Rah Efficient IP

On this blog I’ve long favored economic efficiency.

Economic efficiency is our best wide general analysis tool for finding win-win deals that get people what they want. That isn’t everything, but it is a lot. (more)

On this efficiency basis, I’ve defended many controversial policies, such as blackmail or polygamy. But oddly, I seem to elicit the most opposition by defending the mere possibility of efficient intellectual property! A widely held position and one embodied in law today. If you recall, I argued:

Before barbed wire, it make less sense to farm, or to enforce property rights in land against roaming animals. But after barbed wire, farming and land property rights made a lot more sense. … I’m happy to admit that today intellectual property (IP) is not obviously a good idea. Such property can create large “anti-commons” transaction and enforcement costs … Today, it is often better to rely on other social incentives to innovate. … [But] just as farmers developed barbed-wire, someday I expect IP advocates will develop better forms of intellectual property, and better technologies for marking, sharing, and enforcing such property. Using such innovations, I expect we will allow more and stronger intellectual property. … Which, like barbed-wire, will mostly be a good thing. (more)

Brad Delong responded:

Robin Hanson appears to think that people have the right to send killer robots off to hunt down people who use their ideas without paying. Me? I think this is an example of how thinking too much about property rights can madden the mind. (more)

(Scott Sumner says this “mischaracterizes” me; I agree.)

Matt Yglesias responded:

Robin Hanson is apparently the kind of libertarian who believes in government-created monopolies over the use of ideas: … Are we sad that Isaac Newton was unable to patent a method for calculating instantaneous rates of change? Does Hanson think he should be paying royalties to Michael Spence every time he writes about signaling? … The idea that a person, having shared his ideas with the world, now has the right to call the cops and have people arrested for taking inspiration from the idea without paying for a license in advance seems odd. Which is exactly why historically government regulation of idea-copying has been the exception rather than the rule. (more)

Yes IP’s high costs now make us use it sparingly. But as such costs fall, my guess is that efficient economic institutions will eventually include more ways for users to pay creators of innovations. I make no claims, however, about the exact forms such property and payments will take.

To reduce transaction costs, property rights may expire after a time, and both “usage” and “authorship” may be evaluated at large crude granularities, rather than “every time he writes about [Spence-style] signaling.” There may be random auditing of innovation usage, and folks may buy access to large bundles owned together by those who worked on related innovations. I don’t know if paying for access will be done before or after usage. I also don’t know if such property will be enforced by government monopoly or private law – perhaps people will voluntarily opt into property rights regimes.

What I do know is that enormous value is at stake in getting good innovation incentives and access, value that can probably be increased by better property rights. Economies show a weak a long-term tendency to adopt more efficient institutions, and that tendency is mostly a very good thing.

GD Star Rating
Tagged as:
Trackback URL:
  • Wonks Anonymous

    Could you link to Delong? And who is Scott?

  • matt

    You are thinking too abstractly. In reality the patent system is more of a lottery than property right granting institution. Typically, they get hundreds of applications for a single “invention”, winning a patent doesn’t mean you invented anything, or have any moral claim to your idea, it means your application was processed first. Lottery winners then sell their tickets to mafia groups who can squeeze companies into paying them. This is the system people are debating, this is why they think you are crazy.

    Do you think inventions of the future will be more unique? Do you think that the patent system will be better able to tell what is a unique invention? These are the issues.

    • Douglas Knight

      It is interesting how many commenters on this subject believe that all patents are software patents. It makes to extrapolate, without much certainty, from what one knows, but everyone knows about drug patents.

  • The argument seems to rely on two empirical bets. First that all or many people are motivated to innovate by money. If you look at numerous salient examples – Wright Brothers, Issac Newton, etc – they just weren’t.

    Second that the payoff to everyone is greater than a weak-IP situation. I think monopolies on ideas block developments that could build upon them, while the monopolist makes a return on investment. For instance, would we be better off if nobody could have done DNA work until the original IP ‘owner’ had extracted 50 years of monopoly rent? While they were doing that, is it a safe bet they’d be motivated to innovate?

    • Jeffrey Soreff

      Intellectual property is NOT efficient. The marginal cost to the consumer of strongly enforced intellectual property does NOT match the (tiny) actual marginal cost of copying the information. Yeah, to the extent that innovation is economically motivated (and, as Ruffles and mtraven have noted, this may well be a smaller factor than economists might suppose) it is efficient for innovators to be rewarded by some mechanism – but it is by no means clear that IP is anywhere close to the most efficient choice.

  • The trend in the real world is exactly the opposite — removal of IP barriers is leading to an enormous amount of production and innovation in the open source and open content worlds. I’ve said many times that if I was an academic economist I would be jumping all over these phenomena, exactly because it seems to violate the received models of economics (eg, that people will only produce when paid).

