Arnold Kling

Lessig on Copyright

PRINT
Defending Imperfect Markets... Comment of the Week, 04-09-03...

In 1998, Congress extended copyright terms on both new and existing works. Lawrence Lessig tried unsuccessfully to have this extension overturned by the Supreme Court.

In this interview, Lessig explains the weakness of the economic argument for the copyright extension. First, even for new works, a longer copyright term confers very little incentive at the margin for creators. Moreover,


With respect to existing works, by definition it's not creating any incentives to produce because you're just granting a windfall to works that have already been produced. No matter what we do, Gershwin will not produce anything else. That extension is nothing more than what economists would call "rent-seeking," using property as a cover.

For Discussion. Do you think that those who were pushing for copyright extension would have accepted the economic logic which says that extending copyright terms for existing works has no marginal incentive effects, so that the extension should only apply to new works?



COMMENTS (6 to date)
Robert Musil writes:

In my opinion Professor Lessig's argument is misleading at best. Also in my opinion, it was properly ignored by the Supreme Court in the Constitutional context. But even in the context of legislative policy, I think it goes way off track.

Elsewhere I have written about the "tragedy of the commons" aspect of copyright termination, an aspect which appears to be rather willfully ignored by the proponents of universal short-term copyrights. That whole set of considerations - in my opinion - makes Professor Lessig's argument something a of an economic aside. In fact, the argument seems to be a legal one - derived from some old (and, I believe, misconstrued) Supreme Court language - and I think the Court pointed that out rather pointedly.

Further, Professor Lessig carefully phrases this argument to deny that there will be an incentive to "create" works that have already been created - which makes his argument sound like a tautology. But "creation" of intellectual property rights such as copyrights are not so simple as that. For example, "creation" of the Mickey Mouse franchise - the whole "world" of Mickey Mouse - didn't end with the "creation" of all existing Mickey Mouse cartoons. In fact, creation of meaningful Mickey Mouse cartoon ended long ago. Mickey's current role in popular culture is now mostly derived from the Disney Theme Parks and other Disney products - that is, "derivative works." But the Disney Theme Parks, for example, really just provide a custom-made context for the existing work. Disney's continuing "creation" of Mickey Mouse would likely end if Mickey image became common property. Calling such efforts "marketing" or"maintenance" of an existing work rather than "creation" of a new one is a legal fiction - but an economic error. It's odd that Professor Lessig's current legislative proposal takes particular aim at copyright holders who do NOT "maintain" their copyrights.

I note that Professor Lessig also seems suspicious of much copyright protection of derivative works.

If one wants to believe that Professor Lessig believes his own argument (and it is no criticism of a lawyer that he doesn't), one is led to doubt that Professor Lessig has spent any time with tenants of rented apartments. Do they invest in their apartments the way condo owners do? Of course not. Renters will make only improvements whose value expires when the lease terminates. And at the end of the lease, the apartment has generally been rather substantially "uncreated." That's why there are security deposits under real estate leases.

Paul Zrimsek writes:

Whatever the merits of Lessig's economic argument, it's at war with his constitutional argument. The latter assumes-- for no reason which I have ever seen Lessig or anyone else explain-- that Congress will go on indefinitely passing extension after extension. If this were to happen, it would (pace his economic argument) create new incentives to produce. In other words, the 1998 extension is considered a one-off, unforeseeable and therefore incapable of acting as an incentive, for purposes of economic analysis, but as the harbinger of an apparently inexorable trend for purposes of constitutional analysis.

Bernard Yomtov writes:

I fail to see the contradiction.

Congress has repeatedly extended copyright protection. 1998 was far from the only case. That is the reason Lessig and others think it likely that this will continue. You may disagree, but given the history, the belief seems sensible.

As far as incentive effects, the argument is that extending copyright further has virtually no incentive effect. What is the present value of the extension? Virtually zero.

As far as Mickey Mouse goes, that is one example. Interestingly, though, it suggests that it is possible to do new things with old intellectual property. But this can happen only if the would-be doers have access to the property which in most cases the extension blocks.

