An Evolutionary Approach to Social Welfare

Christian Sartorius

• Publisher: Taylor & Francis
• Year: 2003
• eBook ISBN: 9781134358786
• Edition: 1
• Topic: Business
• Subtopic: Business General
• Index: Business

While not evident immediately, social norms and values play a crucial role in the theory of social choice. In the first half of the twentieth century, the special acknowledgement by economic theory of the autonomy of individuals and their subjective view of the world had led to the serious problem that socially acceptable decisions could not be mad

1: Introduction

1.1: The need for a new approach to social welfare

Since ancient times humans, and philosophers in particular, have been concerned with the forces moving the universe and with the potential of human beings to interfere with them. With regard to prevailing social and religious goals, this potential could subsequently be used to derive normative conclusions concerning the behavior of members of the respective groups. Inspired by the Enlightenment movement, human reason was granted a new role as the eminent means by which man could understand the universe and improve his condition. A prominent example for this change of mentality is presented in Thomas Hobbes’s (1651) Leviathan: while once the state had been viewed as an earthly approximation of the eternal order, with the city of man modeled on the city of God, now it came to be seen as a mutually beneficial arrangement—the social contract—among men aimed at protecting the natural rights and self-interest of each. Coinciding with the end of absolutism and feudalism, and with the beginning of the Industrial Revolution, this emphasis on man’s own faculties and responsibility to further his survival and well-being called for new accounts of political and economic reality and for new visions. In The Wealth of Nations (1776), Adam Smith provided the first major account of this kind: essentially, he describes industrialization as a prominent case of division of labor which allows for a dramatic increase in productivity and, thus, wealth. For this to happen the existence of property rights and of competitive markets is the only, but crucial, prerequisite. (Any further intervention of the state is even considered detrimental.) Without protection of private property, the necessary accumulation of capital would not take place, while markets are needed to exchange products in so far as they exceed the manufacturers’ own demand. By ensuring that supply largely matches demand, competition turns the persistent human desire for improving one’s condition into a socially beneficial agency by pitting one person’s drive for self-betterment against another’s. It is this regulation of the economy as the unintended outcome of the competitive struggle for self-betterment to which the metaphor of the ‘invisible hand’ alludes.

While The Wealth of Nations is no doubt a masterpiece of economic analysis, its political implications are not completely unambiguous. Superficially, the prospect of a prospering economy is very promising—especially against the backdrop of the majority of the population living in miserable conditions: the state just has to reduce, or give up, the major part of its interventions within the economy and enforce a few very essential institutions, and soon all people will succeed in improving their lot—guided solely by the ‘invisible hand.’¹ Since there seem to be winners only, all measures leading to such an economy should find unanimous support. However, closer inspection reveals that, in fact, there are numerous losers as well. First, it may safely be assumed that at least some people—for instance, the holders of privileges in pre-industrial societies or the beneficiaries of the absence of institutions protecting private property—will find themselves benefiting from the old regime more than from the new. Second, the benefit a single person may draw from this change turns out to differ widely. Some people may become very rich while others earn just enough to sustain their own or their families’ lives and may not even have better prospects. For the first time, this aspect of income, or welfare, distribution became a matter of rigorous investigation in the influential works of David Ricardo (1817) and John Stuart Mill (1848). A third source of potential loss, even in a prospering economy, refers to the crucial role of competition. Joseph Schumpeter (1942) was among the first scholars to explicitly acknowledge the crucial role of innovative activities for the remarkable performance of capitalist economies. By his famous notion of ‘creative destruction,’ at the same time, he points out that every innovation carries in itself the potential to devalue existing productive or human capital. Thus, in order for the majority of people to benefit from the innovation-induced growth of their economy, some individuals have to incur some losses in terms of decreasing salaries,² losing their jobs or their companies going bankrupt. Due to the fundamental uncertainty concerning the nature of the innovation, it is impossible to predict who will have to suffer and by how much. It is clear, however, that some people will face substantial costs. Moreover, their willingness to bear these costs is a necessary prerequisite for all people to benefit from the resulting economic growth.

