Welfare economics and second-best theory: filling imaginary economic boxes.

• Author: Wagner, Richard E.

Since the beginnings of the efforts of economists to give their discipline scientific grounding, economists have thought their theoretical efforts had relevance for addressing significant public issues. While the classical economists generally supported what Adam Smith described as the "system of natural liberty," those economists also weighed in on numerous issues of public discussion. The tenor and substance of those efforts is set forth wonderfully by Lionel Robbins (1952) and Warren Samuels (1966). While the analytical default setting of those economists was to support the system of natural liberty, they also recognized the value of sound public policy in supporting that system. The classical economists thought that there could be publicly beneficial activities that the system of natural liberty would be unlikely to do well in providing. They also thought that there were activities provided through commercial transactions that could wreak significant effects on bystanders to those transactions. The amount of education acquired within a society was one such candidate (West 1965), with the care of the poor being another (Himmelfarb 1983). In such matters as these, the classical economists engaged in strenuous debate and discussion that served as a forerunner to the development of welfare economics during the 20th century.

As welfare economics developed, it acquired the same formalistic character as the rest of economic theory came to acquire. It is common to describe the growing formalism as reflecting growth in the strength of analytical techniques; however, that growth also narrowed the domain of economic analysis by replacing plausible reasoning with demonstrative reasoning (Polya 1954). With plausible reasoning, models are vehicles to assist thinking about policy issues; however, there is much of relevance to those issues that cannot be collapsed into formal models--particularly judgment, sensibility, and tacit knowledge. In contrast, with demonstrative reasoning the model itself becomes the object of analysis. Policy discourse becomes a debate over models, in contrast to the classical use of models to assist a debate that ramifies well beyond any model. With the shift from plausible to demonstrative reasoning, the classical tradition of policy analysis grounded in a system of natural liberty morphed into policy analysis grounded in a system of unlimited domain for policy action.

That morphing of domains is illuminated lucidly by Meir Kohn's (2004) comparison of value and exchange as providing antipodal orientations for economic analysis, and with Kohn's analysis being examined at length in volume 20 of the Review of Austrian Economics (Wagner 2007). In short, Kohn's depiction of the exchange orientation conforms to the plausible reasoning that characterized the classical system of natural liberty.

In contrast, Kohn's description of the value framework conforms to the demonstrative reasoning that characterizes the unlimited domain of contemporary welfare economics. This distinction between orientations corresponds to Peter Boettke's (2012) distinction between the mainline of economic theorists that extends back to the classical economists and the mainstream that arose late in the 19th century and dominates economic discourse today. The theory of the second best (Lipsey and Lancaster 1956) belongs to this contemporary system of unlimited domain, not to the system of natural liberty, though Davis and Whinston (1965, 1967) seek to locate some point of contact with the system of natural liberty, as does Harberger (1971). John Clapham (1922) explained that the effort to distinguish between industries with increasing returns and those with decreasing returns represented the creation of analytical boxes that could not be filled, and Arthur Pigou (1922) and Dennis Robertson (1924) extended that controversy. When viewed from the mainline of economic discourse, second-best theorizing, along with its welfare economics parent, is an exercise in trying to fill what are imaginary analytical boxes that cannot be filled in any substantive manner.

Welfare Economics and the System of Natural Liberty

The system of natural liberty recognizes that people are seldom so innocently engaged as when they are making money and that seldom has much good come out of the efforts of people who claim to pursue the public's good, to recur to Samuel Johnson and Adam Smith as two early proponents of that system. This system by no means entails an absence of policy activity by governments, as Lionel Robbins (1952) and Warren Samuels (1966) explain in careful detail. It does mean, however, that the system has a default setting oriented toward individual liberty. Violations of liberty are the exceptions and not the rule in such a system of liberty and justice. In contrast, such violations have become the norm over the past century or so, as no longer is there any general presumption against the deployment of political power wherever the possessors of that power choose to deploy it. The domain of the political is unlimited in contemporary welfare economics, in contrast to that domain being circumscribed within the classical version of welfare economics. This unlimited domain arises because statements about welfare are governed by presumptions about postulated preferences and not by presumptions about the requisites for human flourishing within a system of natural liberty when people live together in close geographical proximity (Cropsey 1950). Hence, welfare economics, along with second-best theorizing, becomes dominated by the demonstrative concerns of form rather than by the plausible concerns of substance, as Cropsey (1950), Nutter (1968), and Yeager (1978) explain to similar effect.

The classical economists were well aware that there were activities of general value to nearly everyone that were unlikely to be provided through normal commercial transactions. In contrast to contemporary public goods theory with its dichotomy between private and public goods, the classical theorists exhibited more subtlety and nuance in their analytical efforts. They would not, for instance, argue over whether lighthouses were private goods or public goods. While Ronald Coase (1974) described how the provision of lighthouses was organized in Britain from tolls collected from ships that came into harbor, this situation is irrelevant for the contemporary dichotomy between private and public goods. The British scheme for providing lighthouses neither demolishes ideas about public goods nor does it deny claims about market failure. It is pointless to use lighthouses to make debating points for one side of the contemporary classification or the other. That classification as used in contemporary analysis illustrates the effort to create and fill imaginary analytical...