Opening Remarks: Henry Smith—exploring the connections between private law and IP.
Session 1: Entitlement Design
Moderator: Rebecca L. Tushnet
Speakers and Papers/Commentators:
Tun-Jen Chiang, “The Paradox of IP”: A paper looking for a theory. Why do we have an IP system as opposed to a prize system? His understanding has always been that IP systems harness private information and that gov’ts in a prize system wouldn’t be able to value the underlying value of the right properly, most commonly b/c judges lack adequate information to do so or judges are too easily politicized in awarding cash prizes, while IP rights determine value of underlying thing automatically through market calibration. If that’s the reason we have an IP system, then one thing in tension with that is that judges in fact calibrate the value of the work/invention in multiple doctrinal contexts, most particularly when they determine the scope of IP rights. Idea/expression dichotomy: Nichols/Learned Hand abstraction test. How to apply? Judges have a feel for what’s about right in scope, and work backwards to find that if a certain level of abstraction gives too much, that’s the wrong level.
Patents essentially have the same problem and resolution: patent scope doctrines of claim construction—a big mess, open-ended, which ends up being used in a way that construes claims according to what judges feel is about right. But what’s about right? We don’t have a better answer than intuitive judgments about the economic value of the right at stake.
That account makes the reason for an IP system rather than a prize system more puzzling. Intuitive differences: IP system, we don’t have to get to the bottom dollar, we just take a stab at it. Both prize and IP systems can be imperfect. If you note that judges calibrate at the edges all the time through doctrines of claim construction or tweaking remedies or fair use, then it opens up the question of whether they should do more. Claim construction: why be bound by whatever the patentee writes, instead of reaching good outcomes within the parameters of the system/incentives? Claim construction could be judges figuring out what the claimed invention ought to be and then construing it by bending it the doctrine to reach that outcome.
Injunctions: why have a rule presuming injunctive relief when infringement is found? There are at least a few cases where injunctions would overcompensate through holdup. If you think judges are ok at tailoring, then they should be able to deny injunctions in those situations. Much of the traditional structure of IP is only explicable if you subscribe to the founding premise that the IP system is designed not to have judges calibrate b/c they’re bad at it, so we need second-order rules that say there is a thing called the work or the invention, determined in value according to market forces. In that view, the idea that judges should just grant injunctions instead of ongoing royalties and that judges should construe claims linguistically make sense to avoid fine-grained inquiry. But judges seem to be doing all sorts of calibration under the hood, and by and large they seem ok at it. Fundamental dilemma.
Commentator: Gideon Parchomovsky: (1) IP paradox (existence or not); (2) proper role of courts in IP space; (3) implications—should we be worried?
Chiang is right on a certain view of IP. Basic dilemma: we are fine with courts defining the asset in IP, where intangible assets w/unclear boundaries are at issue. Idea/expression, invention scope. But many of us feel uncomfortable when courts decide how much money in dollars should be awarded. But defining the scope of the asset of course indirectly determines value. And that’s fine. Is there a paradox?
There is a paradox only if one believes that courts shouldn’t have any say about the rewards for authors and inventors. A lot of people do seem to subscribe to this view. Either this view is wrong or they didn’t think it through. Law is always prior to the market; even Coase says that. Courts should define the asset, and only after that can market transactions take place. This isn’t an IP paradox but a much more general paradox. With Blackacre too, we need courts to determine what interferences are reasonable or not. Drones: is that trespass or not? My rights are influenced by zoning laws. Value of stock is also determined by law.
If you believe scope determinations are inevitable, the question becomes much more difficult. How much leeway do we want to give courts? Legal Realists. The determinations that are inevitable lie within the competence realm of courts. Claim construction as an example. As long as judicial decisions are reasonably predictable, that works. Judges are better at scope than evaluating assets overall.
Does it matter? Matters for pricing function. W/in a certain range, there’s enough determinacy to function. Insurance cos. have a very hard time calculating the value of patents; it’s not just scope but novelty, obviousness, survival through litigation. We also need to worry about adequate incentives to create. Chiang leaves room for optimism—there are still incentives to produce.
Oskar Liivak, “Private Law and the Future of Patents”: Problem of excessive system costs. The costs of engaging with the system is too much to produce gains. Odd if we’re trying to provide a reward for inventors. Dickens, 1850: A Poor Man’s Take of a Patent; shows up in Steinbeck’s East of Eden as well. Private law is attractive: property, contract, and tort—accepted and stable; low administrative costs; cheap to operate—something that patent is not (accepted and stable). Generally speaking, people abide by their duties, but litigation is on the margins. Patents: everything is in the shadow of litigation.
