Spillovers theory and its conceptual boundaries.

• Author: Frischmann, Brett M.

INTRODUCTION

Wendy Gordon has noted that "most of IP law is concerned with internalizing positive externalities." (1) In two recent articles, Spillovers (2) (with Mark Lemley) and Evaluating the Demsetzian Trend in Copyright Law, (3) I challenge the conventional economic theory of intellectual property and specifically the idea that society ought to use intellectual property systems to internalize externalities when feasible.

The nature of the challenge--or the spillovers theory--can be viewed in two ways. I would frame the challenge as an internal one based on, and consistent with, welfare economics. In his reply to the latter article, economist Harold Demsetz seems to accept this view while critiquing aspects of the analysis. (4) Others, such as economist Anne Barron, have critiqued the articles, suggesting that the spillovers theory is inconsistent with welfare economics, necessarily relies on some other noneconomic social theory yet to be specified, and thus is truly an external challenge to the conventional economic theories of IP. (5)

What is interesting about these responses is how they frame a boundary dispute between economic and other social theories of intellectual property. That such a boundary exists is well understood. (6) What seems worth exploring, for purposes of this Article, is how we arrive at and frame the contours of the boundary through a discussion of spillovers. Claims about what lies on one side or the other of the boundary may turn on assumptions and beliefs that might not hold up on close inspection. For example, consider the reason why Julie Cohen believes spillover theory is unlikely to be valued by economics:

The spillovers argument, however, provides no determinate standard.... In the real world, this objection should not be fatal; indeterminacy does not rule out pragmatic policymaking. Within the epistemological confines of economic analysis of law, however, generalized reliance on "externalities" tends to be perceived as signaling a lack of analytical rigor. The problem, in other words, is not the argument itself, but rather these theorists' inability to provide an answer in the terms that their discipline values most highly. (7) Not surprisingly, I disagree with her view of "the epistemological confines of economic analysis of law" and the perceived need for determinacy. (8) Our disagreement rests, I think, on a version or offshoot of the boundary highlighted above. Determinacy is a requirement imposed by the exigencies of policy making or decision making more generally, but it is not an epistemological requirement imposed by economics or the economic analysis of law. If one employs economic (or any other) methodologies to resolve or guide decision making through, for example, cost-benefit analysis, then one inevitably faces difficult questions about what costs and benefits count. As described below, the spillovers theory raises these questions along the boundary noted above.

In this Article, I reengage this debate and the critiques I have mentioned, and explore the boundary between economic and other social theories of intellectual property. I begin with a brief discussion of the conventional economic theories of intellectual property. Next, I discuss the spillovers theory and various critiques, and reflect on a number of conceptual boundaries that surface in this discussion.

1. A BRIEF DESCRIPTION OF THE CONVENTIONAL ECONOMIC THEORY OF INTELLECTUAL PROPERTY

   The basic economic justification for intellectual property rights is that exclusive rights provide the necessary incentives for private investment in creating intellectual resources. (9) Information resources face a well-known supply-side problem common to public goods (10): the inability to cheaply exclude competitors and nonpaying consumers (free riders) presents a risk to investors perceived ex ante (prior to production), and this risk may lead to undersupply. Essentially, in the absence of intellectual property law, there would be a significant underinvestment in some types of intellectual resources because of the risk that competitors would appropriate the value of the resources. Granting intellectual property rights lessens the costs of exclusion, raises the costs of free riding, encourages licensing, and, as a result, allows the owner to appropriate a greater portion of the surplus generated by the production and distribution of intellectual resources. (11)

   Most economic analyses of intellectual property focus on tradeoffs associated with exclusivity. (12) Exclusivity is a supply-side concern that is relevant to assessing how well markets will function. IP rights improve the supply-side functioning of markets for intellectual products (inventions, works, and so on) as well as those markets further downstream for derivative commercial end products. Though each raises additional complications, the reward, prospect, and commercialization theories of IP (13) all take IP-enabled exclusivity as the relevant means for fixing a supply-side problem. The theories differ largely in terms of where in the supply chain IP-enabled exclusivity is needed and the degree of control/exclusivity needed.

   These theories assume that the market mechanism will best aggregate information regarding demand for such investment. Put in a slightly different way, the theories are premised on the notion that private investment in the production, development, and commercialization of IP subject matter will be allocated efficiently on the basis of expected returns in downstream commercial markets, so long as IP rights are available to provide the necessary exclusivity. (14) This premise connects with the idea, articulated well by Harold Demsetz, that markets efficiently aggregate, process, and respond to information about what people want. In particular, the price mechanism provides a remarkably effective signal to producers about where to direct their investments. (15)

   With a few exceptions, (16) there is very little consideration of this contestable premise, and there is no alternative demand-side theory of what "Progress" we want. (17) If we conclude that maximizing social utility is not the end, then market-based efficiency might not be the appropriate metric for evaluating resource allocation. In other words, if we reject maximizing utility as the overarching objective, then there is no dispositive reason to place trust in the market to effectively measure demand. (18) Moreover, even if we stick to utilitarianism, there are good reasons to question whether willingness to pay is a consistently effective mechanism for assessing demand when information systems are involved because of the prevalence of spillovers. (19)

   The supply-side orientation of the economic theories coincides with more general economic theories of property and especially the idea that an important function of property rights is to internalize externalities. Drawing on earlier work, let me briefly explain the connections between externalities, markets, and property theory. (20)

   Externalities, whether positive or negative, are understood to be an important type of "market failure"--at times defined as the absence of a market. (21) The perceived problem is that externalities generally are not fully factored into a person's decision about whether and how to engage in an activity and consequently may have a distorting effect on market coordination and allocation of resources. (22) That is, too few (many) resources may be allocated to activities that generate positive (negative) externalities because those persons deciding whether and how to allocate resources fail to account for the full range of benefits (costs). If those unaccounted-for benefits (costs) were taken into account--internalized--the actors might behave differently, for example, by reallocating their resources in a more efficient manner. (23)

   The distortion manifests on the supply side in terms of undersupply or, in dynamic terms, reduced incentives to invest in what would otherwise be optimal supply. Of course, as noted above, this problem can occur anywhere in the "supply chain." (24) Note that observing the existence and measuring the magnitude of any distortion would depend on a counterfactual assessment of what would have been and generally assumes complete markets elsewhere in the supply chain. (25) Not surprisingly, empirical evidence measuring distortions in incentives caused by externalities (or free riding) is hard to come by. (26)

   The distortion also manifests on the demand side in terms of lost signals about what consumers want and where investments should be directed. The lost signals characterization follows from the notion of externalities as missing markets or unpriced exchanges, and thus depends on the premise noted earlier about the correspondence between market and social demand.

   To avoid distortions associated with externalities, the standard economic solution is to internalize the externalities by pricing the exchanges or enabling missing markets to operate. How is internalization accomplished? For some time, most economists accepted Pigou's view that the government ought to "intervene" via the tax or regulatory system and force externality-producing agents to fully account for their actions. (27) Thus, those who engage in activities that produce negative (positive) externalities, such as pollution (education), should be taxed (subsidized) at a level that takes into account external effects and thus aligns private and social costs (benefits). (28) In The Problem of Social Cost, (29) Coase challenged the Pigovian tradition and added well-defined property rights to the menu of options for dealing with externalities. (30) By definition (within economics), property rights are perfectly defined only in a world without externalities. (31) Of course, the real world is not only afflicted with transactions costs, but also is awash in imperfectly defined property rights and externalities.

   As...