ALBERT

Notes: Conclusions The results of the previous section, estimates of a three equation model from 23 observations, must obviously be regarded as tentative. The consistent positive relationship between number of interest groups and size of government observed with changing sets of included independent variables, changing samples of nations, and treating the number of interest groups as either exogenous or codetermined, does imply rather unequivocally that interest groups are able to influence public policies in such a manner as to lead to increased government size. Beyond helping to reinforce this conclusion, the results of the previous section should be regarded as first steps in the development of a model of the polity that can explain participation in the political process by interest groups and citizens as well as the size of government. The two most important variables explaining government size other than the number of interest groups proved to be population and the percentage of the population voting. The consistently negative relationship between relative government size and population is noteworthy since several recent papers have assumed that the only government output is redistribution. The negative relationship, implying that an increase in population leads to a less than proportionate increase in the size of government, shows that government expenditure exhibits a most basic public good characteristic. The percentage of the population voting, which probably is closely related to the proportion of voters with incomes below the median, consistently has a positive and significant impact on the size of government. The Meltzer-Richard hypothesis that greater participation by low income voters leads to more redistribution and greater government size is strongly supported. The inclusion of both the interest group and voter participation variables in the government size equation relies on theories related to redistributive activities. The voter participation variable posits a direct responsiveness of government outcomes to voter preferences through the operation of the median voter theorem, and implies rich-to-poor redistribution. The interest group theory posits increasing government size through the addition to the public weal of expenditures on goods with disproportionate benefits for certain interest groups. Such expenditures have distributional implications since in the absence of government provision the interest groups would either go without the goods or have to provide them themselves. While the theory makes no explicit prediction about the direction of this redistributional flow, since the largest single category of interest groups in most countries by far is industry trade associations, one might expect poor-to-rich redistribution as the most likely consequence of interest group influence. Thus, the possibility exists that the influence of the two variables on the distribution of income might be largely offsetting, while their influence on the size of government is cumulative. Disaggregating the effects of these and other public choice variables is a promising avenue for future research.

