too many owners each have a privilege to use a given resource and no one has a right to exclude another." By contrast, "a resource is prone to underuse in a 'tragedy of the anticommons' when multiple owners each have a right to exclude others from a scarce resource and no one has an effective privilege of use." They state plainly that "anticommons property can best be understood as the mirror image of commons property," and illustrate the general phenomenon with post-Soviet privatization: "many privatized storefronts remained empty, while flimsy metal kiosks, stocked full of goods, mushroomed on the streets," because fragmented rights had been distributed among various socialist-era stakeholders and "no one could set up shop without first collecting rights from each of the other owners" (p. 698). Applied to biomedical research, the problem is distinct from the ordinary cost of a patent monopoly: "the tragedy of the anticommons refers to the more complex obstacles that arise when a user needs access to multiple patented inputs to create a single useful product," so that "each upstream patent allows its owner to set up another tollbooth on the road to product development, adding to the cost and slowing the pace of downstream biomedical innovation" (p. 699).

Two mechanisms (p. 699). The authors identify two distinct routes by which an anticommons can arise: "either by creating too many concurrent fragments of intellectual property rights in potential future products or by permitting too many upstream patent owners to stack licenses on top of the future discoveries of downstream users."

• Concurrent fragments. Patents held by separate owners on individual gene fragments, or on individual members of a receptor family that a firm must screen against as a set, force a single downstream developer to assemble many complementary licenses before it has an effective right to proceed. The paper reports that "a recent search of the Lexis patent database disclosed more than 100 issued U.S. patents with the term 'adrenergic receptor' in the claim language" (p. 699), and that faced with this bargaining challenge, "firms choose between diverting resources to less promising projects with fewer licensing obstacles or proceeding to animal and then clinical testing on the basis of incomplete information" (p. 699).
• Stacking licenses. Reach-through license agreements (RTLAs) on upstream research tools give "the owner of a patented invention, used in upstream stages of research, rights in subsequent downstream discoveries" (p. 699) — a royalty, an exclusive or nonexclusive license, or an option on whatever the tool is later used to discover. The paper's examples are Cetus's initially resisted attempt to attach RTLAs to PCR, and DuPont's licensing terms for oncomouse and cre-lox technology, under which the license terms "permit DuPont to leverage its proprietary position in upstream research tools into a broad veto right over downstream research and product development" (p. 700).

Why private bargaining may not fix it (pp. 700–701). The authors allow that repeat-dealing communities sometimes build institutions — patent pools in the automobile, aircraft, and synthetic-rubber industries, copyright collectives in music — that reduce the transaction costs of bundling rights, and hedge that "patent barriers to product development may be a transitional phenomenon rather than an enduring tragedy" (p. 700). But they give three specific reasons a biomedical anticommons may be harder to dissolve than others:

1. Transaction costs. Many upstream owners are public institutions "with limited resources for absorbing transaction costs," the rights cover a heterogeneous mix of techniques, reagents, and sequences that are hard to value comparably, and bargaining costs tend to fall early in a project, before its value is known (p. 700). Antitrust risk compounds this: a 1975 consent decree dismantled the aircraft-manufacturers' patent pool, and "the antitrust climate changes from one administration to the next" (p. 700).
2. Heterogeneous interests of rights holders. Public and private owners pursue different goals — the paper's example is NIH's dispute over co-ownership of AZT patent rights, sought to keep prices low rather than to maximize royalty income — so that "when owners have conflicting goals and each can deploy its rights to block the strategies of the others, they may not be able to reach an agreement" (p. 700).
3. Cognitive biases. A worked example: of 50 upstream patents, one (unidentified in advance) will be the key to a discovery a downstream developer will pay $10 million for; each owner's rational ex ante asking price is $200,000, but each irrationally overvalues its own contribution, so "the upstream owners collectively will demand more than the aggregate market value of their inputs" and the deal fails (p. 701).

The authors' own use of "holdout" (p. 700). Discussing why a biomedical anticommons may be more persistent than other patent anticommons, Heller and Eisenberg write that "the lack of substitutes for certain biomedical discoveries (such as patented genes or receptors) may increase the leverage of some patent holders, thereby aggravating holdout problems." This is the authors' own word for the concurrent-fragments dynamic, not a label supplied by later commentary.

Remedies discussed (pp. 700–701). The paper points to patent pools and copyright collectives as institutional responses that have "evolved to facilitate licensing transactions" in other industries, while flagging antitrust exposure as a real deterrent to new pools. Its policy recommendation, stated as the paper's conclusion, is: "policy-makers should seek to ensure coherent boundaries of upstream patents and to minimize restrictive licensing practices that interfere with downstream product development. Otherwise, more upstream rights may lead paradoxically to fewer useful products for improving human health" (p. 701).

