Saturday, February 13, 2010

Intellectual property

This expands a comment I made here. Skepticlawyer herself kindly looked over it, so I have incorporated her longer comments (slightly edited so they agree in audience) as indented quotes.

Intellectual property (IP) comes in three basic forms: trademarks, which are a form of branding and operate as long as they are being commercially used; patents, which operate for limited times (typically 20 years or less); and copyright, which lasts for the lifetime of the artist plus some period (now 70 years).

I once heard Kenneth Arrow say in a lecture that he could never resolve his thoughts on intellectual property because he could not determine how to balance that before the fact, we want to encourage people to create things, but after the fact we want as much access as possible to maximize the benefit gained from the thing created. Since the benefit from disseminating the knowledge protected by patents is the most extensive (in both scale and scope), it is hardly surprising that they have the shortest lifespan of the forms of intellectual property. Since trademarks are a form of labelling—so their wrongful use is typically straightforward fraud, pretending to be something you are not—it is hardly surprising they last as long as the brand is operative. They are still a form of IP, since they invoke reputation (and other associations), but a fairly unproblematic form of IP.

But the dilemma Arrow pointed to is, in a sense, true of anything useful that we create. What makes intellectual property distinctive is:
(1) use of the created thing can occur, repeatedly, without degrading the thing itself; and
(2) it is hard to block access to the thing created once it is in the public space.
Note that neither point is as simple as saying they are non-rivalrous (though they generally are) and non-excludable (though exclusion is typically difficult): i.e. are public goods.
Skepticlawyer comments:
Not always public goods, but almost always non-rivalrous; club goods are also non-rivalrous.
Where IP is non-rivalrous but at least somewhat excludable, they are club goods. But an odd sort of club, since the only benefit of being in the “club” is avoiding legal hassles. That is, “membership” only involves use of the IP that is entirely usable without the “club”. The more difficult exclusion is, the more salient that is.

To see the implications of non-degradability, suppose someone clears land for a farm that proves highly productive. Allowing anyone to use that farm however they liked would quickly degrade the farm. We would lose the benefit of the thing created, even without considering the disastrous incentives created to not engage in difficult productive activity.

But if someone composes a song, then any number of people can play or listen to the song without it being in anyway degraded. On the contrary, that people use it “keeps it alive”.

Even the issue of the incentive to create is not quite the same in both cases, since the author can get kudos from the repeated use of his or her work. Though the issue of degradation is not entirely absent: copyright can protect vulnerable contributors to (for example) a documentary.

It is relatively easy to stop outsiders degrading the farm: one can fence the farm, the borders can be marked and the normal processes of the law used to protect the physical boundaries. There are marked boundaries, physical barriers and legal recourse: the latter is more effective the easier trespass is to identify. For then it is more likely to be prosecuted while such prosecution will be less needed the more widespread is respect for delineated boundaries (which likelihood of prosecution can, of course, further encourage).

Property rights depend quite fundamentally on the existence of a general willingness to abide by them and sufficiently effective legal recourse to back them up, as well as sufficient clarity in boundaries that they can be usefully enforced and easily respected. Hence the legal concept (laches) that one can lose rights by not exercising them: if you do not take a boundary seriously, why should anyone else? (Or even, in some circumstances, know they are there.)
Intellectual property has real boundary (and other) problems. It is a lot easier to put fences around a farm than around a song or the knowledge of how to do something. The farm is a singular physical thing. A song is an arrangement of things that can be repeated any number of times. (To be somewhat Aristotelian about it, the farm is property in a thing, owning a song is property in the form of a thing that can have any number of instances.) Copyright is about controlling copy right, after all.
Skepticlawyer comments:
This touches on the point that armacny raised in comments: there are fundamental differences between corporeal property and non-corporeal property—choses in possession as opposed to choses in action. IP law—in certain crucial respects—tries to treat the two as identical, but they are not; the differences, while subtle, bleed out everywhere in practice.
Identifying trespass on a farm is policing a specific physical space. Identifying trespass on a song requires policing public and private spaces in general. A farm has a clear boundary: what level of commonality in arrangement of things constitutes intellectual trespass is a lot more unclear. (Such as deciding a common 10-note sequence constitutes “use” and thus trespass and so chargeable use—since it is the power to block usage which gives one something to charge for doing or not doing.) Identifying and policing boundaries is far more difficult for intellectual property than ordinary property. Though trademarks are easier to deal with than patents or copyright, since trademarks are in regular use on physical things.