Are patents private law? No: empirically. They’re just not accepted and stable like these other institutions of property, contract, and tort. Does the public feel an obligation to obey? No. Could patent be private law?
Patent theory is incompatible with becoming private law. Sichelman said the same thing about remedies. His solution is to rid us of private law concepts to become consistent. Worried that we won’t get efficiencies w/o private law, so suggests getting rid of dominant patent theory and build something compatible with private law.
What makes private law special? Purpose: enable socially beneficial activity; allow coordination of activity. Structure: duties to avoid harming others. System of rules. Criteria for efficient operation: we can comprehend our duties (H. Smith on information costs and modularity); we feel an obligation to obey. HLA Hart’s discussion of the internal viewpoint on an institution; less worry about litigation and more about what one ought to do. Systems become cheap to operate under those conditions.
Dominant patent theory: purpose, promote progress in useful arts. Structure: provide incentives to induce activity. Patents are a reward to patentees subsidized by those that infringe. Once you set up the world this way we’re in trouble from a private law perspective.
Duties: pay patentee whenever infringing; no real sense of harm to patentee. Criteria for smooth operation: can we comprehend when to pay? No. Do we feel an obligation to pay? No. Only obliged by litigation (risk).
What to do? Think about getting away from incentives. Private law does incentivize through property, but it’s generally not by telling people to produce/telling us it’s there for ensuring property owners they’ll make money. Get away from the idea of subsidy. Ask what we want to have in the world: what interactions, coordination behavior? Once we think that way—what could third parties do to harm a business model—we start moving towards something better. Don’t do things to harm people trying to supply their tech to other users. We could comprehend what to avoid (people who are commercializing/transferring tech) and what causes harm (copying). Could even tell a story about what’s wrong with a complete defense for independent inventions. Could feel an obligation—a regular business interaction, not just misappropriation.
Purpose of judiciary: What do judges think they’re doing? Learned Hand: We don’t see how our decisions ultimately affect the end result, levels of innovation. No internal viewpoint, just reward system that no one has a good feel for. Missing essential characteristics to make it easy/efficient to operate. Not necessarily morality but acceptance of what harms to avoid; if the tech community could understand it, so could judges.
Commentator: Adam Mossoff: Failure of fit between a lot of the way we theorize the patent system as a public regulatory system that pursues economic goals by subsidizing innovation versus what we see in operation, a private law type structure. Two points: 19th century; empirical claims.
Shifting to a focus on market transactions as a proposal: in talking to people who work in the innovation industries, this is their perspective, but also doctrinally on to something, b/c many courts conceptualized patents in the 19th c. this way, which led to patent licensing market. Common to hear in American/British system discussion of monopolies; this is true in the same sense that American political system came from Britain: came from it, but also broke from it. So too w/patents. British patent system was economic regulatory system, first to file; not viewed as property rights in England. A patent was a personal monopoly privilege; couldn’t be transferred. James Watt was an academic researcher, but he had to find a business person to work with when he got a patent; he couldn’t just sell it.
US broke w/that approach; first to invent, but more importantly recognized patent as property, bringing a normative structure to thinking about patent. History has a lot of policy dispute; you can find judges who say otherwise in the 19th c., but the majority rule is courts citing to property doctrines in patent cases. They did this when they dealt w/commercialization in the marketplace: they say “we are not like England.” This wasn’t a matter of remedies (not primarily); about the ability to frame conceptually w/ability to use normative principles—the right to use, the right to dispose of the right. A private ordering presumption. Adopted common law concepts of assignments and licensing. As a result, economic historians say, there’s an explosion in economic activity involving patents. Apple licenses its patents actively.
Too much myopic focus on litigation; we do this b/c we are lawyers, most of us w/litigation backgrounds rather than transactional; litigation is also public, not behind the scenes. But there’s a huge amount of transaction/licensing. Talking about breakdown of patent system w/o considering existing market transactions is armchair empiricism; benefits to patent owners exist under the current system. Patents contribute $6 trillion to the US economy through transactions, licensing. That’s the denominator against which to measure the costs. Get into empirical side. Lots of empirical work: patents double your chances of getting startup funding.
