
Michael Heller on Gridlock and the Tragedy of the Anticommons 11/02/2009
What this covers
Michael Heller, a law professor and economist, joins Russ Roberts on EconTalk to develop his argument about ownership fragmentation as an obstacle to innovation and wealth creation. Rather than restating his central claim about the tragedy of the anticommons, Heller ranges across concrete cases where too many separate rights-holders block productive assembly: drugs that never reach patients, creative works trapped in copyright limbo, spectrum frequencies left idle, and high-tech products snarled in patent thickets. The conversation moves between the abstract—mathematical symmetry between commons and anticommons gridlock, the difference between too few and too many owners—and the granular, where individual copyrights on songs or patent claims on genes make whole projects economically unfeasible to pursue.
The discussion spans medicine and innovation, where Heller examines the "drug discovery gap" and explains why Golden Rice and gene therapies stumble against patent fragmentation; copyright and the arts, where the hidden cost appears in rap music that shifted from sampled collages to single samples for licensing reasons, and documentaries like Eyes on the Prize that vanished from circulation; telecommunications, where spectrum rights structure wireless capability; and law, where he contrasts solutions ranging from patent pools and fair-use doctrine to registries and the blunt instrument of eminent domain. Throughout, Heller maintains that the solution is not fewer property rights but better-designed ones—legal redesign that makes assembly cheaper rather than eliminating ownership altogether. Roberts occasionally pushes back on whether gridlock is the real culprit or merely one factor among many in innovation slowdowns.
Heller argues that beyond the familiar 'tragedy of the commons' (too few owners, overuse) lies a symmetric and largely invisible 'tragedy of the anticommons' (too many owners, underuse), where fragmented ownership of patents, copyrights, spectrum, and land blocks valuable resources from being assembled and used, destroying wealth and lives.
- Privatization can overshoot: the spectrum from commons to private is only half the range, with anticommons gridlock on the other side
- Patent and copyright fragmentation suppress drugs, creative works, and high-tech products that never come to market and are therefore invisible losses
- Solutions lie not in eliminating property rights but in redesigning them to make assembly of complementary rights cheaper (pools, registries, fair use, timely redefinition)
This asset isn't compiled yet
You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.
Both the tragedy of the commons and the tragedy of the anticommons are different versions of transaction cost failures—it is difficult to negotiate with everyone to forbear (commons) or to get everyone to contribute their piece (anticommons)—and James Buchanan and Yoon proved their mathematical symmetry.
“one one way to understand a tragedy of the anti-commons like a tragedy of the Commons is that these are both different versions of transaction cost failures”
When you have complementary inputs—like all the sections of train track to get from here to there, or all the patents needed for a single treatment—missing any one yields nothing, so the wealth-maximizing arrangement is to combine them all under single (monopoly) ownership rather than fragmenting them.
“the notion of complementary inputs you have a series of complementary inputs you need all of them you need all of the sections of train track to get from here to there and if you miss any one of them you you don't get from here to there it's not like you get three quarters ever you get nothing... separate ownership is actually suboptimal with the wealth maximizing approach when you have complementary inputs... is monopoly ownership you want them all combined in a single owner”
The class action is one of the American legal system's mechanisms for solving gridlock by aggregating a series of claims, no one of which is worth pursuing individually but which collectively generate a lot of economic value—and the Google Books settlement uses exactly this structure, though it has its own gridlock problem as stakeholders hold it up for a bigger share.
“the class action is one of the mechanisms that we have in the American legal system for aggregating a series of claims no one of which is worth pursuing but collectively generates a lot of economic value”
Property law historically assumed an owner owned a column from the heavens to the center of the earth, but when airplanes were invented this would have given thousands of landowners trespass claims against overflights; the gridlock was solved by a timely legal redefinition declaring 'heaven' to be 1,000 feet up, above which landowners have no rights.
