Welfare Economics and Second-Best Theory

Section 1.1 explains that, when the optimum is “maximizing economic efficiency”¹ and the set of sufficient conditions for that optimum’s achievement is the Pareto-optimal conditions,² The General Theory of Second Best³ concludes that, if one or more of the Pareto-optimal conditions will not be fulfilled, increasing the fulfillment of the remaining conditions will not in general even tend to increase economic efficiency (because individual departures from those conditions [individual Pareto imperfections] will in general be as likely to counteract the misallocative tendencies of the remaining departures to any given extent as to compound them to that extent). Section 1.2 explains why economists use an economic-efficiency-analysis protocol that ignores and is invalidated by The General Theory of Second Best. Section 1.3 lists some of the standard economic-efficiency conclusions this General Theory undermines. Section 1.4 outlines the protocol for economic-efficiency analysis that I claim responds economic-efficiently to this General Theory—a protocol that focuses both on the impact of a choice on the relevant economy’s Pareto imperfections and resource allocations and on the different ways in which such imperfections interact to cause or not cause the various categories of economic inefficiency an economy can generate (most of which conventional economic-efficiency analyses ignore).

This chapter argues that the economics profession has not responded appropriately to The General Theory of Second Best (Lipsey and Lancaster in Rev Econ Stud 24:11–32, 1956). Section 2.1 explains the deficiencies both of two influential arguments for the economic efficiency of ignoring Second Best and of the arguments for the economic efficiency of 6 decision-rules that are alleged to respond appropriately to Second Best. Section 2.1 also argues that—although many articles that do respond to Second Best establish relationships that can inform the protocol for economic-efficiency analysis that is economically efficient—taken as a whole, this valuable literature is deficient in that it (1) ignores most of the categories of economic inefficiency whose magnitudes choices can affect and most of the Pareto imperfections that affect these categories of economic inefficiency and (2) largely ignores the inaccuracy and allocative cost of relevant theoretical and empirical research. Section 2.2 criticizes in broad terms various subsets of the applied-Welfare-Economics literature. Section 2.3 explains that (with only a few limited exceptions) the substantial number of theoretical-and-applied-Welfare-Economics textbooks that I and others have examined either totally ignores Second Best or references and then largely disregards it. Section 2.4 delineates and criticizes 7 additional arguments that economists have claimed justify their and our profession’s ignoring Second Best. Section 2.5 recounts a number of instances in which economists responded not only incorrectly but in doubtful good faith to Second-Best-Theory talks I gave. Finally, Sect. 2.6 delineates 6 reasons that may account for the economics profession’s failure to respond appropriately to Second Best.

This chapter has two sections. After claiming that definitions of such concepts as “the impact of a choice on economic efficiency” should be evaluated by their conformity to professional and popular understanding and their ability to contribute to the analyses in which they will be used, Sect. 3.1 articulates and elaborates on an equivalent-variation-oriented definition of the concept that it argues best satisfies these criteria. Section 3.2 then contrasts this definition with and explains its superiority to the Pareto-superior/Pareto-inferior definition of increases/decreases in economic efficiency and to the definitions of “the impact of a choice on economic efficiency” implicit in the Kaldor-Hicks test for economic efficiency, the Scitovsky test for economic efficiency, and the potentially-Pareto-superior test for economic efficiency.

Section 4.1 states that, on my definition, first-best-allocative-efficiency (FBLE)-analysis protocols and concrete FBLE economic-efficiency analyses proceed on the assumption that the Pareto imperfection whose alteration’s economic efficiency is at issue is the only Pareto imperfection in the relevant economy. Section 4.1 also points out that the FBLE magnitude of any Pareto imperfection is zero. Section 4.2 explains that, on my definitions, second-best-allocative-efficiency (SBLE)-analysis protocols and concrete SBLE analyses are the protocols and analyses that would be economically efficient if all theoretical and empirical research could be executed perfectly accurately and financed at no allocative cost. Section 4.2 also indicates that the SBLE magnitude of any Pareto imperfection is the magnitude of that imperfection that minimizes the amount of economic inefficiency generated in the relevant situation. Section 4.3 states that, on my definitions, third-best-allocatively-efficient (TBLE) protocols for economic-efficiency analysis and TBLE concrete economic-efficiency analyses are, respectively, the protocols and analyses that are ex ante economically efficient, given the inaccuracy of relevant research and the allocative cost and benefits of executing and financing it and assessing its economic-efficiency implications. Section 4.3 also states that the TBLE magnitude of any Pareto imperfection is the magnitude that a TBLE concrete analysis will deem economic-inefficiency-minimizing. Section 4.4 contrasts my definitions of “first-best,” “second-best,” and “third-best” with the usages of two other prominent Second-Best theorists.

This chapter has five sections. Section 5.1 specifies the symbols I use to reference, respectively, each of the seven types of Pareto imperfections an economy may contain. Section 5.2 specifies the symbols I use to reference, respectively, private-benefit, private-cost, (private) profit, allocative-benefit, allocative-cost, and allocative-efficiency figures. Section 5.3 specifies the symbols I use to reference, respectively, three categories of resource-uses. And Sect. 5.4 specifies the symbols I use to reference, respectively, particular categories of resource allocations.