Molly S. Van Houweling, “Disciplining the Dead Hand of Copyright: Durational Limits on Remote Control Property”: Connections b/t tangible property law and copyrights. Blackacre and Black Beauty. Controversial endeavor: Lemley argues against propertization of IP; Mossof argues for. Peter Menell has expressed skepticism; Liivak has said that it too often corresponds w/patent absolutism. Michael Carrier: property has lots of limits, edge cases, complications as compared to the monolithic version often presented as “property” when it comes to IP; we can learn from property’s limits.
Conventional wisdom differentiates Blackacre and Black Beauty in time b/c tangible property rights can be infinite while constitution requires IP rights to be limited. Why might that be? The special nature of © as a prohibition of conduct remote from the persons or tangibles of the party having the right. May be infringed a thousand miles from the owner w/o his awareness, ever. This right couldn’t be recognized or endured for more than a limited time: from White-Smith v. Apollo. Remote control property over time makes us particularly nervous.
Why? Special notice problems. Don’t have notice helpers as we do w/traditional tangibles—owner on the land; fence in front of the owner; limits don’t have to do w/boundary crossings but w/use limits, more confusing than conventional non-remote control property rights. Interferes w/ basic intuitions about what we can do as owners of tangible things. Problems can get worse over time as we lose track of the owner/the way they wanted to limit use.
Special obsolescence problems. Non-possessory rights are especially likely to become out of date. Difficulty of finding absent owner to renegotiate. Non-possessory rights can be fragmented and overlapping.
And yet: expanding duration of copyright makes it not-so-limited. Perpetual copyright on the installment plan; some copyright enthusiasts want it to be infinite—unfair disadvantage v. other property owners. What skeptics and enthusiasts share is idea that there is a fundamental difference between permanent property rights and “limited times.” But problems of remote control property plague tangible property as well, and therefore we see duration limits in the tangible realm. Rule against perpetuities; ex ante durational limits for servitudes and future interests; periodic recording requirements for servitudes and future interests; ex post termination/modification of servitudes and future interests; adverse possession; statutes of limitations. Note the ones that operate on nonpossessory interests like servitudes and future interests especially. Prohibition remote from the persons or tangibles of the party having the right, as in White-Smith. Recognition in tangible property that this can become problematic—Md. legislature notes change of conditions in restricted tract or neighborhood surrounding it—the usefulness of many reversionary interests vanishes. Not practical to obtain releases b/c owners are numerous and scattered—similar to orphan works and other duration problems. “Dead hand”—owners do exist, but they are unfindable/too many—they might as well be dead.
So some states have ex ante duration limits on restrictive future interests: 30 years in Md. Another solution is periodic recording requirements for nonpossessory interests in order to make them trackable/provide adequate notice. Cal. approach. Ex post termination of obsolete restrictions—Mass. approach, must provide substantial benefit to a person claiming rights of enforcement at the time rights are asserted. Obsolescence can also lead to remedial adjustment—remedy is damages and not injunction.
Bringing this home to ©: think about how we might use some of those tools from tangible property to address the problems in © with increasing duration/obsolescence/changed conditions. Google Book Search: a © that seemed valuable initially now just threatens to lock up works that could be put to beneficial uses. Paul Heald’s work showing the power of © to keep books out of print, likely against the wishes of the authors themselves. Reinforces other suggestions such as recording requirements (Sprigman etc.) and scope adjustments (fair use over time) and remedial adjustments (Copyright Office orphan works proposals). © is problematic private law, but a lot of the heartland of private law is also problematic—servitudes. Copyright reform in the private law tradition.
Commentator: Julie Cohen: Private law skeptic. Framework bleeds through, intentionally or not. VH’s project is to find firm analogical footing for temporal limits on ©, and is persuaded that time is/could be relevant lever for tailoring. But not sure analogy proves very much about how relation b/t © and time should work. Some of the rules VH examines, including Rule Against Perpetuities—it’s pretty clear that the real property system worries about dead hand control, but not clear how that cuts. Both in original formulation and various modern formulations, rule is supposed to provide balance by allowing 2 generations of control but not much more. That’s exactly why Congress lengthened the term—the same accommodation.
Servitudes: problem is more fundamental/revealing limits of private law project. Remote control of chattels embodying © works and uses. But servitudes aren’t nearly as disfavored as the paper presents them. They aren’t just vehicles for remote control by atomistic isolated owners; they’re powerful, flexible vehicles for communal ordering—create residential neighborhoods, commercial development (anchor tenants), and for exactly those reasons the thrust of the Restatement is greater liberalization for CCRs, noncompete covenants for shopping centers; get increasingly favorable treatment over time. Offer tools for ordering that cut across atomism of traditional real property: collective benefits, collective externalities. So why is the paper taking an atomistic view of servitudes? Is it the private law methodology/toolkit inherently atomistic, emphasizing clear boundaries and transactions rather than communal ordering. That atomistic orientation is particularly problematic for understanding servitudes and for understanding IP.