“people's intuition is that when you own piece of land you own up to the heavens and down to the center of the earth... if you want to overfly from from New York to DC you would necessarily fly over the separate ownerships of thousands of people one of whom potentially would have a trespass claim against you... part of the solution to that was to say that you own up to the heavens but as a matter of law heavens is 1,000 feet up”
The real tragedy of the anticommons is invisible: unlike the visible polluted air or depleted fishery of a commons, the anticommons manifests as drugs never invented because scientists shy away from uncertain property terrain and cell phones that never reach market because firms know they cannot assemble the underlying rights.
“tragedy of in the ocean you can see that there's you know catching the same number of fish or lobsters tragedies of the ante Commons tend to be invisible it's the drugs that aren't invented because scientists shy away from doing research in areas with it uncontrolled the underlying property rights it's the cell phones that don't come to market”
Just as a tragedy of the commons (too few owners) leads to overuse of a resource, a tragedy of the anticommons (too many owners) leads to underuse, because each owner can block the others from putting the resource to any productive use.
“instead of too few owners in a tragedy of a common we can have too many owners in the Tri the ante Commons that ever resource being wasted in the economic sense through overuse in a commons a resource can be equally wasted through under use in an anti comment when you have too many owners that you have the potential for them blocking each other from putting the resource to any productive use”
The documentary 'Eyes on the Prize'—assembled from clips from 80 archives, ~300 photos from 100 archives, and ~120 songs—won an Emmy in 1987 then disappeared into a vault, unable to be burned to DVD or rebroadcast because the filmmaker could not control and reassemble all the hundreds of separate small copyrights needed.
“a guy named Henry Hampton a filmmaker the documentary about dr. Martin Luther King and to make this documentary film Hampton pulled video clips from 80 archives he had 300 photos that he pulled another hundred archives he used about a hundred twenty songs when he broadcast eyes on the prize it won an Emmy in 87 and then it disappeared the film went into a vault it couldn't be burned onto DVD it couldn't be shown again on TV because he didn't control... all of those hundreds of separate small pieces of Rights”
The nature of rap music changed because of copyright gridlock: artists like Chuck D of Public Enemy used to rap over collages of hundreds of tiny samples, but today rappers use a single sample because licensing one is feasible while licensing 300 is virtually impossible—so the legal/gridlock environment, not changing taste, made an entire style of artistic expression effectively illegal.
“Chuck the Public Enemy he used to rap over basically what he called the collage of sound hundreds of tiny samples... today rappers don't do that they have a single sample and the reason they have a single sample so they can license one sample fairly easily the licensing 300 is virtually impossible so it wasn't that taste changed in hip-hop music it's that the legal environment made the gridlock environment made a whole style of artistic expression illegal”
The solution to anticommons gridlock is not to eliminate rights but to first notice that fragmentation across drug development, telecom, airports, and creative arts is the same problem, then design tools that make it easier for entrepreneurs to assemble rights cost-effectively—such as patent pools, which allow a DVD player to use several hundred required patents via a single pooled licensing fee.
“the first and most important step is to notice that there is a problem that's a fragmentation can be extremely costly... to think about tools for making it easier for entrepreneurs in the market to assemble rights where it's cost effective to do so... we have what's called patent pools the reason that your you can pop a DVD disc into any DVD player is that the several hundred patents that are required to make that work are all pool together and you pay one licensing fee to the entire pool”
Private ownership usually creates wealth, but too much ownership has the opposite effect: when too many people own pieces of one thing, cooperation breaks down, wealth disappears, and everybody loses—a free-market paradox called gridlock.
“private ownership usually creates wealth but too much ownership has the opposite effect it creates gridlock gridlock is a free-market paradox when too many people own pieces of one thing cooperation breaks down wealth disappears and everybody loses”
Golden Rice—engineered by Ingo Potrykus and Peter Beyer to produce vitamin A and potentially save millions of lives and prevent millions of cases of childhood blindness—was ready to go but infringed roughly seventy US patents, and only reached potential market through a humanitarian license after the inventors shamed the patent owners.
