Law and economics: realism or democracy?

• Author: Smith, Henry E.
• Position: Twenty-Seventh Annual National Federalist Society Student Symposium

Is law and economics anti-democratic? One hears complaints from many quarters that law and economics is a form of technocracy that cuts off legitimate debate and suppresses other important values that people hold dear. (1) On this view, law and economics privileges efficiency and focuses on quantifiable values to the exclusion of other, less measurable values that could have found expression through the political process. These concerns are central to debates in areas ranging from environmental protection to intellectual property. The irony in these complaints is that they are offered by commentators who are heirs of the legal realists, many of whom would in the same breath decry excessive formalism and applaud judicial sensitivity to policy. There may not be an inherent contradiction here, but there is a tension in practice.

Law and economics and democracy are not enemies, but I contend that legal realism--or its lingering aftershocks--causes law and economics to be more technocratic and less democratic than necessary. While legal realism as a movement itself may be dead, it rules us from the grave. As the saying goes, "We are all realists now." (2) There is nothing wrong with law-and-economics-inspired theories as theories--or with legal realism as a theory for that matter. Analyzing law and legal relations in their smallest parts and considering micro incentive effects (to the extent data is available) are worthy exercises, but without some sensitivity to institutional detail and competence, the tendency is to substitute the wisdom of the analyzing expert, especially in courts and agencies, for the collective wisdom emerging either from democracy or tradition. (3)

Many movements in legal thought draw on legal realism, (4) and law and economics is no exception. Coase's articles on the FCC (5) and social cost (6) are hyperrealist in their assumptions about property, especially in their adoption of the most extreme version of the bundle of rights conception of property. (7) In the bundle of rights conception, property has no content on its own but instead emerges from policy-driven decisions about the actions that people might take. (8) Things are merely a backdrop to this fine-grained analysis of potentially conflicting activities, and rights to exclude from things have no particular status as a starting point.

These assumptions were understandable in light of Coase's goal of demonstrating that, in a world of positive transaction costs, it matters how entitlements are assigned. But when it comes to using Coase's insights, his hyperrealist assumptions have been allowed to steal the show. (9) In Coase's analysis of nuisance, we expect judges to figure out ex post which of the conflicting parties should be awarded each stick in the bundle of rights. (10) And in making these decisions, the questions of "who invaded what" or "who caused what to whom" do no work at all. (11) In contrast to traditional and everyday notions of property as a right to things that is good against the world, Coasean agnosticism about causation leads one to see both the trampling animals and the trampled-upon crops as the cause of conflict. And under this conception, one is to ask whether fists or noses cause punches, or, for that matter, which are the cheapest cost avoiders. (12) None of this accords with non-economic intuition. (13) Although causal agnosticism is a useful theoretical construct and fine as far as it goes, it does not go very far: for transaction-cost reasons--not to mention basic moral reasons--causation is unidirectional. We have made ex ante decisions about what counts as an invasion, (14) and absent good reasons--and sometimes good reasons exist--we should stick to those decisions.

Now it might be thought that this technocratic tendency in fine-grained analysis is specific to property. Such a view seems unlikely when we consider that Coase and many of the bundle theorists are basically treating property as dissolving into torts. Echoes of this are to be found in current debates over intellectual property, in which skeptics of intellectually property would like to see more of a tort or regulatory regime than a property regime in IP. (15) In any case, the legal realist strand that became law and economics tackled torts relatively early, (16) and torts has featured much more largely in law and economics scholarship than has property proper. Torts seems tailor-made for the type of technical approach that legal realist-style law and economics offers. This Essay, therefore, will concentrate on torts and argue that even here, on the best terrain for legal realist law and economics, the technocratic tendency has led to similar, if less dramatic, results.

A word about technocracy and democracy is in order. I am using "democracy" and "technocracy" in a special sense, one in which they potentially conflict. In arguing against "technocracy" I am not opposing well-informed decision making of all sorts. Instead, I am making the narrower point that modes of legal decision making that ask judges to use a great deal of contextual information have their inherent limits. The argument is based on the presence of information costs, a limit that could be regarded as "technocratic," but in a different sense. The information cost argument here is at a meta level: In evaluating a system of decision making, one might want to use all available information and techniques, even if these reveal limits, within the system, on our ability to use information.

In other words, I am making a meta-level, realist-style argument for a certain degree of formalism in ordinary legal decision making, where formalism is (relative) invariance to contextual information. (17) By contrast, combating formalism and thereby disregarding these limits to the use of contextual information is quite characteristic of legal realism in practice. Law and economics is only one branch of the tree whose trunk is legal realism proper, and many of the criticisms of thoroughgoing antiformalism apply to these other approaches as well. But today's topic is law and economics, and more particularly antiformalist post-realist law and economics. Unconstrained contextual decision making tends to put more power in the hands of decision makers within the system--often unelected judges in the case of the common law--and this power tends to conflict with democracy to the extent that such derisions are difficult to reverse in the political process. Moreover, the information cost considerations for which I argue tend to point towards greater reliance on everyday morality, associated with the people, generally. Highly refined all-things-considered utilitarian decision making tends to conflict both with this popular morality and the congruent, more modest decision making that can be economically justified at the higher, systemic level.

There is nothing inherent in analyzing legal relations at the systemic level that would necessarily lead to technocracy, so it is worth considering why law and economics, and the economic analysis of torts in particular, partakes so heavily in those aspects of legal realism that emphasize expert decision making.

Like some strands of legal realism, the economic analysis of torts tends to emphasize, if not elevate, the role of the judge. This elevation is somewhat ironic in light of the role of juries in tort law, as opposed to administrative decision making. The common law was fertile ground for the first generation of law and economics because bilateral interactions, including litigated disputes, are easier to model than the complex simultaneous interaction of many parties in taxation or regulation. Thus, in the sphere of primary actors, the bilateral impersonal tort-like conflict between the activities of A and B, or the more personal contract between A and B, make torts and contracts easier to understand than property, which is often about impersonal interactions between owners and the "rest of the world." In property, multiple parties may have some claim on a single resource and multiple systems may overlap. Some interactions, such as the tragedy of the commons, were amenable to economic tools, (18) but the bilateral interactions at the center of torts and contracts made these areas a top priority in law and economics. Further, in a world with zero transaction costs, solving every problem would be costless; (19) it takes some effort to remember that the choice of analytical unit itself has transaction cost implications in the real world. Common-law litigation thus looks more amenable to economic analysis.

From there, it was a short step to focus on judicial decision making and, in older law and economics, the kinds of cost-benefit analysis that judges might undertake. When analyzing an interaction, the benefits of fine grain are apparent--they are the point of the exercise--but the costs are less apparent. True, any analysis should take "administrative costs" into account. But as is quite apparent with the bundle of rights, the cheapness of the baseline in rem right to exclude is easy to overlook. In the case of property, the convenience of the baseline stems in part from the diffuse nature of the processing costs. To the extent that these benefits of the baseline inhere in the system as a whole and are usually left implicit, they are especially easy to ignore in fine-grained analysis that puts a premium on articulated rationality.

It is somewhat ironic that law and economics overlooks the system-wide benefits of simplicity, because economists have long known that global systems show significant local variation. Partial equilibrium and general equilibrium are two very different things. Law and economics rarely rests on a general equilibrium analysis. But partial analyses must be taken with a grain of salt; it is characteristic of complex systems that a subpart may not share properties with the whole. (20) Similarly, the theory of the second best...