This article addresses the limitations of the market failure approach to discussions of public goods. The authors'' argument draws specifically from transaction cost economics, which they feel contributes to a better understanding of issue for/against government intervention in the market. Zerbe and McCurdy argue that the case for eliminating market failure through the internalization of externalities is flawed, and that governments should intervene in the marketplace only when they have the ability to lower transaction costs.

Zerbe and McCurdy begin by giving a brief history of the market failure concept, from the traditions of Pigou and Samuelson. Included in this summary is a working definition of market failure from a welfare economics position that is: a circumstance where the pursuit of private interest does not lead to an efficient use of society''s resources or a fair distribution of society''s goods (p.559). This definition refers to the free market''s inability to supply a sufficient amount of public goods, which then necessitates government-led solutions. Accordingly, market failures represent a necessary, but not sufficient condition for government intervention.

Zerbe and McCurdy then explain why market failure analysis is conceptually flawed. The failure of market failure is in fact both the inability of the concept of market failure to address transaction costs and the reliance on consumer preference as an explanatory mechanism. The authors attribute the causes of market failures to externalities, which arise entirely from transaction cost, or the cost of choosing, organizing, negotiating, and entering into contracts. Moreover, since unpriced transaction costs are ubiquitous, this gives rise to a situation in which externalities and hence market failures can be found wherever transactions occur (p. 563). Hence, eliminating transaction cost, in their mind, should be the goal of policy analysts. The transaction cost associated with internalizing the cost of negative externalities will never be lower than the net monetary impact of the externality (p 562). That is, the transaction cost associated with the reduction of a producer''s supply will always be greater than the social/environmental cost of the offending externality and therefore such a remedy would not be efficient. Secondly, Zerrbe nad McCurdy address the difficulty welfare economics has in revealing consumer preference. "The market failure concept is not inherently empirical and as such cannot provide answers to empirical questions" (p. 571). The article then goes on to describe several examples when markets can successfully handle potential externality problems. These examples include lighthouse; land tenancy; bees and crops; and common property.

So when should government intervene in the market place? Efficient and optimal government intervention in the marketplace must begin by examining transaction costs and net benefits, not market failures per se. Since externalities are ubiquitous, using externalities as a basis for intervention can at best be arbitrary- and is in no way guaranteed to be an efficient means of addressing the externality. Anytime government can reduce private transaction costs or its own cost of provision, it should do so regardless of whether or not an externality exists. It need not wait for an appearance of an externality to effect a justification (p. 565). The approach Zerbe and McCUrdy take aims to provide solutions for traditional market failures and takes advantage of government''s ability to lower transaction costs, usuallyl through its power of coercion.