Which means, of course, that it relies a lot more on the formal structure of enforcement (courts) and informal structures of enforcement (respect by individuals) primarily because the legal property rights align so poorly with economic property rights (control of the thing). So much so, that what is surprising is not that there are difficulties with intellectual property, but that the idea has any legs at all.

Revealingly, it seems to have legs in direct proportion to clarity of boundaries, difficulty of access and balance of benefits. TV stations, cinemas, books and other “hard copy” materials are relatively straightforward for intellectual property since they are so public in operation. The operators are also very much in the business of dealing with intellectual property as both beneficiaries and potential transgressors, so they have counterbalancing incentives. That they can operate intellectual property regimes is not surprising. Indeed, the main role of the state may well be to cut down on transaction costs by providing a single set of rules plus dispute resolution tribunals.

Manufacturing patents are in a somewhat similar situation. The main exception being where incentives are very one-sided. So we would expect poor attention to intellectual property by countries whose manufacturers are far more users of intellectual property than creators of it. Places such as China or Pakistan, for example.
Skepticlawyer comments:
Bit of IP history people may find amusing: the USA was notorious for abusing intellectual property in the 18th and 19th centuries; they were constantly doing to European inventors then what China is doing to American inventors now. This even overspilled into literature, and was still going on in the 20th century. That’s why there were so many ‘pirate’ US editions of Lord of the Rings and practically the whole oeuvre of such noted authors as George Eliot and Charles Dickens—something that annoyed both Eliot and Dickens no end, as they were probably the two most successful novelists in the world at that point.
There is a not dissimilar pattern with respect to strong and weak interpretations of international law: C19th US was all for strong interpretations of international law. Post 1945 US, not so much.

The rise of computers and the internet make intellectual property (specifically copyright) issues much trickier because it means that use is far less public, far more decentralized and private, so much harder to spot. We also get far more people who are direct users of intellectual property but not creators of it, so have little incentive to respect it. In other words, the particular model of distributing copyrighted products—due to the dramatically cheaper and easier access and shift in the balance of benefits among those with such access—is under severe technological stress (to put it mildly).

But one has to careful here. The desirability of different forms of ownership of some arrangement of notes/words etc may still allow sale of intellectual property even when free versions are easily available. When Baen Books put On Basilisk Station up on their website as a free e-book it promptly hit the NYT bestseller list: i.e. people went out and bought versions of it.

Is this people respecting intellectual property? Or is it that the internet acts best as a “billboard” and people just like having the hard copy version? Some performers, such as Prince, hold that giving music away free on the internet is still worth it because enough people will still go and buy the CD or whatever that it is worth one’s while. It is better to be seen to be “with it” than trying to enforce unenforceable property boundaries with all the arbitrariness that involves. Of course, whether this strategy would work so well if everyone did it is another question.
Skepticlawyer comments:
There is quite a bit of research to be done on this very peculiar form of altruism out there. It is related to those city councils that have a ‘donation’ system for parking. Once they’ve sacked their parking meter people and reallocated police enforcement elsewhere, the council often finds it makes as much—or more—money from its parking meters than it did when there was a draconian system of enforcement in place. Both Linkin Park and Radiohead have also released free versions of their music online at various times; in both cases, the bands finished up making MORE money than they would have had they pursued a ‘normal’ studio release for which everyone had to pay.
If humans did not have strong basic norms of ownership, any strong form of private property would be quite impossible. (Chimpanzees, for example, do not seem to be able to develop norms of ownership of things when in the wild.) This seems to be a manifestation of the strong human norm of ownership (which I would argue arises from us being a tool using and pair-bonding species: chimpanzees are neither).

It is clear enough that “intellectual property” is not a nonsense-construct or a mere exercise in exploitation. On the contrary, it seems a natural way to deal in ideas. Up to a point. But, like all property law, it relies on being generally respected and enforceable at the margin.