While this list is certainly not exhaustive, it shows some of the major reasons as to why, even in a growing economy, a significant number of people will incur substantial losses. For the individuals concerned, their condition will in no way improve by their knowing that for the whole of society gains by far outweigh losses. On the contrary, from the purely individual perspective, it is even quite plausible that the divergence between their own misfortune and other people’s good fortune should make them perceive their own situation as even worse. For this reason, accounting for relative changes in individual well-being should be expected to play an important role in the assessment of welfare as yielded by a given state of or change within an economy. However, while the complexity of the interactions and, thus, the individuals’ mutual influences on their respective well-being increased constantly over the last centuries, economists seemed to be unable to account for this change by the development of effective procedures for the aggregation of social welfare from individual well-being. The history of economic thought—from utilitarianism to social choice theory—even testifies to a rising uneasiness and incapability of dealing with the problem of interpersonal comparisons of well-being, eventually culminating in Arrow’s (1951) impossibility theorem. While summarizing the latter development, sections 1.2 and 1.3 also allow for the identification of two of its major causes. First, from the methodological perspective, there is a marked tendency to render the individual the central unit of analysis. Moreover, while with the rise of subjectivism the individual is granted complete autonomy over her decisions and actions, the interdependence of different individuals’ decisions and of the preferences on which these decisions are based—that is, the social influence on them—is basically rejected. This closely relates to the second point, that questions of allocation are clearly given priority over questions concerning the distribution of goods. While the allocation of goods according to individual preferences is far from being a trivial problem, there is at least no need for the invocation of normative assumptions that could not easily be shared by a large majority of humans: efficient allocation in the sense that, for a given set of individuals, maximum total utility is derived from the suitable allocation of a given set of goods is generally considered as a positive contribution to welfare. The latter argument can obviously not be maintained for most questions of distribution. Stances as different as capitalism and egalitarianism are cases in point. While most people basically accept a worsening of their personal situation as long as the realized costs appear to be sufficiently small, temporary in nature, and the chance of being compensated for them by future gains seems to be realistic, the extent and the quality of tolerated divergence of living conditions varies widely between cultures. The latter fact may indeed appear somewhat discouraging for an economist trying to develop a (general) theory that describes the contribution of states or changes of distribution within a group to the welfare of this group.

Nevertheless, for people belonging to a specific cultural context, these figures are often characterized by a high degree of homogeneity. The latter fact appears less surprising once the mutual interaction between the members of a given group is recognized as giving rise to their reasonable expectation that personal misfortune does not last forever but, rather, is a kind of precondition for the potential rise of good luck. It is the nature of this social contract that people feel committed to it as long as the incurred burden consistently relates to the potential benefits to be realized. Evidently, the social values underlying and constituting a social contract could easily serve as measures for at least culturally specific comparisons of costs and benefits incurred by different members of the same society. Why, one is tempted to ask, didn’t economists make use of it as a natural means for the interpersonal comparison of well-being and for the aggregation of individual well-being to social welfare? The answer, again, is twofold and relates to what was said above. First, without being able to explain how different societies arrive at their respective sets of social values or normative principles, it would be impossible to develop more than just a series of culture-specific theories, each of them based on a unique set of normative assumptions not shared by the members of all other societies. Moreover, being an emergent property of the interaction of the members of a community, the social values underlying any social contract can be assessed only with difficulties even greater than those that already led to the rejection of introspection and the adoption of subjectivism and revealed preferences as the guiding methodological principles in economics. Utilitarianism and John Rawls’s (1971) Theory of Justice are typical examples of theories hampered by exactly the latter two kinds of problems. Section 1.4 will elaborate on the fundamentals of interpersonal comparisons of well-being and on the way in which these fundamentals are reflected in two of the best-known principles underlying the aggregation of social welfare.

Social welfare and its reconstruction from individual well-being will of course be the core issues of the work to follow. However, inspired by the conclusions drawn from the understanding gained in sections 1.2 to 1.4, the approach employed in this work will differ markedly from most, if not all, other approaches to this subject. First, social interaction and interdependence will be acknowledged as a necessary precondition for the interpersonal comparison of well-being and, thus, for the aggregation of social welfare. However, it will also be acknowledged that drawing on specific normative assumptions will never give rise to a generally accepted theory. Obviously, the two conclusions appear to be at least to some extent contradictory. How can this contradiction be resolved and the two conclusions reconciled? The key to the solution of this problem relates to the conjecture that different societies employing their respective sets of norms and social values do not coexist independently. Rather, it will be presumed that groups or societies have to compete for a given set of resources and that the sets of norms and values they hold influences their respective probability to prevail or to be displaced—with obvious consequences for the welfare in the respective groups. A more detailed outline of this evolutionary approach to social welfare is provided in section 1.5. Most remarkably, this approach is a positive approach: Instead of relying on social norms and values as specific normative assumptions underlying the analysis of every single social context, it includes social norms and values as variables that explicitly influence their holders’ welfare. Doing so, it includes distributive effects in social welfare and, at the same time, bridges the gap between different groups.