IP is the subject of pervasive intermediation: intermediaries manage all the fractional and cumulative uses; also the production of tangibles & intangibles related to IP involves lots of intermediaries—publishers, movie studios, tech employers—they are production intermediaries. Layers of intermediation embedded in systems of IP production and dissemination. From an institutional POV that means that legal institutions need to be able to manage problems of access for cumulative/fractional use and they need to constrain the intermediaries in some way.
Time could be important in designing such institutions, particularly for orphan works. Even so worries about remote control/dead hand don’t get us far into the institutional design questions. Study IP directly w/o a filter needing analogies.
RT: The theme I extracted from these three: Anything that can’t go on forever, won’t.
Liivak says; For the most part these institutions are self-enforcing. The stake-holders participating in these institutions know their rights and duties and they largely abide by them.
[How do we know that? Counterexamples. Property: mortgages. Contract: consumer contracts. Do you know if your cable provider is complying with its contract with you? Tort: medical mistakes that cause harm)
Van Houweling: consider servitudes—they don’t work as well as other parts of property law, but we don’t reject them. Question raised is whether it’s worth the candle in patent law. Our current system fars fall short of dealing w/pathologies in the way servitude law has evolved, by requiring proper notice for example as well as durational limits.
Henry Smith: Is this all about the difficulty of asset definition? Calibration: trespass is a first cut, nuisance is a refinement. All-nuisance, all-use all the time world would be pretty difficult (but is the world we have in IP?). Maybe the reason it seems like a paradox is that we aren’t clear on what the resource is in the first place. Servitudes: we have remote control, but also incomplete separation too. We don’t allow many servitudes on personal property, but at least we know what the property is. More like water law—can’t go as far in defining.
Chiang: In every area of property we have a problem of asset definition—always has implications for value. We don’t have a well-theorized second-order rule or methodology to do asset definition, whereas in tangible property law we might not have a complete, gapless, ambiguity free rule, but it’s more tethered to first-order intuitions about the underlying asset value. Trespass isn’t defined by the worth of the property, though nuisance has more of that flavor, but still less than © or patent. To the extent that asset definition falls back on first order intuitions that’s about asset value, that’s a circular definition.
Van Houweling: yes, we’re all getting at a problem of asset definition. Chiang’s paper prompted recognition of tangible property cases w/ such problems—mistaken boundary claims, Manilo v. Gorski; no injunction to make encroaching building be torn down, but calculate damages instead. Upsets the idea that there’s a paradox b/t clear rules and private ordering v. judges having to do valuation on occasion. Q is whether it’s rare or endemic in a way that makes it a paradox.
Ted Sichelman: I try to show that if we adhere to a private law framework we don’t promote innovation optimally. Multicomponent products, independent inventions—we care differently about notice costs, clear boundaries. If we try to move from regulatory system (theory and goal) to private property focus we’ll have a mismatch. Other situations, private property works quite well; just b/c you have a regulatory goal doesn’t mean you can’t use private property and private law to promote that goal. Don’t think it’s all or nothing. Wanting everything to look neat and tidy = mismatch.
Liivak: we might well agree on doctrinal details. Worry that framing affects analysis. Independent invention might be a good test—if a private law view has the flexibility you’re interested in, maybe I can convince you.
David Kappos: You’re looking for a system to help people understand patent boundaries better. ECJ & German courts have been working on this in standard-essential patents, setting forth a rubric requiring the patentee to make a clear offer of a reasonable royalty license. The erstwhile licensee must respond clearly; if there’s not agreement, the potential licensee is required to post a bond and there’s a resolution path. Does that shine a light to further clarify property rights?
Liivak: sounds very interesting. Maybe it was implied, but I think there’s an important distinction b/t ex ante and ex post licensing—when you have tech you haven’t independently invented, I’m all for such negotiations. If it’s instead allowing or preferring ex post licensing where the patentee looks around at the world to find payors, I’m less sympathetic—don’t see social benefits. Any distinction b/t ex ante and ex post, putting thumb on scale for real tech transfer, would be good.