Notes: Summary and conclusions What I have proposed for consideration is a view of human behavior which is a complement to both the rational egoist model of individual behavior, and the sociobiological model of species behavior. Cooperative behavior can benefit all members of a group and increase each individual's chances of survival. Rather complicated cooperative behavior patterns are observed in some species, and give testimony to the importance of genetic heritage in explaining behavior. But anyone who has ever observed small children at play must believe that the instincts for selfish action dominate those to cooperate in humans. What humans inherit is a capacity to learn, to discriminate one situation from another, to generalize from past experience. But cooperative behavior among humans is learned. The usual depiction of this learning process by modelers of rational egoistic behavior is to assume that learning takes place within the context of the game. Since learning takes time, a repeated game is required to achieve cooperation. The most successful strategy for teaching one's fellow-player to cooperate appears to be the tit-for-tat strategy. It is interesting how closely Axelrod's (1984) description of tit-for-tat's success in bringing about cooperation parallels the psychologist's description of learning. The cooperative behavior of one player is rewarded by the cooperative behavior of the other. Noncooperative behavior is punished. ‘The emergence, growth, and maintenance of cooperation ... require an individual to be able to recognize another player who has been dealt with before [read stimulus]. They also require that one's prior history of interactions with this player can be remembered, so that a player can be responsive' [read one has a conditioned behavior pattern] (p. 174). Axelrod emphasizes that ‘there is no need to assume that the players are rational. They need not be trying to maximize their rewards. Their strategies may simply reflect standard operating procedures, rules of thumb ... habits ... The actions that players take are not necessarily even conscious choices’ (p. 18, and again at p. 173). Axelrod thus clearly believes that the tit-for-tat strategy as a description of behavior applies to situations, in which the rational portion of the rational egoism assumption is inappropriate. Throughout the book he moves back and forth from examples for which rational egoism in a prisoners' dilemma might fit, and others where it will not. He spends a whole chapter on ‘The Evolution of Cooperation in Biological Systems’ (written with William D. Hamilton, Chapter 5). The parallel between the basic principle of the evolutionary approach, ‘whatever is successful is likely to appear more often in the future’ (p. 169), and the basic principle of behaviorist psychology, ‘positive reinforcement increases the future likelihood of operant behavior,’ is again obvious. Operant conditioning describes the learning process of the individual in adapting to the environment, natural selection describes the evolutionary process of species adaptation (Notterman, 1970: 13). The common thread running through each of these paradigms from selfish gene, through thirsty rat, to homo economicus is egoism. And, I submit, the only assumption essential to a descriptive and predictive science of human behavior is egoism. What then of the rationality assumption? A moment's reflection will reveal that the only use to which we put the rationality assumption is to add precision to the predictions stemming from the egoism postulate, as when we operationalize rational egoism by assuming the individual maximizes an objective function. Now rigor is an important and useful property of a model, so long as it does not come at the expense of realism. Fortunately, the public choice analyst can have both the rigor of mathematical modeling and the realism of assuming only egoistic motivation, if he/she treats the rationality assumption as an as if behavioral assumption. Over 35 years ago, Armen Alchian (1950), answered critics of the profits maximization assumption in the theory of the firm by arguing that in a competitive environment the less profitable firms perish, and the surviving firms adopt decision rules as if they had been consciously trying to maximize profits, whatever the criteria actually employed to make decisions. Competition for survival selects gene structures as if the evolution of the species were maximizing the probability of survival. Where cooperation has significant advantages for all members of a group, the group will adopt mores and laws which condition people to behave as if they were maximizing a function like (1) with θ = 1. Over time, those social institutions for conditioning cooperative behavior will survive that maximize group survival chances. The weaker the social conditioning of cooperative behavior is, the more individual behavior will resemble the maximization of (1) with some θ 〈 1. In the limit, when individual decisions do not impinge of the welfare of others, individual behavior resembles the maximization of (1) as if θ = 0. Thus, behavior can be modeled as if it was motivated by conscious choices to maximize different objective functions, with the specification of the objective function varying to reflect the level of analysis (species, group, individual), the adaptive history of the subjects, and the context of the actions. For most of us this is a comforting thought, since we have been conditioned to working with optimization models. But there is an important difference between as if maximization based on a pure egoism postulate and rational egoism as usually employed. The difference lies in determining what goes into the objective function. A model of man based on as if maximization of a socially conditioned or an evolutionarily molded objective function, is a model of adaptive behavior. Thus, what I am proposing is that the rational egoism postulate be replaced by an adaptive egoism postulate to model human behavior. The difference between these assumptions is similar to the difference between the rational expectations and adaptive expectations assumptions used in economics. To describe an individual's expectations at time t using the rational expectations assumption we look at what happens after t, to describe them under adaptive expectations we look at what happened before t. Rational egoism, as usually modeled, considers only the consequences of different actions, the payoffs in the strategy matrix. When modeling the behavior of an individual, who has played the game several times before, who has been rewarded for choosing certain strategies, and punished for others, this forward looking objective function can give accurate predictions of an individual's actions. For much of what we do in public choice as in economics, this rarefied setting for making choices may come close enough to matching the context in which isolated individuals act to give us reasonable descriptions of human behavior. But when we attempt to explain behavior in more complicated social contexts, as we so often do in public choice, charity, voting, crime, then we shall add to the descriptive power of our models by recognizing that man has not only a future, but also a past. George C. Homans (1958, 1962, 1964, 1967, 1974) has argued for many years that a unified sociological theory can and should be built on behavioralist psychology principles to displace the many discordant theories of modern sociology. Were economics and rational politics to replace rational egoism with adaptive egoism as their fundamental behavioral postulate, the possibility would exist for constructing a common methodological foundation for all of the social sciences. For once interdisciplinary research within the social sciences would not be hampered by basic differences in methodologies. The potential gains in knowledge from cross-fertilization are enormous. Certainly these gains are worth the

Notes: Abstract Constitutional questions have resurfaced in the last few years with regard to the European Union. The Maastricht Treaty is a kind of constitutional reform. The rejection of this treaty by Denmark in its first referendum, and the subsequent debate of the treaty that has taken place, raise questions about both the nature of the European Union, and the process by which its rules (constitution) get revised and approved. Similar questions surround the entry of Austria and the three, additional Scandinavian countries into the Union. Perhaps, no event called the European Union's constitution, or lack thereof, into question more that way in which a new President of the Commission was chosen during the summer of 1994. This paper will reexamine the European Union's structure and procedures from a constitutional perspective. Drawing upon the author's recently completed book, Constitutional Democracy, the paper will review the purpose for the Union's existence, the implications of this purpose for the structure of the Union (federalism versus confederation), the rules for making collective decisions in the Union, citizenship, and in particular, the procedures by which the European Union's “Constitution” is written and revised. The constitutional perspective adopted is taken from public choice. The political institutions of the European Union are examined relative to those rational, self-interested individuals would choose to advance their interests.

Notes: Abstract This paper tests whether the results from standard structure-conduct-performance [SCP] models estimated at the industry level are sensitive to the degree of heterogeneity of the firms in the industries. Industries are separated into homogeneous and heterogeneous categories depending on whether the profit rates of firms within an industry converge on a common value or not. In “homogeneous” industries we find that both the long-run projected returns on assets for the industries and Bureau of Census price-cost-margins are well explained by variables usually included in SCP models, as in particular industry concentration. In contrast, few if any of the usual SCP-model variables are statistically significant in the regressions for heterogeneous industries.