1.2: From utilitarianism to positivist subjectivism

Once—in the course of the Enlightenment—man had come to believe that human reason was supposed to play the most eminent role as a means to the understanding and control of his environment and, thus, to the improvement of his condition, he also needed a moral guideline telling him to which ends these means were to be directed. This guideline was provided by utilitarianism and Jeremy Bentham as its leading proponent. Basically, Bentham believed that the maximization of pleasure and the minimization of pain are the driving forces and determinants of human action. Here, pain could also be interpreted as a sanction imposed by another agent intending to influence one’s own action. Moreover, he regarded pleasure-seeking and pain-avoidance as the basis for a normative criterion for human behavior. In contrast to other, particularly earlier, ethical theories, utilitarians are consequentialists; that is, they focus upon the consequences of an individual’s action rather than upon its intrinsic value or upon the motives of the agent. It is also important that the emphasis on pleasure-seeking and pain-avoidance does not imply egoism, the view that a person should pursue her own (narrow) self-interest even at the expense of others. The hedonic calculus employed by utilitarians included—and had to do so, from their perspective—the pleasure and pain of everyone likely to be affected by a given act. This logic immediately becomes evident as underlying Bentham’s (1789) Introduction to the Principles of Morals and Legislation: a legislator seeking to maximize the happiness of the whole community could make it unprofitable for one person to harm another by laying down certain penalties for different acts and, as an effect, by creating an identity of interest between both parties. Altogether, the calculus was supposed to yield the ‘greatest pleasure of the greatest number.’

For Bentham, pleasures and pains, of course, were of various kinds and different in quality, so much so that he invested considerable effort in their classification (1789:ch. 5). As Warke (2000:8–11) shows, the use of the term ‘utility’ (which he adopted from Hume) for Bentham implied irreducible multidimensionality. Moreover, he basically assumed that the perception of pleasures or pains belonging to the same category differs among individuals. As a consequence, aggregate utility for a group of n persons each experiencing m different types of pleasure and pain results in an object expanding in an nxm-dimensional space. According to the ‘index number problem,’ it is impossible to derive a ranking or a general measuring rod for the valuation of a series of states that differ in more than one independent quality. Thus, in order to arrive at a decision relevant for the entire group, nxm different expressions of pleasure or pain have to be reduced to a single value, the aggregate utility, by employing some intrapersonal (over all m) and interpersonal (over all n) utility weights. As to the latter part of the problem, Bentham, who used to distinguish thirty-two circumstances (such as age, sex, and education) influencing the individual’s sensibility in terms of pleasure and pain (1789:ch. 6), was well aware of the difficulties raised by the attempt to assess and to compare the utilities of persons that differ in so many respects. He therefore ruled out all empirical problems by adopting an ethical assumption, that is, by imposing a strictly egalitarian rule that became the standard assumption shared by all modern utilitarians: ‘Everybody to count for one, nobody for more than one’ (Bentham 1827, as quoted in Warke 2000:fn. 8). As a consequence, the suitable weights on every individual’s utility for making ethical judgements or legislative enactments are equal across all individuals concerned. With regard to intrapersonal comparison of utility, the situation is more difficult—and this is why, for Bentham, the distinction between different qualities of happiness is a crucial one: assuming equivalence between different sources of happiness, utilitarians would have to accept even malevolent behavior if the pleasure drawn from the malevolent act by the perpetrator exceeded the pain suffered by the victim. Only the distinction between different qualities of happiness makes it possible to specifically impose sanctions to suppress one (e.g. mischievousness or malicious glee) kind of happiness as opposed to another. But even this possibility rests on the assumption that their relative expressions within the average individual can be determined in terms of average weights. Just this act of determination, however, relies on the subjective perception of individuals such that even strictly impartial utilitarians could be expected to arrive at different conclusions concerning representative weights for various pleasures and pains. Obviously, Bentham was aware of this problem; while he even proposed ‘the faintest of any [sensation] that can be distinguished’ (Bentham 1782 as quoted in Warke 2000:11) to represent a kind of unit of intensity, he frankly acknowledged that measurements of and in terms of these units must remain imprecise. Contrary to common opinion, Bentham did not believe that either aggregate or individual utility could be measured precisely; he did not use these concepts except for qualitative investigation.