Mark Seecof writes:

Robert Musil's comment is vacuous: any new works Disney might create using Mickey Mouse will get new copyright terms. The question is whether Disney's *old* works should carry indefinite protection, including a right to forbid "derivative" works (a "right" which itself is mostly rent-seeking, not creative incentive). Of course, the derivative-works right is what Musil wants to protect, to prevent what he calls a "tragedy of the [IP] commons." He suggests that Disney would flush poor Mickey down the drain if it lost the copyright to Steamboat Willie, because other folks might (assuming they somehow overcame Disney's immortal trademark claims) draw (gasp) unflattering pictures of Mickey. This might not be so (after all, Disney persists in developing its own Snow White properties without a copyright on the original fairy tale), but if Disney did drop Mickey, it would replace him with a new, newly copyrighted, character--so the change in the law would have *incentivized* Disney's creativity! It is true that it *might* be more economically efficient *for Disney* to stick to Mickey, but it's an open question whether that would increase welfare *overall,* which is of course the purpose of copyright laws.

I continue to be amazed by the prevalence of "halfway" economic analyses that presume that maximizing welfare for a monopolist (by enabling him to extract the maximum consumer surplus) also maximizes the public welfare (welfare overall). This presumption only holds when creativity by anyone other than the monopolist is ignored. Public welfare will increase when consumers can spread their surplus around, fuelling increased productivity through competition.

Opposed to Musil's "tragedy" argument one can imagine *better* choices in, say, DVD's of old cartoons--we have no reason to assume that Disney's staff are better than any possible competition at authoring cartoon anthologies or supplying critical commentaries [note 1]. Indeed, we have every reason to think that Disney's current monopoly (gained as Lessig irrefutably shows by rent-seeking politics long after Walt's initial creativity had reaped its promised reward) makes it slothful and thus reduces creativity.

Note 1: supposedly one is free to criticize another's copyrighted work, even if that means copying small portions. As a practical matter, one of the best ways to criticize a motion picture is to synchronize "comments" to a presentation. This is presently done by film professors in dimmed lecture halls, or by putting "commentary tracks" on DVD's. DVD's, that is, which carry a copy of the motion picture under criticism. Neither Disney nor anyone else will authorize DVD copies of their works with (other than fawning) critical commentary tracks, so "synchronized criticism" is virtually unobtainable (the supply of seats in academic showings being very, very small and correspondinly pricey). It's hard to see what benefit accrues to the public (as opposed to, say, Disney) from these restrictions on older works decades after the copyright terms originally offered as a quid-pro-quo for their creation have expired.

Robert Musil writes:

I hadn't seen this last post before.

It's hard to know what to do with a comment as off-base as "any new works Disney might create using Mickey Mouse will get new copyright terms." Of course, any "new work" that anyone creates gets "new copyright terms." But that's almost completely irrelevant. The issue is not just "old" versus "new" works. That's just one of Lessig's shell games, one that was properly and tartly rejected by the Court.

Disney wants to retain its copyright in the character of Mickey Mouse - a copyright that allows Disney (for example) to stop any else from creating Mickey Mouse cartoons at all. That allows Disney to control public associations with Mickey, which has a lot of value.

Of course copyright exists to maximize consumer welfare. That the copyright holder may obtain some market power is not to the contrary. Nobody even proposed that Congress has legislated copyright laws to enrich monopolists before the Disney-haters came on the scene.

The rest of Mr. Seecof's post is even more confused. His introduction of "fair use" considerations in connection with the right to criticise is a complete distration. The First Amendment trumps the copyright laws to that extent, but so what? Nothing there helps in understanding copyright economics.

His assertion that stripping a copyright owner of his property "incentivises" the former owner to be more creative by making replacement property is just perverse - almost pathological. By that theory the Czar's cossacks "incentived" Russian jews to be more creative every time the cossacks burned the jewish homes in a pogrom.

One could go on and on, but what's the point?

That Lessig's arguments are leading to rants such as Mr. Seecof's is powerful evidence that Lessig himself is way off base. But there's a lot better evidence of that - starting with the Court's essentially contemptuous and correct dismissal of Lessig's arguments.

Mark Seecof writes:

It seems that Robert Musil thinks that copyright is some kind of natural property right which would endure forever unless Congress "stripped" the owner of it. This contrasts with the view, expressly adopted by the Constitution, that copyright should be a temporary right created by law, as a reward for creative effort. Were Congress to abstain from extending the copyright on Mickey, it would expire in accordance with the terms on which it was granted. The Eldred case concerned whether Congress was required to so abstain. Since the people who created works copyrighted long ago have long since finished their creative effort (for those works), the question is whether Congress has power to give a new copyright (for that is what the extended term is) to people (corporations) *now* who did not create the work to be protected.

