Friday, March 20, 2015

Intellectual Property as Global Public Finance

I wrote the following post for the Balkinization blog symposium for the upcoming Innovation Law Beyond IP 2 conference at Yale Law School, where I will be presenting an early work-in-progress with Daniel Hemel. You can read my post on Balkinization here, and you can see all posts in the symposium here.

The conventional justification for IP is that information is a public good (i.e., it is nonrival and nonexcludable), and making information excludable through IP allows it to be efficiently supplied by private markets. Both sides of this account have been questioned: not all information has the characteristics of a public good or can be made excludable through IP, and propertization is not the only way the state compensates public-goods providers. As Daniel Hemel and I analyzed in Beyond the Patents–Prizes Debate, the state also encourages information production through mechanisms such as tax incentives and direct spending. And one challenge for domestic innovation policy is recognizing that like conventional public finance mechanisms, IP facilitates a transfer from consumers to innovators, and that the off-budget nature of this IP “shadow” tax should not affect the innovation policy choice.

In Intellectual Property as Global Public Finance, Daniel and I examine information production at the global level, where conventional public finance mechanisms are lacking. Many information goods are global public goods (or quasi-public goods), so under the conventional account, global coordination is needed to prevent countries from free-riding on each other’s information production. Global IP treaties such as the TRIPS Agreement help solve this global coordination problem by requiring countries to contribute to the extent that they use the information produced under IP laws (with defection punished by trade sanctions). And in the global context, the off-budget nature of IP laws may be an asset, as it facilitates creation of this stable Nash equilibrium in a way that maps onto very different national public finance regimes.

If this were the full story, one would expect to find little state investment in non-IP innovation mechanisms for which free-riding cannot be prevented. And yet governments at all levels do invest significant resources beyond IP in producing information goods.

Daniel and I offer a number of hypotheses to explain these investments. For example, producing information goods has local production externalities, so nation-states may compete to attract innovative individuals and firms. Relatedly, rent-seeking may cause countries to use information subsidies to circumvent free-trade limits on industrial subsidies, and may cause industry interest groups to lobby for grants and tax credits to extract subsidies from the state. Nonmonetary motivations such as altruism and the pursuit of prestige may supplement these incentives for high-profile goods.

Notably, many of the motivations we describe are supported by the existence of international IP agreements. To the extent that public investment in information production can be translated into patentable inventions and copyrightable works, TRIPS allows countries that produce information goods to appropriate a larger share of the benefits from their products. IP laws thus encourage the provision of grants, prizes, and tax incentives for information production at the domestic level.

Finally, we think the standard account misses the point that global production of an information good depends on how the benefits of that good are distributed across countries, which can occur in three distinct ways:
  1. Aggregate global benefits exceed the cost of producing the information, but domestic benefits are smaller than that cost in every country individually.
  2. Domestic benefits exceed the cost of the information in one—and only one—country, perhaps because demand for the information is geographically localized.
  3. Domestic benefits exceed the cost of the information in multiple countries.
The conventional free-riding narrative is based on category #1, for which the incentives seem to resemble a global prisoners’ dilemma: the world is better off if each country contributes, but each country has an incentive to defect. As noted above, TRIPS offers a partial solution to this collective action problem.

Category #2, however, presents no coordination problem: the one country for which benefits exceed costs will have an incentive to produce the good using domestic innovation policies. This is true even if there are significant spillovers for other countries: market actors do not need to fully internalize the benefits of their investments in order to have sufficient incentives to invest.

And category #3 presents an anti-coordination problem (more akin to “chicken” than a prisoners’ dilemma): too many countries have incentives to invest, so each has some incentive to wait and hope the others will produce the information first. This coordination challenge is markedly different from the challenge traditionally considered by public goods theorists, and it may become more significant as larger numbers of countries increase their capacity for information production. For a chicken-type challenge, one imaginable solution is for all countries but one to sign a treaty committing not to produce the good; the one remaining country would thus have an incentive to produce the good independently (as in category #2). More broadly, we think that chicken-type coordination for information goods deserves future attention.

We do not mean to suggest that the current system of using IP as a form of global public finance for producing information goods results in socially optimal investments. Rather, we think that any realistic system of incentivizing information production involves distortions. Identifying the factors that motivate global investments in information-production despite the appeal of free-riding may enable more robust predictions as to which sorts of information goods are likely to be underproduced both with an IP-based system and without.

No comments:

Post a Comment