In economics, the half century following the 1860s was characterized by a major change in the understanding of value—from the classical ‘labor theory of value’ to the neoclassical notion of marginal utility. While classical economists were quite aware of the phenomenon later known as decreasing marginal utility (see Stigler 1950:310), they had as yet been incapable of using this notion to reconcile the utility derived from the consumption of a good with the demand for it and its market price. This changed significantly after William Stanley Jevons (1871), Carl Menger (1871), and Léon Walras (1874) marked the outset of the ‘marginalist revolution’ with the almost simultaneous, and independent, publication of their books. Together with constrained maximization, marginal utility called for formal analysis and, thereby, for a one-dimensional variable as the basis for algebraic operations. Among the three marginalists mentioned above, Jevons was most explicit about the concept of utility he applied. He thereby adopted a kind of intermediate position between the classical conception and the view adopted by other neoclassical economists: as Warke (2000:12) concludes, Jevons, an ardent admirer of Bentham, accepted as definitive these same elements of value, yet managed to measure in a determinate single dimension the utility flow from an act of consumption.’ In Jevons’s (1871) own words this transition manifests itself as follows: adopting the Benthamite perspective, ‘utility must be considered as measured by, or even as actually identical with, the addition made to a person’s happiness. It is a convenient name for the aggregate of the favourable balance of feeling produced—the sum of the pleasure created and the pain prevented’ (Jevons 1871:53f). When, however, it comes to economic calculus:

it is convenient to transfer our attention as soon as possible to the physical objects or actions which are the sources to us of pleasures and pains […and to] employ the word utility to denote the abstract quality whereby an object serves our purposes and becomes entitled to rank as a commodity.

(Jevons 1871:44f)

Jevons avoided the problems arising from weighing various components of utility by associating with each type of commodity consumed only a single kind of pleasure. Moreover, for the consumption of different kinds of goods he considered all pleasures as homogeneous.

While one-dimensional homogeneous utility may be suitable for the solution of simple economic problems relating to the consumption of ordinary goods, it is not sufficient for providing answers to ethical questions. As we may recall, this was the major reason for Bentham to develop his taxonomy of pleasures and pains. Since Jevons was nevertheless concerned with moral issues, he had to use a different approach: to account for the possibility that utility may be drawn from doing one’s duty as well as from consuming an ordinary good, and to rule out the case that the latter may exceed the former and that, thus, the person could be allowed to give her consumption priority over her duties, he introduced a two-level lexicographical utility scale. He assumed a ‘lower calculus’ that after incurring a given cost (e.g. in terms of labor) allows people to satisfy their ordinary, self-directed wants and desires to the largest extent possible, and thus to accumulate wealth. At the same time, a ‘higher calculus’ determining moral right and wrong would indicate how this wealth and the corresponding costs are to be divided among all members of the society (Jevons 1871:32). It is the most convenient property of this lexicographic approach that by simply assuming ‘moral indifference’ for a given situation, normative questions can easily be faded out.

This leads to a third aspect of welfare in which Jevons departed significantly from Bentham: decisions of social relevance usually require a balancing of the individual contributions to that decision. In the context of moral decisions, both Bentham and Jevons assumed a kind of natural (= innate) inclination that made people comply with given moral rules. While these rules may not be identical for all persons belonging to a given community, there will be a given set of rules for any person, and, as a rule, the (higher) utility calculus will turn out to make people obey the rules regardless of other, private, preferences. This is different with regard to ordinary decisions related to consumption, for instance. Bentham, though assuming a certain degree of variability of individual susceptibility, adopted the egalitarian stance and deliberately set all utilities equal. In contrast, for Jevons, these differences of susceptibility and the empirical problem of scrutinizing another person’s mind and of assessing her utility are sufficient reason to repudiate interpersonal utility weights.

Evidently, Jevons’s conception of utility was the result of an almost painful attempt to reconcile his utilitarian persuasion with the practical needs related to his scholarly endeavors. But the need for accessibility to mathematical operations was not the only cause for the change in the meaning of utility. Another cause was a difference in perspective. While Jevons was concerned with utility as the net happiness the individual could derive from consumption of a commodity, Walras and Menger focused more on exchange and on market phenomena. For examp...

Table of contents

• Cover page
• An evolutionary approach to social welfare
• Routledge frontiers of political economy
• Title page
• Copyright page
• Preface
• 1: Introduction
• Part I: Evolution, behavior, and learning
• Part II: Coordination, cooperation, and social welfare
• Notes
• References