Research Frontiers

This chapter discusses how the SWF framework can be generalized to accommodate variable-population cases and differentiated individual responsibility. The framework, as presented in earlier chapters, assumes a fixed-population setup (the very same individuals exist in all outcomes). Conversely, a variable-population setup allows for individuals who exist in some outcomes but not others. Extending the SWF framework to this case means grappling with the philosophical literature on population ethics—specifically, grappling with the ethical significance of non-identity and deciding how to make ethical comparisons between outcomes with different total population sizes. Earlier chapters also focus solely on the pattern of well-being in outcomes—ignoring that two individuals who are at the same well-being level may be differentially responsible for their condition and thus have unequal ethical claims to a well-being improvement. The economic literature on equality of opportunity (EOp) provides a structure for generalizing the SWF framework to reflect differentiated responsibility.

The Institutional Role of the SWF Framework

This chapter discusses the appropriate role of the SWF framework within governmental institutions. The topic is divided into three parts. First, under what conditions is it legally permissible for governmental officials to employ the framework as a policy-evaluation methodology? The framework provides ethical guidance; what is the legal role of such guidance? Second, to the extent the SWF methodology is used within government, what weight should officials accord to citizen ethical views in specifying the methodology (in choosing a well-being measure, rule for ranking well-being vectors, and uncertainty procedure)? This part of the chapter discusses “empirical social choice.” Third, is it advisable for the framework to be implemented across the full range of governmental programs, or only more narrowly—with respect to tax-and-transfer policy? Some economists argue for the latter position—contending that CBA should be the preferred assessment procedure for non–tax-and-transfer policies.

Which SWF Should We Choose?

The choice of social welfare function (SWF) is an ethical matter. This chapter considers the ethical arguments for and against specific SWFs. It builds upon chapter 3, which presented the leading SWFs and divided the landscape of SWFs into various regions (depending upon their axiomatic properties). First, the chapter discusses the ethical pros and cons of the utilitarian SWF, as compared to continuous-prioritarian SWFs. Next, it engages the ethical debate between continuous-prioritarian SWFs and other non-utilitarian SWFs. The utilitarian SWF has a key advantage under uncertainty, as compared to continuous-prioritarian SWFs: it has an uncertainty procedure that respects both the ex ante Pareto principles and a very plausible axiom of Dominance. Conversely, the utilitarian SWF is insensitive to the distribution of well-being (it violates Pigou-Dalton). Non-utilitarian SWFs do not perform better under uncertainty, as compared to continuous-prioritarian SWFs; but there may be other ethical grounds for preferring these.

The Landscape of SWFs

One key component of the SWF framework is a rule (the SWF) for ranking well-being vectors. This chapter presents the major such rules used by SWF scholars or suggested by the philosophical literature: the utilitarian SWF (which adds up well-being numbers); the continuous-prioritarian family of SWFs (which sums well-being numbers plugged into a strictly increasing and concave transformation function); the leximin SWF; the rank-weighted family of SWFs; and the sufficientist family. The chapter then discusses the key axioms that are used in the literature to categorize SWFs: the relatively uncontroversial axioms of Pareto Indifference, Strong Pareto, and Anonymity, and the more contested axioms of Pigou-Dalton, Separability, and Continuity. The landscape of Paretian, anonymous SWFs can be divided into various regions, depending upon these latter axioms; and the utilitarian, continuous-prioritarian, leximin, rank-weighted, and sufficientist SWFs can be placed within this landscape. Axioms for applying an SWF under uncertainty are also discussed.

Measuring Well-Being

This chapter discusses the well-being measure: a key component of the social welfare function (SWF) framework. This measure, w(·), assigns well-being numbers to individuals in outcomes so as to reflect admissible well-being comparisons (of well-being levels and/or well-being differences). In order for the SWF framework to function, these admissible comparisons must include interpersonal as well as intrapersonal comparisons; the chapter explains why. It then shows how von Neumann/Morgenstern (vNM) utility functions can be used to construct an interpersonally comparable well-being measure that respects individual preferences. A different preference-based well-being measure, the equivalent-income measure, is also reviewed. Although the preference view of well-being is dominant in the SWF literature, w(·) may instead be based upon a non-preference view of well-being, such as an hedonic or objective-good account. The chapter concludes by considering why some economists have been skeptical about interpersonal comparisons.

The Social Welfare Function

This chapter provides a synopsis of the social welfare function (SWF) framework. It presents the key components of the framework, which are then developed in greater detail in subsequent chapters: an interpersonally comparable measure of well-being, which converts each outcome into a list (“vector”) of well-being numbers, one for each person in the population; a rule (the SWF) for ranking well-being vectors, such as the utilitarian SWF or a continuous-prioritarian SWF; and a procedure for ranking policies, understood as probability distributions across outcomes. The chapter then discusses the relation between the SWF methodology and ethics. This framework provides an ethical evaluation of governmental policies—an evaluation that is consequentialist and, specifically, welfarist. Finally, the chapter contrasts the SWF framework with cost-benefit analysis (CBA); and explains how the framework succeeds in producing a well-behaved ranking of outcomes and policies, notwithstanding Arrow’s theorem.

Implementing the SWF Framework

The SWF framework is a fully generic methodology for policy assessment. One important application in the economic literature concerns taxation (termed “optimal tax” scholarship); but the framework is applicable to any type of governmental policy choice. This chapter illustrates the implementation of the SWF framework, using the regulation of fatality risks as a case study. Risk regulation is chosen because this is the major application of cost-benefit analysis (CBA) in the U.S. government. The chapter focuses on the utilitarian and continuous-prioritarian SWFs. Utilitarianism gives priority to the young in reducing fatality risks and to those with higher income. Continuous prioritarianism intensifies the preference for the young, but mitigates or reverse the preference for those with higher income. CBA is significantly different from both utilitarianism and continuous prioritarianism. It markedly intensifies the utilitarian preference for reducing the risks of those with higher incomes; and, unlike both SWF-based approaches, is insensitive to the distribution of policy cost.

Introduction

This chapter is a brief introduction to Measuring Social Welfare. The book provides an overview of the social welfare function (SWF) framework for assessing governmental policy. This introductory chapter motivates the framework via some stylized examples of policy choices involving environmental protection, redistributive taxation, infrastructure planning, climate change, and health policy. These examples illustrate recurring features of policy choice: that policies affect multiple dimensions of individual well-being; have “losers” as well as “winners”; produce a range of well-being changes (losses or gains) to individuals starting at a range of well-being levels; and may require a tradeoff between overall well-being and inequality. The SWF framework is able to take account of these recurring features of policy choice, and is applicable to a wide range of policy domains.