Relation to the Georgist Case

Heller and Eisenberg's own framing already supplies the structural link: the anticommons is, in their words, "the mirror image of commons property" (p. 698) — where the commons problem is too many privileges of use and no right to exclude (overuse), the anticommons is too many rights to exclude and no privilege of use (underuse). Read as a general structure rather than a biomedical-specific one, the parallel to Georgist rent theory follows: a scarce resource generates a return to whoever can restrict access to it, and splitting the right to exclude among several parties does not eliminate that return — it multiplies the points at which a toll can be charged. Each rights-holder can extract value simply from occupying a position with the power to say no, unrelated to any contribution to the resource's value. That is the same rent logic Georgists apply to location: it is the power to exclude, not any service rendered, that lets a payment be collected.

The two mechanisms the paper identifies map onto two structures already documented on this wiki, though the authors draw neither comparison and the mapping below is offered as an interpretive reading, not a claim either author makes. The concurrent fragments mechanism — many separate owners of complementary, non-substitutable rights who must all agree before a single product can be built — is structurally the same bargaining failure as the holdout problem documented there for land assembly: many blocking rights-holders needed for one output, each with leverage disproportionate to their own share's worth. Heller and Eisenberg use the word "holdout" themselves for this dynamic in patents (p. 700), which is a direct textual link rather than an inference. The stacking licenses mechanism — an upstream owner retaining a reach-through claim on a downstream discovery not yet made — is closer to the contract-design logic on Intellectual-Property Rents: a rights-holder structures its claim to capture value from a future, as-yet-undefined product, the same "veto right over downstream research" (p. 700) that page's discussion of reach-through royalty contracting addresses from the licensor's side.

None of this is the paper's own subject. Heller and Eisenberg write about biomedical patents and say nothing about land, location, or land value taxation; the extension of "fragmented exclusion rights over a scarce input let each holder extract a toll unrelated to any contribution" to location specifically is drawn here as interpretation, applying a structure the authors describe for patents to a different kind of scarce resource.

Nuances and Limits

The paper is explicitly hedged rather than categorical about how persistent any given anticommons will prove — the authors describe possible institutional fixes and call the tragedy potentially "transitional" (p. 700) in the same breath as listing three structural reasons (transaction costs, heterogeneous interests, cognitive biases) that the biomedical case specifically may resist those fixes (pp. 700–701). The paper is also about a government-created legal structure (the allocation of patent rights), not about land tenure, and the underlying scarcity is different in kind: a patent's scarcity is legally constructed and could in principle be redesigned by policy (narrower claims, coherent boundaries, non-exclusive licensing norms), whereas locational scarcity is physical. Applying the paper's logic to land is an extension beyond its scope, not a restatement of it.

The paper credits its underlying theory of anticommons property to Heller's companion article, "The Tragedy of the Anticommons: Property in the Transition from Marx to Markets," Harvard Law Review 111(3): 621–688 (1998) (their footnote 3), and restates that article's central illustration — post-socialist Moscow storefronts left empty despite kiosks in front of them "stocked full of goods" (p. 698) — as part of its own introduction. The Harvard Law Review article itself has not been read in full here; what is stated about it beyond the Science paper's own text rests on that article's published abstract and on Heller's later self-summary of the same example, and it is cited only as the origin of the term and the theory, graded accordingly in Sources below.

Bears On

• Holdout Problem (Land Assembly) — the concurrent-fragments mechanism is the same many-blocking-rights-holders bargaining failure documented there for land assembly, and the authors name it "holdout" themselves (p. 700)
• Intellectual-Property Rents — the stacking-licenses/reach-through-license mechanism is the originating case for the licensing-as-rent-protection dynamic that page documents
• Land Monopoly — the structural mirror image: rent extracted through fragmentation and blocking rather than through concentration and exclusion

See Also

• Holdout Problem (Land Assembly)
• Intellectual-Property Rents
• Land Monopoly
• Economic Rent
• Quasi-Rent

Sources

1. Michael A. Heller & Rebecca S. Eisenberg, "Can Patents Deter Innovation? The Anticommons in Biomedical Research," Science 280(5364): 698–701 (1 May 1998). DOI: 10.1126/science.280.5364.698. Read in full (all four pages and footnotes) — used for the definition of the anticommons and its mirror-image relation to the commons (p. 698), the Moscow storefronts illustration (p. 698), the two mechanisms of concurrent fragments and stacking licenses/reach-through license agreements with their worked examples (p. 699), the DuPont/Cetus licensing examples and the authors' own use of "holdout" (p. 700), the transaction-cost, heterogeneous-interests, and cognitive-bias arguments and the patent-pool/copyright-collective remedies (pp. 700–701), and the conclusion (p. 701). A-claim throughout — peer-reviewed Science article, read in full.
2. Michael Heller, "The Tragedy of the Anticommons: Property in the Transition from Marx to Markets," Harvard Law Review 111(3): 621–688 (1998) — cited by Heller & Eisenberg (1998, footnote 3) as the origin of the anticommons concept. Full text not accessible at last review (2026-09-25); used only for the origin of the term and the Moscow storefronts example, sourced to the article's own published abstract and to Heller's later self-summary of the same argument and example (Michael Heller, "The UNE Anticommons," Manhattan Institute conference paper, 2004, citing back to "Heller, supra note *, at 621–26"). C-claim — abstract and the author's later summary only; full text not consulted.