Section 6.1 defines the general concept of a “distortion” in a private-benefit, private-cost, or profit figure, the concept of the “aggregate distortion” in such a private figure, and the concept of the “aggregate percentage-distortion” in such a private figure. Section 6.2 defines three basic categories of profit-distortion formulas: conventional, otherwise-Pareto-perfect (oPp) profit-distortion formulas, onesy (1E) formulas for the “aggregate distortion in the profits yielded by a specified resource allocation,” and step-wise-distortion—(SW)D—and step-wise aggregate profit-distortion—(SW)ΣD(Pπ)—formulas. Section 6.3 indicates the symbols that stand for each of the categories and subcategories of distortions that Sects. 6.1 and 6.2 define. Section 6.4 concretizes the concept of the 1E formula for the aggregate distortion in the profits yielded by a specific category of resource allocation. Section 6.5 concretizes the concept of the (SW)D formula for the aggregate distortion in the profits yielded by any resource allocation by deriving such a formula for the same category of resource allocation on which Sect. 6.4 focuses. Section 6.6 outlines the way in which the 1E and (SW)D formulas previously derived would have to be revised to take account of various complications previously ignored. Section 6.7 explains why the study will assume that it will prove to be TBLE for economic-efficiency analysts to use (SW)D rather than (1E) formulas for any aggregate profit-distortion.

Section 7.1 analyzes the step-wise-monopoly (imperfections-in-seller-price-competition-generated and imperfections-in-QV-investment-competition-generated) distortion in the private benefits that a specifiable unit-output (UO) producer did obtain/would have obtained by selling a specified unit of output it produced/could have produced, that a specifiable QV investor did obtain/would have obtained by using a specified QV investment it created/could have created, and that a specifiable production-process researcher did obtain/would have obtained by using the production-process discovery its specified PPR project did make/would have made. Section 7.1 also analyzes the step-wise-monopoly (SW[M]) distortion that the combination of imperfections in seller (price and QV investment) competition and traditional fair-rate-of-return public-utility-pricing regulation can generate in the private benefits yielded by the creation of some QV investments and the execution of some PPR projects. Section 7.2 analyzes the (SW[M])D in the profits yielded by various categories of economics-marginal resource allocations.

This chapter focuses on the six types of Pareto imperfections other than the imperfections in seller competition with which Chap. 7 is concerned. This chapter does not attempt to be as comprehensive as Chap. 7. Instead, each of its six sections makes some salient comments on (1) the step-wise distortions that all exemplars of one of the six types of Pareto imperfections other than imperfections in seller competition generate in the private benefits, private costs, and profits that members of various functionally-defined categories of resource allocations, respectively, confer on, impose on, and yield the resource-user to which the relevant category of resource allocations allocates resources and (2) the causal or correlational relationship between the magnitudes of the step-wise private-benefit, private-cost, and profit distortions for indicated resource allocations generated by all exemplars of the type of Pareto imperfections on which the section in question is focusing and the magnitudes of its counterparts for one or more of the same step-wise distortions that all exemplars of another type of Pareto imperfections generate.

Section 9.1 criticizes various first-best antitrust-related economic-efficiency-related arguments. Section 9.2 criticizes (1) the standard Law & Economics argument that holding tortfeasors liable in accident and pollution cases will equate the profitability and economic efficiency of all avoidance-move decisions and (2) the conventional first-best analysis of the economic efficiency of shifting from a negligence/contributory-negligence legal regime to a strict-liability/contributory-negligence regime on realistic assumptions about the prevailing legal and social realities. Section 9.3 criticizes the standard argument for the conclusion that too little R&D of all kinds is done from the perspective of economic efficiency, discusses the economic efficiency of various policies that have been proposed to stimulate R&D, and makes some preliminary comments on policies that might be adopted to reduce R&D-related misallocation. Section 9.4 criticizes various first-best-economic-efficiency public-finance-related arguments.

This chapter indicates some of the non-analytic-protocol matters that a government must handle third-best-allocative-efficiently for its overall approach to economic-efficiency prediction/post-diction to be TBLE. The issues it references include (1) whether to legally obligate non-government actors to supply relevant information, whether and how to compensate them for doing so, how to evaluate the information they supply, (2) the criteria to be used to select relevant government employees, the training and supervision to give such employees, (3) the compensation to be paid such employees, the criteria to be used to decide whether to retain, promote, or fire them, (4) whether the employment-contracts of such employees should limit their ability to become employees of, do paid consulting work for, or give paid speeches to specified sets of non-government actors during and after their government service, (5) the identity of the non-government actors that should be entitled to provide information to economic-efficiency analysts and the time-constraints and page-constraints that should be imposed on any such entitled information-providers, (6) the obligations of economic-efficiency analysts to justify their conclusions, and (7) the nature of the process for reviewing economic-efficiency analysts’ conclusions (including the rights of specified parties to challenge those conclusions, the duty of the analysts to respond to challenges, and the obligations of supervisors to justify their ultimate economic-efficiency conclusions).

The conclusion briefly summarizes the content of this study and expresses the hope that the economic-efficiency-analysis protocol it recommends or some superior variant of this protocol others develop will be used and will increase both the economic efficiency and the overall desirability of government decisions.