    What motivates people in these contexts is an interesting open question. But it is clear that property rights has nothing to do with it. So, at minimum, you need to actually show, with reference to the real world rather than economic theory, that more IP will lead to more production.

  • Russell Wallace

    The reason for the negative reaction to your posts about IP is that IP law as it currently stands destroys far more value than it takes. (Granted that this is not true of every form of intellectual property, and a drastic cutback in IP law would be preferable to its total abolition, still total abolition would be an improvement on the current state of affairs.)

    Now if you say that some future better form of IP law will not be so destructive, well, without knowing anything about the form such hypothetical future laws might take, I can’t say one way or the other. If you have proposals on how to make IP law less destructive, even if such proposals require hypothetical future technology to implement, that would be interesting.

    But as long as most of the text of your post consists of abstract advocacy of IP without comment on specific forms, criticism will be both appropriate and forthcoming.

    • lemmy caution

      I agree.

  • K

    By “efficiency” I assume you refer to the Pareto type? But how can you argue in favor of pareto efficiency if I’m not consulted on whether I consider to be fair the terms of my exclusion from the use of an idea that I might have come up with myself? I want to be consulted *first* before you negotiate away my freedom on my behalf.

  • David C

    I agree with some of Russell Wallace’s points. It would be better for you to be more specific, but at the same time, I’m surprised you’ve gotten so many negative arguments for simply stating that you expect IP will be drastically improved. What’s most interesting is that nobody seems to be arguing why they think IP law won’t improve.

    • Re: “nobody seems to be arguing why they think IP law won’t improve.”

      IP law will improve – by becoming less constraining and restrictive. Due to:

      Votes – consumers want their MP3. Technology – is on the side of the users.

  • rapscallion

    Economic efficiency makes no sense as a criterion. How can anything ever be economically inefficient relative to a possible counterfactual?

  • Killer robots? Perhaps Brad was experimenting with his medications.

    It seems to me the devil’s in the details here; I could imagine truly enormous transaction costs. And those transaction costs will almost entirely be felt in the early days as the ideas are congealing – i.e. inside your window.

  • The usual problem with IP is that it costs too much and you don’t get a chance to evaluate it before you buy it and see how good it is, and once you have paid for it there is the reluctance to discard it if you find out it isn’t worth what you paid for (which is non-refundable).

    What would be a better system would be if you could use any and all IP that you wanted to, and then after some period of time, payments would be made based on the value generated by your use of that IP. If you don’t generate any value because the IP sucks, no payment to the maker of the IP. If you do generate a lot of value from the IP, then the maker of the IP gets a cut.

    There would be a tremendous incentive to generate the best and most productive IP, and tremendous incentive to utilize the best and most productive IP. The owners of legacy and crappy IP would not like this system.

    How to implement it? The problem is that any prospective payment system is something that is susceptible to cheating. It should be possible to implement an “IP rights manager”, that keeps track of IP usage and the value generated from that IP, and then calculates a fair cut for the owner of the IP. If most value is generated through the use of IP, then IP rights managers could keep track of it. Maybe it would start out as an artificial currency, like Linden dollars, but would eventually have a conversion factor.

    Something like this could include negative payments for crappy IP that wastes time and doesn’t meet advertised performance. Spam would be a negative payment too. Marketing already is the largest part of the cost of bringing products to market. The costs to implement the IP rights manager could be charged to marketing. There could even be positive payments to try out IP, and if it is successful it would be used more and generate value-added which the maker of the IP would get a cut.

    I think the problem is that people would be greedy and would game and cheat the system (on both sides) such that it couldn’t work.

  • Re: “Yes IP’s high costs now make us use it sparingly. But as such costs fall, my guess is that efficient economic institutions will eventually include more ways for users to pay creators of innovations.”

    Sharing will become increasingly easy, more and more people will be inconvenienced by anti-sharing laws, and “internet time” will make the durations of those laws look silly – so the costs of the intellectual monopoly laws to society look set to continue to rise.

    • The high cost of IP is only due to the monopoly power of the entity that controls the IP. The high profit from that monopoly controlled high cost allows the monopoly holder to erect barriers to entry to prevent competition by other holders of other IP.

      That is what M$ did, they had a monopoly on operating systems early on, and they used that monopoly to stifle competition. M$ jiggered around with the OS until products that competed with M$ products didn’t work, for example Lotus123. The only reason that Apple was able to compete was because their hardware and software was incompatible.

      With M$ having a monopoly, it could run the computer software market to maximize its profit while maintaining a monopoly, not to maximize efficiency or innovation. We don’t know what IP was not developed because M$ stifled the competition.

