Abstract: *Department of Political Science, State University of New York at Stony Brook. 1"Democratic Organization: A Preliminary Mathematical Model," Public Choice 16 (Fall 1973), 17-26. 2Essai sur l'Application de l'Analyse a la Probabilite des Decisions Rendue a la Pluralite des Voix, Paris, 1876. 3The Theory of Committees and Elections, London: Cambridge University Press, esp. pp. 159-178. As Black has pointed out, Condorcet's discovery of the paradox of cyclical majorities, in significant ways, anticipated Arrow's Impossibility Theorem. (See Kenneth Arrow, Social Choice and Individual Values, 2nd Edition, New York: Wiley, 1962, and Duncan Black, "An Examination of Professor Arrow's Impossibility Theorem," Vienna, 1968.) The most extensive treatment of Condorcet's mathematical contributions to political and economic theory is Gilles-Gaston Granger, La Mathematique Sociale du Marguis de Condorcet, Paris: Presses Universitaires de France, 1956.

Abstract: What determines the extent to which a democracy will provide goods and services to its citizens? Differences in taste can be invoked to explain why communities, like individual citizens, consume different menus. And variations in wealth as well as taste may help to explain why one community may elect to spend only half as much as another in the same state on the education of its children. But will the process of majority rule yield an efficient division of resources between private and public goods? When market fail, will government do better? The democratic process portrayed in this paper resembles in certain respects the town meeting procedure used by many New England communities to determine simultaneously the level of educational expenditure and the consequent tax rate. But the paper simplifies many aspects of the democratic process in order to show that the principle of majority rule may encounter difficulties in even the most elementary circumstances. The ballot-box allocation considered in this paper involves equal shares for all in a collectively distributed good financed by a proportional tax. 1 It will be assumed for the most part, that all citizens have

Abstract: This paper has attempted to demonstrate how the bandwagon effect would alter the benefit-cost decision inherent in voting. It was shown that, other things held constant, the existence of a bandwagon would lead to a situation where the favored candidate receives an increased number of votes at each cost. However, it was also noted that when all things are considered, it is most difficult to predict the overall effect of the announcement of a runaway election due to the simultaneous effect of rational abstention.

Abstract: Economists have known for a very long time that most social states have more than one equilibrium. In particular, there is the agreement equilibrium that corresponds to some point on the Pareto optimal frontier, and that represents cooperation among the members of the society. There is a second, well-known equilibrium—the independent adjustment equilibrium—which occurs when, for one reason or another, agreement is not possible. The independent adjustment equilibrium is normally less satisfactory to the parties than the cooperative equilibrium. Ideally, the role of the state is to affect this independent adjustment equilibrium in ways that lead to superior outcomes. Economists, although occasionally dealing with the independent adjustment equilibrium, have normally confined their research to the cooperative equilibrium.

Abstract: In this paper, we have examined the form of special interest benefits as determined by the interaction between interest groups and political parties. We concluded that the party is interested in granting benefits with long expected lives, since in this way the amount which can be appropriated by the party is increased. However, it is often true that long lived benefits have high deadweight losses; this explains the inefficient form in which subsidies are often granted. The theory developed here has several other implications. First, there should be a hierarachy of benefits granted to groups. Second, there is no presumption of consistency in benefit granting legislation, since benefits are sold on a piecemeal basis. Third, one law can benefit several groups, including certain customers of the industry. Fourth, regulatory commissions can themselves become an interest group. Fifth, pressure groups could be formed by political organizations. Sixth, benefits, once granted, should be repealed less often in a two party democracy than in a multiparty democracy. Finally, shorter lived benefits would be favored by politicians or parties which had safer tenure. Our model also has certain implications for optimal term of office of elected officials. In a recent article, Posner (1974) has argued that the current theory of regulation does not predict that expensive (inefficient) methods of conferring benefits would be used; and he argues that the current theory is generally lacking in testable implications. This paper has provided an extension of the current theory, which predicts the form of benefits and does have refutable implications.

Abstract: The empirical tendency of government programs to expand with time has long been known to economists and political scientists. While various explanations have been offered that government is a superior good, for example, or that taxes can be raised during periods of "social crisis" but cannot later be cut back1 these tend to be unsatisfactory: most are not borne out by available evidence,2 and none explain why different programs or governments should grow at different rates. This paper, while by no means trying to solve the entire problem, is an attempt to present and test a logical explanation for this growth, in terms of the decisions of rational voters, as it may have occurred in the United States in recent years.

Abstract: The interdependent utility model developed here should be viewed as a closer approximation to reality than the independent utility function selected by Pareto and honored Allen when he led in its adoption by English-speaking economists. The strengths and weaknesses of the interdependent model are similar to those of the traditional model.

Abstract: In an article that appeared in a recent issue of this journal, Peter Bernholz shows that "logrolling implies the presence of the Arrow-paradox at least for two issues with two alternatives."' In this note I will present an alternate proof of this point, which is simple and direct and covers any number of issues with any number of alternatives. I will also point out that this argument has been made before indeed, it has previously appeared in this journal. Consider several issues, A, B, etc., each including several alternatives; e.g., al, a2, etc., for issue A. Each distinct combination of alternatives, one from each issue, specifies a "platform," or "policy package," or (to use Bernholz's term) a "complex alternative;" I shall use the term "platform." Consider also an odd number N of voters with strong preferences over the platforms. I assume (as does Bernholz) that voters' preferences are "separable" or "independent" with respect to the several issues that is, each voter can establish a preference ordering of the alternatives in any one issue considered alone. For example, with respect to issue A, voter i prefers a1 to a2 regardless of how other issues may be resolved; put otherwise, he prefers any platform including a1 to the otherwise identical platform including a2. (This assumption certainly is not *University of Maryland, Baltimore County.

Abstract: Ever since Anthony Downs published his pioneering work on political action in a democracy, it has been widely acknowledged that assumptions about maximizing behavior can successfully be extended to the public sector. In 1975, eighteen years later, the student of public economics can draw on a large and rapidly growing literature dealing with voters, politicians and bureaucrats whose behavior is analogous to that of maximizing consumers, resource owners or producers in models of the private sector. Those who have followed the development of this literature will be aware, however, that it still lacks integration. Voters or their representatives seldom appear together with bureaucrats in the same article. As a result, the production of public goods fails to be considered in most models of output determination. In much of the recent literature on public choice, the median voter is king and the bureaucrat remains his largely invisible servant. In his recent book, Bureaucracy and Representative Government, William Niskanen [1971] attempts to remedy the situation. While much of the book is devoted to what may be called the theory of the bureau as a firm, Niskanen proceeds in later chapters to combine his conclusions on production with those of a rigorously formulated model of the demand for public goods. On the demand side, he goes beyond the common median voter model, describing a process of government by legislative committee which leads to quite different and novel results.1 Although Niskanen considers both the demand and the supply side, he does not use the integrated analysis throughout the book. In this paper, we show that some of his conclusions cannot be sustained if this is done. A simple diagrammatic

Abstract: The analysis of a city of equals, while admittedly removed from the real world, does yield an important finding: in spite of the presence of increasing returns to scale, the optimal city size is finite. Moreover, the application of the theoretical developments of the theory of clubs to the concept of the co-operative city and to coalitions within the city seems difficult but possible. It may provide valuable insights into programs of public acquisition of urban land and the creation of New Towns, and into the analysis of the urban political process, respectively. Also, I might conjecture that, with this transfer, there is considerable scope for further development of the areas of inter and intra-urban migrations, and of a general theory of urban growth and decay.