“some scientists English focus and Peter Boyer realized that one simple solution to the problem of major problem child blindness around the world and child mortality is stems from vitamin A deficiency... in the process of engineering vitamin A rice that they call Golden Rice they discovered that that new type of rice infringed on roughly seventy US patents”
A central registry is valuable when the social value of identifying owners is high and the cost of maintaining it is low (as with land titles and car DMVs), but for copyright the rights are both enormously numerous and enormously fragmented over time (creators die, lyrics vs performance vs use rights), making a copyright registry complicated to establish.
“for copyright you have an enormous number of works and an enormous number of Rights in them which fragment quickly over time as the original creator dies or in the music context the difference between the person who writes the lyrics the person who performs it there's different kinds of uses... the rights are both enormous numbers of them and enormously fragmented”
Empirical work by Bessen and Meurer ('Patent Failure') finds that, leaving aside pharmaceuticals and chemicals, the net effect of patents on economic growth is wealth-destroying—they suppress innovation rather than create it—and firms like Intel and Google would rather protect IP through trade secrets, process protections, and first-mover advantage than through patents.
“they have a book called patent failure... they've tried as best I can as economists to measure the overall effect of patents on economic growth and their findings are pretty startling which is that leaving aside pharmaceutical and chemicals... outside of that area the net is that patents are wealth destroying they suppress innovation they don't create it”
Among rights-management agencies (ASCAP, BMI and global equivalents), the biggest concern is no longer piracy but gridlock—it is so hard to assemble the rights people need for sensible, wealth-maximizing uses that people either forgo the use or become pirates, so piracy itself is an artifact of the difficulty of breaking through copyright gridlock.
“the biggest concern is not piracy anymore the biggest concern is is actually gridlock that it is so difficult to assemble the rights that people need for reasonable economically sensible wealth maximizing uses that people end up being either either not using them sorta forgoing them altogether or being pirates that the piracy itself is an artifact of the difficulty of having sort of breaking through gridlock in the copyright area”
Cutting-edge medical treatments like gene therapy and cancer diagnostics require access to many separately patented genes, each owned by a small biotech startup that demands a price reflecting potential profits from the whole therapy, creating a holdup problem where assembling all the patents becomes prohibitively costly.
“imagine walking into Nora thorium and it's filled with the owners of all the relevant patents to one gene therapy treatment or one diagnostic tool each of those patents is separately owned and unless you can assemble get agreement from every single patent owner in the room the tool or the therapy doesn't work... they each demand the corresponding price a price that has in mind potential profits that you might make from the therapy as a whole”
Current patent law (the 1952 Act) was written before DNA and the internet, supporting an industrial style of innovation with a close link between a single patent and a single product; but modern wealth generation is about assembling hundreds to tens of thousands of patents in a single product, so patent law is mistuned to today's innovation style.
“the current basic patent law which is designed in nineteen was written in nineteen fifty two before we discovered DNA before the internet... that style of innovation isn't the way America innovates anymore... wealth generation today is largely about assembling hundreds thousands for tens of thousands of patents in a single product but a patent law that is absolutely not tuned to that style of innovation”
Fair use is part of the original copyright bargain (dating to the 1700s) giving creators limited rather than absolute rights; one previously unnoticed virtue of fair use is that it functions as a place to solve gridlock—the wider the exception, the easier it is to assemble small fragments (e.g., music samples) into new creative works.
“one of the previously unnoticed virtues of the fair use exception is that it becomes a place where you can solve gridlock so the wider that exception is the easier it's assembled small fragments under one new umbrella to put together a collage of small snippets of music samples into some new creative work”
Elinor Ostrom's Nobel-winning insight identified an intermediate solution between commons and private ownership: under certain conditions—repeat play, cross-cutting relationships, small numbers, shared community—groups can manage common resources effectively through emergent norms enforced not by the state but by the players themselves, and this can be wealth-maximizing.