(Note that even with CTEA, in the USA a copyright is just supposed to disappear quietly when the term is up. This is not "stripping" the ex-copyright-owner of anything; if it were he could sue for redress under the 5th Amendment "takings" clause. It's analogous to a lease on a house expiring. The tenant may have to move out, but he hasn't been "stripped"--he has just used up the value of his lease.)

The limited-term view of copyright, to which I subscribe, does not object to the copyright holder's *temporary* monopoly on copying the work in question. Indeed, it is that monopoly which provides a mechanism for society to pay for creative work. Consider other mechanisms which have been proposed, such as: eliminating copyright, but awarding prizes from public funds for works deemed valuable by a committee. That scheme, and so many others in the same spirit, would likely fail like any "funding for the arts" scheme: it would be captured by the politically active and used to reward their supporters regardless of a work's value to the public. By contrast, the copyright scheme gives the copyright owner a chance to extract from the public as much or as little as his work is worth to them, over a given period of time, in competition with various substitutes (in the broad sense: the public might prefer, say, golf lessons to purchasing an overpriced book, or might prefer Bugs Bunny cartoons to Donald Duck for aesthetic or financial reasons).

In this scheme, long-run consumer welfare increases because creators will invest in creation when they can expect to reap a reward. If there were no copyright, they would likely create less because mere copiers would reap the rewards of the creators' efforts. But with copyright, the public will pay for creative works in rough proportion to their overall value (of course, the copyright holder might misprice his works and force a sub-optimal level of consumption; since he would likely suffer as much or more than the public from that, we may expect the market overall to minimize such problems).

However, this view of copyright does not favor extending copyright terms past their original lengths. (1) The public offers a copyright term as a quid-pro-quo for creative work. The copyright holder's is given a temporary monopoly to enable him to extract a suitable price for his work, based on its value to the public once it is offered to them. The grant of temporary monopoly is just an apolitical pricing mechanism. Since the actual value of the prospective monopoly will depend on the value of the work to the public without reference to any particular political action, the public need not pay for the abuse and inefficiency likely in a more particularized payment scheme. (2) Once the proposed bargain (so-and-so-many years' monopoly in return for the work) has been accepted (the work published), the public is under no obligation to pay more. By analogy, having agreed to trade "the pick of the litter" for a stud service, why should the owner of the dam surrender any more of the offspring, regardless of how pretty they might be? Indeed, for the public to pay more to politically-active copyright holders, years later, undermines the whole scheme, which was to insulate the public from the excess costs of political payment schemes.

The history of CTEA suggests that it was was pure political rent-seeking. It gave copyright terms to people who did not create anything, and who had (this is key) received the full value for which the creative works were bargained away long ago.

Now, suppose that Congress thought too few works were being created... what could it do? Well, it might try increasing the reward for creativity, by offering a longer copyright term. Would that be unconstitutional? For all of Musil's sniping at Lessig (and his unworthy suggestion that I am merely Lessig's dupe) I don't think Lessig claimed that it would (although some of the amici offered arguments for that view). I think Congress could constitutionally offer a longer copyright term for new works. It might be unwise, but I don't think it would be unconstitutional. However, that is NOT the question of interest. The question of interest is whether Congress can lengthen the terms of existing copyrights--copyrights on works created long ago. Musil doesn't want to draw the distinction, but it can hardly be obscured. The public offered certain copyright terms to creators long ago, terms that the public (well, Congress, anyway) considered fair and which, N.B., the creators decided to accept! So with respect to the oldest works, the public's bill has been paid. By what right do current owners of old copyrights demand *more* money (years)?

This brings us to two other issues. One is Musil's confused notion about a tragedy of commons in IP economics. Look, "tragedy of the commons" is shorthand for the way a resource may be destroyed when people have a right to consume it but no right to exclude others from doing so. If the resource is finite, it will likely disappear. Note that people who create new copyrightable works DO NOT CONSUME ANY EXISTING RESOURCE. Musil's "tragedy" simply does not exist. Musil complains that without a copyright holder enforcing a ban on derivative works, too many people would CREATE new works. I repeat: not destroy existing works, but create new ones. Since no resource is being consumed, no "T of the C" can occur. Musil wants to prevent the creation of new works--that is, he wants to prevent expansion of a resource (the supply of creative works for public consumption). Note that copyright ensures that the new resource has an owner (for a time) so it cannot be destroyed by overuse. Even when ownership passes to the public, the resource still cannot be overused, because making a new copy DOES NOT REDUCE OR INJURE THE ORIGINAL. It is IMPOSSIBLE to overconsume an IP resource by copying it. (It is quite possible to make creation of a new IP resource uneconomic by copying without payment to the creator.)