  • Really non-obvious good inventions are typically simple, because of Ockhams razor and the K.I.S.S. principle.

    This means that the real landscape if intellectual “properties” is not amenable to being cut into separate regions like farmland, since new principles, techniques etc. totally change the landscape, and overlap enormously with other smaller intellectual regions.

    The most obvious example of this is all the silly patents where the only “invention” is using a computer to do stuff. And from my personal digging in patents: I am quite tired of all the silly claims which essentially states: “This inventions principle runs on a computer which is a computer.”

  • J Storrs Hall

    The major problem here is that the property model — usufruct monopoly — is poorly suited to ideas (tho it works well for physical objects and land). Surely you can come up with a better paradigm. You want production incentives, but you also want costs of using previous ideas to reflect the actual costs. Several recent studies, including Steven Johnson’s Where Ideas Come From, indicate that free communication of ideas is more important in production of new ones than monetary incentives to innovators.

  • TStockmann

    More bluntly, you have absolutely no evidence that any additional protections will spur useful innovations other than a silly and inaccurate analogy and “common sense.” IP protections “may” expire? The best thing about this discussion is that it reveals a genealogy of morals, from the vaguely apologetic language and conditional language in the U.S. Constitution about this kind of licensing:

    To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries;

    to a notion that the right derives from anything other than demonstrated usefulness. And that leaves aside naked money politics that led last decade to the travesty of extending the period of these exclusive rights for works that had already been produced, the actual creators being in many cases dead and the corporate interests fully rewarded for their initial investment.

  • b

    It seems I’m the 20th person to post the same thing, but perhaps I could offer a brief summary: your statement that stronger IP rights are efficiency-improving is simply an expression of personal faith. Even simple statements like ‘patents are positively correlated to innovation’ are maddeningly difficult to demonstrate empirically, so it’s jarring when somebody goes even further and states that speculative changes to IP law must improve efficiency.

    It’s great that you have personal faith, but don’t be surprised when others have different beliefs based on different information.

  • Vlad Tarko

    I think about IP in the following terms. Consider imagination as a resource. At a certain moment in time there is a given amount of imagination available. What IP does is to influence the direction in which it is used. With no IP, it is generally used to make quasi-identical copies of already existing successful items (these are low-risk, so more of them are made). What IP does is to force innovators to come up with things that are sufficiently different from already existing things (such that they are not covered by the existing patents). From this point of view, IP may be a good thing to have, if we want fast enough innovation in certain areas (e.g. medicine). (I’m not sure how this could be framed in terms of “efficiency”, it’s more about having certain desirable social goals.)

    Consider some examples that illustrate this theory:

    (1) India does not recognize IP when it comes to drugs. As such, it is a great producer of generic drugs, but, as far as I know, no novel drugs have come from India. This is understandable, as the search for new drugs is a costly and uncertain process.

    (2) Open source pieces of software are largely copies of existing successful software (typical example: LibreOffice is a slightly improved clone of Microsoft Office 2003, the same thing is true about open source picture, video & music editors etc.). The notable exceptions seem to be the open source technological platforms (like Linux, Apache, MySQL, Java etc.). There may be a reason for this, as being free gives them a strong competitive edge over proprietary platforms (even if the latter are, perhaps, better quality). Moreover, it may be a desirable thing for platforms to change slower.

    (3) Music since piracy became widespread has not been very innovative. From 1950s till the 1990s something radically new music appeared every decade. No longer so in the past two decades. (There may be an alternative explanation: the increased availability of old music provided strong competition to new acts.)

  • Pingback: Innovation isn’t about progress | The Uncredible Hallq()

  • Pingback: Overcoming Bias : Why I’m Not Libertarian()

  • Poelmo

    Actually, India now has domestically developed drugs. Saying the West’s (especially America’s) draconian IP-system is necessary to develop new drugs is like saying that paying protection money to gangsters is necessary to prevent people getting their legs broken by men with baseball bats. Most scientists and inventors don’t care about IP and royalties, all they seek is the thrill of being the first, the joy of figuring out how something works, the immortality of being in the history books and to just live comfortably while working on their next project, what they don’t seek is billions in profit (they won’t get those anyway in our current system, upper management and the shareholders will). However, we live in a world where clever shareholders and executives have made sure they control all the resources (both physical and intellectual) that scientists and inventors require for research. Then they turn around and say: “see, scientists and inventors need us, and we need money and IP-rights, so money and IP-rights are necessary for research”.

    One more thing, and I’ve said this before: what exactly is fair about making money off of population size? Why should a creator keep charging money for downloads of his ideas, beyond the cost of his research? Why should an American creator get four times as much money as a German creator?