“the circumstances in which groups are able to manage resources effectively even though they are Commons... having repeat play having relationships that go across that the resource and other resources having a small small number of people having a you know sharing the same church... those sorts of norms can be the basis for wealth maximizing behavior that you don't need to look on the one hand to Leviathan to the state... there can be this intermediate solution between Commons and private that can be wealth maximizing”
Patents are generally good and necessary—without biotech patents there would be no biotech revolution—so the gridlock insight is not anti-patent; rather, privatization is not the endpoint but an optimum between overuse and underuse, and the goal is not the most property rights but the best-designed property rights.
“patents are a good thing right if it weren't for biotech patents there wouldn't be the biotech revolution... privatization isn't the endpoint it's an optimum between overuse and underused between Commons and anti-commons that what we're aiming for is not the most amount of property rights or were aiming for is the best design property rights”
Roughly 20 million books are under American copyright but out of print—neither freely usable (public domain, pre-1923) nor purchasable from publishers—so this bulk of 20th-century knowledge is effectively inaccessible and unsearchable; Google Books attempts to solve this fragmented-ownership gridlock by digitizing them and routing fees to a central registry that tracks down owners.
“there's a roughly 20 million books under American copyright that are under copyright but aren't in print can't go to the bookstore... unless you happen to be near a library that happens to have a copy you have no way of accessing it... what Google has done is for those 20 million books they digitize those and they're trying to make them available to people at the word level”
There is a 'drug discovery gap': over the last 30 years roughly 40,000 DNA patents were granted and drug R&D investment rose steadily, yet the number of drugs that actually treat disease steadily declined, and patent gridlock is an important part of the cause.
“in the united states in the last 30 years we've had 40,000 DNA patents roughly granted we've had a steady increase up and up and up year-over-year in drug R&D investment but the amount of money going into the pipeline and researchers been going steadily up the number of drugs that actually treat disease has been steadily going down we have what I call a drug discovery gap”
Patent reform is stuck because Congress cannot move forward until the two big opposing sides reach bipartisan agreement, but the relevant divide is not Democrat versus Republican—it is pharmaceuticals versus IT—and breaking this requires a supermajority that is the political equivalent of the economic gridlock the book describes.
“the bipartisan there isn't Democrat and Republican the bipartisan there is pharmaceuticals and and IT”
Modern high-tech products like cell phones (thousands of patents) and online banking (potentially 10,000 patents) make it both impossible to discover all the patents you need and impossible to know whether their claims cover your product, so no high-tech product can avoid infringing dozens or hundreds of patents.
“your cell phone several thousand patents right on any particular cell phone there's no way to do online banking without infringe or without dealing with potentially 10,000 times and it is given the way our patent system works it is both impossible to discover the patents that you need and impossible to if you can discover all them to know whether or not the claims cover your product or not you simply cannot design a high-tech product that isn't infringing dozens or hundreds of pens”
Eminent domain solves the holdup problem by eliminating individual ownership when assembling a large bundle of property rights, but it is a last-best 'nuclear option'—abused (raising Public Choice problems, as in Kelo), broadly hated, and not the right approach for intellectual property, where the analogous government 'march-in' right on patents would crush pharmaceutical companies' incentive to invest the ~billion dollars it takes to develop a drug.
“you can always solve the problems too many owners by eliminating primary ownership but that to me is really a last best solution”
Paradoxically, the patent system exists to promote innovation, but more patents—too many patents—can mean fewer life-saving innovations because they create gridlock.
“the reason we have the patent system is to promote innovation paradoxically more patents too many patents can mean fewer life-saving innovation”
Over 90% of the airwaves in America are dead air (unused), because spectrum property rights were defined as thousands of tiny, geographically located, non-transferable, limited-use licenses that make it nearly impossible to assemble national high-speed wireless networks.