Of course, Musil does have a logic, even though he reaches the wrong conclusion. He thinks the "resource" in question is "public appetite for a complex of related creative works," e.g., Disney's "continuing creation" of Mickey-related theme-park attractions and tchotchkes and so-forth. Musil thinks that, for example, new non-Disney representations of Mickey Mouse would diminish the "public appetite" in question (perhaps by a route like this one: a Mickey knock-off could show Mickey doing something bad. The public would then refuse to view Disney's Mickey stuff because they would fear that all Mickeys are bad).

Well, wrong premises lead to wrong conclusions. In the real world "public appetite" for something is not a resource waiting to be consumed. Musil confuses supply and demand. The public demands entertainment, and the purveyors of Mickey supply it. If they supply it well they can reap a reward; the copyright system is designed to *facilitate* that reaping.

Musil compounds his error by supposing that a supply of non-Disney Mickey material must destroy the value of Disney's Mickey stuff. Of course, non-Disney Mickey material might be a *substitute* for Disney Mickey stuff, thus driving down the price (toward marginal cost of reproduction for copyright-expired works). But just as Lawry's The Prime Rib restaurant can prosper on the same street (La Cienega Blvd.) in Beverly Hills as Burger King, Disney could make good money selling their Mickey stuff even if other people were free to offer inferior (by Musil's estimation) Mickey material. A claim that we should preserve Disney's monopoly on any piece of Mickey stuff, *after* we have already paid full price for it (that is, after its original copyright term has expired) is like a claim that we should shut down all the non-Lawry's prime-rib restaurants in the USA.

In the case of restaurants, we all understand that competition and choice serve the public best. We protect Lawry's right to the name "Lawry's" to promote competition: the name makes it possible for the public to seek out a particular sort of meal (and if Lawry's quality declines, the public can go elsewhere). In the same spirit, we would (we already do) protect "Disney's" name. But this has little to do with copyright. If Disney's copyright on Mickey expired when it was supposed to, Disney could still produce Disney-branded Mickey stuff. They could compete with other brands of Mickey stuff the way Lawry's competes with other brands of prime rib. Disney would do fine; indeed, as the expert on Mickey creation it would likely dominate the Mickey market.

Of course the price of ex-copyright Mickey stuff would fall. So what? The public has already paid for the original creative effort! The price of new stuff might fall too, since it would compete with the old stuff. Again, so what? The proposed cure for that (giving Disney a monopoly) is worse than the disease (open competition in Mickey material). And, though Musil dismisses the distinction, *new* Mickey stuff would be copyrighted, so the new competition to serve "Mickey demand" would be orderly... new Mickey creators would have to produce better work than Disney to outsell Disney on copyrighted material. Musil's wacky "tragedy" argument doesn't come anywhere near to analyzing the real world, as I pointed out before by referring to Snow White. Disney can't stop competitors from selling alternative Snow White depictions... yet Disney still makes zillions from its Snow White. If Musil's "tarnishment" (my word) theory were correct that should not happen!

Musil's "Cossack" analogy is deeply flawed, and if his faint suggestion of anti-Semitism on my part was intentional then I resent it. I don't propose to rob anyone. Since a copyright term is a quid-pro-quo for a creative work, once it has expired after due enforcement (for 56 or 95 or however-many years) the work is properly regarded as bought and paid for. If I buy a sackful of potatoes from a farmer, that may well incentivize him to plant some more, especially if I've made a standing offer to pay him for a new sack (like the copyright law does). If I'm willing to pay the farmer TWICE for the same sack of potatoes, he may well invest less in planting. Why should he grow more when he can just sell me the old sackful over and over again? Copyright holders like Disney may see political contributions as cheaper than creative investments (especially after risk-adjustment, since new creative works are much riskier than old, proven, ones). Since a new cartoon might cost several million dollars and the favor of a Congressman seems to retail for a few tens of thousands, holders of old copyrights undoubtedly calculate along these lines every year.

I'll try to come back and address Musil's error with regard to fair use (which I think makes him misunderstand my film-commentary example).

Comments for this entry have been closed
Return to top