Priyodorshi Banerjee

B

Some of these theories (as a cluster, often called choice theory) proved to be extremely influential. By the end of the 1960s, they had succeeded in establishing themselves as one of the main pillars of a newly coalesced dominant school of thought, neoclassical economics. Typically, such a theory used fairly sophisticated mathematics, displayed analytical elegance, and often derived implications which were relatively simple and intuitive, at least within the accepted framework of economic thought. Also, as far as the fundamental motivational issue was concerned, the typical assumption was that an agent received satisfaction (or utility or payoff) if and only if he consumed physical goods and services. A singular advantage of this generation of theories was their structural sparseness and consequent analytical sharpness.

Even from the end of the 1950s, however, various criticisms began to emerge, both of the mathematical choice-theoretic structure as a class, as well as of individual theories. The critiques typically focused on the framework of assumptions in a given theory. Since any set of assumptions

0$12+$5

received sustained attention from other social scientists. To that extent, asking how a theoretical decision-making unit’s economic choices are affected by its senses of identity certainly constitutes a significant departure from the perspective of neoclassical econo mics, even if it does not open substantive new avenues for the social sciences as a whole. Two attendant questions are immediately raised, however: first, is there a substantial methodological departure, and second, do the answers and insights generated differ enormously from those arising from a more traditional analytical approach?

Methodological Approach

Methodologically, the authors’ approach is as follows. First, they introduce the idea that there are two broad types of goods which agents can consume and derive satisfaction from – one that satisfies basic tastes, and another that is not so basic, and whose value derives from norms, i e, social rules regarding behaviour. Second, they expand the idea of the utility function to take into account this second type of good. One part of the utility function is exactly as in traditional models, and describes how goods satisfying basic tastes contribute to satisfaction, while a new term is additively introduced to describe the relationship between utility and consumption of goods satisfying non-basic tastes. The innovation thus lies in the introduction of this second term, which follows the form of what in the technical literature is called a “loss function”. In other words, there is a predefined optimal level of consumption, and any deviation from that level leads to a loss of utility. This is, of course, quite different from the other (traditional) part of the utility function, where there is no such predefined optimum, and indeed, more consumption always leads to higher levels of utility. The predefined optimum is interpreted as the norm or the socially specified level, as given by an agent’s identity grouping. So the closer an agent is to the consumption level specified by the identity marker, the higher the satisfaction that is received.

Introducing these kinds of loss functions as part of the utility function is an ical, interpretation of identity and norms and their relationship to utility, and by extension, to decision-making. However, this particular way of thinking of two kinds of goods, with distinct roles and effects, which respond to different environmental stimuli, does not appear to sit well with the idea that identity or other non-economic variables are as important as the ones that traditional theories focused on. No justification is provided as to why there are two (and only two) types of tastes nor why goods satisfying such tastes have such different impacts upon utility. There is also no discussion on how to empirically identify such tastes, or how to determine when a good satisfies one type of taste versus the other. Despite these weaknesses, the approach does have the advantage of providing a simple inter pretive link between traditional utility theory and the notion of identity. It also has the advantage of being a relatively simple technical extension of received structure, thereby allowing comparability and retaining tractability.

As far as the explicatory or predictive aspects of their ideas are concerned, the rewards seem to be more in the domain of the promised than of the present. The authors propose that their approach is (more) intellectually satisfying (than others), and that it will lead to “more reliable” and “sturdier” economic models (pp 7-8). Thus, while a significant part of the book (especially parts two and three) comprises applications, the focus seems to be more on the presentation of simple illustrative examples, rather than detailed analysis. The authors apply the framework developed in part one to various problems in the economic analysis of education, race, gender and organisations.

Overall, these four chapters contain many interesting examples, but tend to get a little repetitive. The most interesting is chapter five on organisations. Organisations can be viewed as informationally closed collections of economic agents who despite differences, and possibly conflict, of interest among themselves, have voluntarily associated with each other for some common purpose. Most commonly, the organisational form receiving theoretical attention from economists is the firm. There has been rapid development