Rachel Sachs: In terms of needing a reason to get rid of the dominant theory in patent law—why do we want this to look more like private law? Is there a fundamental flaw in the theory? Is it just expense? Can private law do it better for some reason? Is there an overlapping consensus, and is it clearly cheaper to do that under private law? Maybe we don’t feel obligation to IP b/c it makes lawbreakers of us all, but if IP is an essential human right, that might matter.
Liivak: I start out talking about system costs, but we need more reason than that. Before we get to tailoring, we have the bigger Q: how much innovation do we need compared to other activities; dominant theory is premised on knowing a certain number that’s unknowable v. shoe stores or Thai restaurants. We don’t need to know if we go to the private law model and set up a neutral platform for those who innovate. (If the rest of the economy is efficient.) Rather than IP exceptionalism where we distort markets, think of it as neutral platform and don’t try to maximize innovation v. other things.
If you take reward theory seriously, your first go round has to be ok with trolls. They’re just collecting the promised rewards, including from independent inventors. That theory has come up short on the ground.
Van Houweling: linked to Cohen’s point: why look at system costs adjustment and why not challenge the system if we’re not getting the benefits of community from real property servitudes. Likewise, remote control rights do assist in coordination/long term planning v. author only owns rights in manuscript. Doesn’t prove system is worth its costs; what we certainly shouldn’t have is the pathological system without any of the tools used to moderate the problems.
Brett Frischmann: For Chiang: Demand manifestation in systems driving allocation of resources—ex ante investment decisions are the core of patent; we can tolerate some ex post errors if ex ante is driving most decisions. For Liivak: duty to obey in private law—empirically needs verification. Dave Hoffman on how different generations approach contract differently; if the premise is wrong, that pulls the rug out from the project. If you’re going to do private v. public law or the hybrids, you have to do comparative institutional analysis. Why do you want the alternative model you advocate? Just saying the end is “innovation”—what does that mean? What the normative objective is and the means.
Chiang: Literature on patents v. prizes begins w/ the valuation issue—not the whole story but the first cut. As for systematic point, yes but I’m not sure it provides an answer in that both patents and prizes are about systematic overall incentives rather than individual fine tailoring. Doesn’t matter if courts get it wrong in one particular prize or patent as long as there’s no systematic error.
Jonathan Barnett: You’ll never find a Delaware or NY or English judge deciding what the damages are unilaterally; typically they’re what the parties said in the contract, or expectations with a market benchmark. If we take that back to patent, it seems like something similar is happening there. Hard to get a market benchmark for claim construction, but for damages, we want to look at how the market is valuing that right b/c a patent is simply a way to allocate intellectual resources into IP modules, so damages try to replicate the transaction that would have occurred—either through reasonable royalty or injunction. With contract, judges choose specific performance when there is no benchmark and they want the parties to negotiate/reveal value through that exercise.
What happens when you go to monetary damages in patent—your market benchmark becomes less accurate b/c there isn’t a rich market any more for those patent assets. That’s why injunctions are valuable.
Chiang: that’s one side. One wrinkle: it only works in contract b/c you have a second order methodology for asset definition that doesn’t depend on valuation. We think it’s improper for the court to say the plaintiff made a huge contribution so I will construe the contract in a way that expands defendant’s obligations to the plaintiff.
Liivak: for Frischmann: trying to focus on tech transfer—that’s the basic definition of innovation. Wants to build that theory around micro-transactions.
Patrick Goold: Idea that private law is cheaper—we had the same debate in 1980s torts about private litigation or regulation. The idea was that regulation would be cheaper b/c litigation was so expensive—Sugarman said “do away with tort system.” Main critique is that when self-enforcement does break down, litigation costs were massive. Would amplify that in patent where the costs are already great.
Greg Vetter: How open would Liivak be to technological tailoring/less uniformity if the resource at issue will never work very well as a property right. E.g., software only for ©, not patentable.
Oren Bracha: Theme in the papers: “property is a neutral platform”—for Bracha that’s like saying the earth is flat. Property is not neutral; it is a high-stakes system for allocating resources, power, and power backed with the coercive power of the state behind it. Requires substantive justification, which might be different across us—natural rights, efficiency, something else. You can’t just take a characteristic of property and call it neutral.
Michael Meurer: For Liivak: put some discussion of citation practice by scientists in this paper. Seems we have a private legal system there that works reasonably well; scholars rely on scientists citing people they should cite; the comparison has been drawn.
Van Houweling: property isn’t a neutral platform and it can’t just be wound up and left to run; property system intervenes to correct pathologies.
Liivak: He’s not saying property is neutral, just better than what we have, and he should be more careful about his distinctions.