Which means it is possible to over-reach. Law that cannot be enforced is likely to become arbitrary, which undermines respect for it, or become a dead-letter, so rather pointless.
Skepticlawyer comments:
Legal philosopher and jurisprudential scholar Joseph Raz first made this argument; he was the one who argued that morality was most often separate from law for the simple reason that it has to apply to citizens independently. If morality depended on law for its enforcement, we would need a ratio of one citizen to one policeman in order to stop us all from killing each other, or raping each other, or whatever. The main reason most people don’t murder or rape is not because there is a law against them, but because they are wrong. For this reason Raz argues that if someone suggests that a moral principle is ‘universal’ and thus requires legal enactment/enforcement, any truth claims they make about public compliance with the moral norm must be examined with very great care.
US interests tend to be very keen on intellectual property—and have a lot of sway with US policy makers—because a very significant proportion of US exports are the products of intellectual property (computer games, films, porn, the products of patents). Hence contemporary free trade agreements with the US (either bilateral or multilateral) have strong IP provisions.

Which have tended to get stronger—there is something to the story that IP gets extended to protect Mickey Mouse. “Getting stronger” here means “more likely over-reach”. Just as property rights can persist when the state denies that they do (hence black markets) so the state can notionally create property rights beyond its competence to enforce or which are sufficiently badly delineated so as to prove of limited use or even negative use (if they cannot generate enough respect for them such that they waste resources). The further any property rights regime moves from economic property rights (who actually controls what) the more likely it is to fail. A property rights regime which most people can see they are getting something from will work much better than a property rights regime that significant numbers of people see as not giving them enough benefits: taking the public with you is not incidental to property rights, it is basic to it. (Even enforcement of widely agreed property boundaries is, in a sense, keeping faith with people).

If there is substantial “extra legal” action, then that means the property rights regime is failing. The answer to that is to realign the regime so as to maximize the range of people who are benefiting from it: and thus to maximize the extent it is self-enforcing. Extra enforcement is an easy answer: it is only the correct one if very widely accepted boundaries are not being enforced. (A poll result saying a majority of people are in favour of banning something does not constitute the needed acceptance: such superficial expressed preference is not the very broad revealed preference required.) If the problem is that too many people do not agree with those boundaries in the first place (typically because the benefits seem too one-sided), then one has to move the boundaries.
Skepticlawyer comments:
In sum, laws need to align as much as possible with self-interest as economists understand it; to the extent that they don’t, said laws are pissing into the wind. Seriously. This is my main argument against ‘moral’ and sumptuary laws of all stripes, and is in part derived from Raz’s explorations of the issue.

Indeed, sumptuary laws provide a revealing example, because medieval sumptuary laws performed much the same role as modern laws against “identity theft” do: they were to stop people passing themselves off as something they were not. In a society with limited literacy and very strong social role dynamics, this mattered. They acted as genuinely informative “social trademarks”. But they were also transitional. Earlier, the knightly class (in the broad sense) so dominated the economy that mere merchants would not be able to afford the “wrong” clothing. Once merchants began to become wealthy, the laws were needed, since the merchants could afford as good, or better, clothing than the knights and the nobility. As the commercial revolution took hold, literacy spread and the state became much more important in providing public order, sumptuary laws became (where they persisted) an inappropriate holdover propping up a structure of privilege which had lost connection with social functioning.

Remind folk of anything we discuss above?

What some interests would like to get away with in IP is well beyond what current technology supports: a trend that is likely to get worse rather than better. IP law and practice needs to work with who actually has control of what and what boundaries are sustainable because of real enforcement capacity given the balance of benefits and people's sense of fair boundaries.

ADDENDA And who thinks charging a 24-year old lives-with-parents gamer A$1.5m for uploading a game sold to him a week early by a retailer by mistake is fair?

FURTHER A web “insider” casts a sceptical eye on some of its effects and implications, including for the value of creative work.

AND ALSO Richard Stallman gives an excellent lecture on the history of copyright law (and why 'intellectual property' is a confusing and distorting notion that does bad things to one's thinking).


  1. Oooh, I loved sumptuary laws... I mean, they stratified colours and furs and everything. But the basic take-home point is that people thought they were stupid and didn't want to follow them. King Knut holding back the tide comes to mind - it really doesn't work.

    Yes, that's a lovely analogy, Lorenzo.

  2. Ta :) I am grateful to skepticlawyer for the comment that sparked it. (We use the "identity theft" analogy when we teach students our Fashion & Dress presentation.)

    And, to be fair, there was a period when they worked: but that was when they fitted people's sense of what you needed to know about someone.