“over 90% of the airwaves in America are dead air that is they're not used at all and that's important artifact of the way we define property rights in broadcast spectrum we have thousands of owners with tiny geographically located non-transferable limited use licenses that make it virtually impossible to assemble for example national high-speed wireless networks”
Spectrum gridlock—what Thomas Hazlett calls the 'tragedy of the telecommons'—has cost the US economy trillions of dollars, and the US fell from number one in wireless/information economy capacity to near the bottom of the top 20 over the last 20 years.
“Tom Haslett actually at George Mason where you are is calculated that the cost of spectrum gridlock what he calls the tragedy of the telecom ins a riff on this anti-commons idea cost the US economy trillions of dollars”
Universities and university presses are on the wrong side of the copyright debate—favoring strong copyright and limited fair use—because their fair use policies are set by general counsel offices thinking of the university as a property-rights owner, not by faculty who want the widest dissemination of their work and are paid in prestige rather than cash.
“university presses like... Princeton Press and universities in general are often on the wrong side of this debate they come out typically in favor of strong copyright and limited fair use and I think as educational institutions... that's actually the wrong side for them to be on usually professors want the widest dissemination of their work”
Fair use doctrine is not fixed but is defined by what people are actually doing ('reasonable use'); rights holders try to 'brush you back from the plate' to shrink the strike zone, so the only way users retain rights to use snippets is to assert them—and faculty should be more aggressive about asserting educational use.
“what fairy doctrine is at any point is the is what's reasonable... it's like in baseball they're trying to brush you back from the plate and the more they brush you back... that changes the strike zone so the only way that we get to use snippets your thirty five words or educational use is this is if we assert it I think that faculty should be should be more aggressive about asserting educational use”
Under use only became a recognized word in English (and a legitimate Scrabble word) three or four years ago, whereas overuse has been a word for four hundred years, reflecting that under use simply isn't part of our core cognitive structure for thinking about resources—we treat resource problems as a one-way ratchet.
“overuse has been a word in English for four hundred years the problem of under use... is almost completely invisible under use only became a word in English in Scrabble only three or four years ago... under use simply isn't part of our sort of core cognitive structure for how we think about resources we have to have a one-way ratchet when we think about it”
There are many private, quiet ways to solve gridlock/holdup—such as assembling land without a big announcement and buying parcels quietly—and there have been many successful examples.
“there are a lot of private solutions that's these kind of problems that that happen and at work you don't make a big announcement that you want to buy up a bunch of land you're gonna do it quietly and of course have been many successful examples of this to avoid the holdup problem”
Privatization aligns the individual's incentive with the resource and ultimately with social welfare, because if you take too many fish out today there are none left to reproduce for tomorrow, making a single decision-maker willing to invest in conservation.
“what privatization does is it align the individual incentive with a resource ultimately with a larger social welfare you take too many fish out today there's none left reproduced for tomorrow”
Heller experienced gridlock first-hand assembling images for his own book—for example, a Quaker Oats square-inch deed to Alaska land (a 1950s cereal marketing campaign that sold tens of millions of deeds) took weeks of negotiation and a quitclaim agreement with PepsiCo's IP lawyer to use as an illustration.
“I have a deed to a square inch of land in Alaska... a great marketing triumph by Quaker Oats cereal and they put these deeds in cereal boxes... tens of millions of these square inch deeds... just to use that picture in the book... took a weeks of negotiation I had to track down... Quaker Oats it's not part of PepsiCo... we ended up working out a quick claim agreement”
EconTalk could not license the opening bars of 'If I Had a Million Dollars' by Barenaked Ladies because the publisher demanded a high price and the rights were complicated by separate lyrics, music, national, and international rights, even though the podcast makes no money and is for educational purposes.
“my first choice was actually the opening chord the opening bars of if I had a million dollars by the Barenaked Ladies and we've had a great deal of trouble trying to get the publisher of that music... there's lyrics and music there's national there's international rights... they wanted a very high price which was I thought strange because we don't make any money off this site it's for educational purposes”
Heller welcomes being on EconTalk.
“Michael welcome deaqon Todd it's great to be here”