The Sources of Public Choice
In the Wealth of Nations published in 1776, Adam Smith convincingly explains the fundamental importance of freedom for the prosperity of a nation. Incidentally, the year of the publication of the Wealth of Nations coincides with the year of the Declaration of Independence of the United States of America.
Not incidentally, however, Smith’s book has enormously inspired the US Constitution framers in writing the Federalist Papers in the years 1787-88. In Britain, Smith’s home country, the echo of the Wealth was more mixed. Smith has been praised as an anti-mercantilist. However, his commitment to liberty has become less acknowledged. Smith had to convince his compatriot readers against the theories of competing writers. In 1780, only a few years after the publication of the Wealth, Jeremy Bentham published his Introduction to the Principles of Morals and Legislation in which he proposed utilitarianism as a leading doctrine for the British public. Bentham fought for a moral society against phenomena such as usury and prodigality. His follower John Stuart Mill saw utilitarianism as a principle to obtain a better society. He has adopted the ideas of an equal sacrifice of taxation first proposed by Horace and Jean-Baptiste Say (Faccarello 2006). Mill, too, thought that the rich and poor citizens should both contribute to the state, but as utility of income and wealth was assumed to decline with higher income, Mill’s equal sacrifice principle implied a higher taxation of the rich compared with the taxation of the poor. The amount of the total sacrifice, in fact the amount of money to be raised has not been questioned. The expenditure side of the budget was not explicitly considered. Therefore public choice has remained outside economic analysis. Bentham and Mill both had a large influence in Britain. It is due to their publications that utilitarianism became a leading philosophy in nineteenth- up to twentieth-century Britain. As utility was regarded as cardinally measurable, welfare maximization became a policy goal for governments. Welfare maximization had its own justification. It was imposed and hence independent of individuals’ choices. So the spirit of liberty of Adam Smith has been crowded out of public policy.The utilitarian view was contested by three neoclassical economists: Carl Menger of Vienna, Stanley Jevons of London and Leon Walras of Lausanne who succeeded in explaining prices in a market economy as a result of consistent individual market evaluations of private goods around 1870. However, individual market evaluations were alien to the British utilitarians who neglected to notice that the great neoclassical triad has opened a new research programme from private to public goods. The study of the economics of public goods has been exiled to the Continent to Germany, Italy and Sweden.
In Germany Adolph Wagner of Berlin was an important trailblazer. Wagner did not believe much in the relevance of individual choices in the public sector, but he understood the relevance of public goods, which were unknown to the then leading British public finance economists Bentham and Mill (see Blaug 1978). On the one hand, Wagner gave an important signal to the economists of the Italian School of public finance to study public goods on the basis of the marginal utility principle, which they appreciated from reading the publications of Menger, Jevons and Walras. On the other hand, the Italians have already become familiar in terms of quid pro quo for public goods, inherited from
French economists A.-R.-J. Turgot and M.-J.-A.-N. Caritat de Condorcet (Faccarello and Sturn 2010) and from Francesco Ferrara.
Antonio De Viti de Marco (1888) and Ugo Mazzola (1890) concluded that the principle of choice had to be extended from private goods to public goods: this was a new world, quite different from the utilitarian world of Bentham and Mill. But how should choice be organized as public goods were consumed collectively? All individuals had to contribute according to their joint evaluation, said De Viti de Marco, and each individual according to his or her own evaluation in terms of his or her “godimento” (pleasure) as Mazzola has suggested.
How could this bridge be built? De Viti de Marco thought that the actual state should become something like a cooperative. Mazzola thought that coercion was necessary (Fausto 2006: 77). However, coercion would disregard the idea of individual evaluation: in fact the heart of the whole Italian scientific venture. No doubt; marginal evaluation had to be voluntary. At this critical point the Swedish economist Knut Wicksell made the decisive contribution. He linked the idea of quid pro quo inherited from Turgot, Concorcet, Ferrara and Sax and combined it with the institutions of the parliamentary decision process:The principle is as such nothing, but the interest principle, the well known principle of reciprocity between contribution and return whose range of application and usefulness I try to extend in two directions: On the one hand, following the example of Sax and his followers, towards applying the modern view of marginal utility and of subjective evaluation consistently to public services and private individuals’ response payments... and - herein I could not indeed quote a forerunner - by linking the principle... with the form which tax power has adopted nowadays... namely the parliamentary principle of tax approval and by trying to establish the conditions under which the principle of quid pro quo... can be brought into effect automatically. (Wicksell 1896: vi-vii, author’s translation)
To Wicksell it was necessary that individuals’ consent was voluntary. So the link to neoclassical economics was established, and the result was indeed a choice. Wicksell effectively closed the gap from private to public goods; the problem which was left open by Menger, Jevons and Walras, who only considered private goods. The importance of linking choice from private to public goods has never been fully grasped by the established Anglo-Saxon School of public finance until today. James Mirrlees, for example, admittedly acknowledges the existence of public goods in Diamond and Mirrlees (1971a, 1971b).
He spared an amount of resources for public goods in his theory of optimal taxation. However, he leaves the question unanswered how public goods should be evaluated and how many public goods are desired (Blankart 2014).The light disseminated by the Swedish and Italian economists remained unnoticed until Richard Musgrave popularized their views in his 1939 paper “Voluntary exchange theory of public economy”. However, the fundamental attack on the welfare approach of the Anglo-Saxon School was due to the publication of Lionel Robbins’s book An Essay on the Nature and Significance of Economic Science (London 1932 [1935]). Robbins argued that the economists should stop studying normative issues about the welfare of a society, but rather analyse positively how individuals use scarce means to achieve alternative ends. So Robbins has repatriated choice in the centre of economics.
Robbins gave confidence to economists studying positive economics in the new world of public choice. Duncan Black (1948) and Kenneth Arrow (1951 [1963]) were the first who tested the new trail. They extended the study of free markets to the study of free democracies. Their first results were a great surprise. Black found that separate individual orderings, when compared pairwise, might end in a cycle. He popularized earlier but forgotten results by the French scientist Marie-Jean-Antoine-Nicolas Caritat de Condorcet (1785) and the English logician Charles Dodgson (1876).
Arrow (1951 [1963]) has proven that, when voters have to choose between three or more distinct alternatives, individual rankings cannot be converted in a consistent set of collectively ranked preferences if individuals’ domain of alternatives is unrestricted, if dictatorship is to be absent, if Pareto efficiency and independence of irrelevant alternatives have to hold.
If one or more of the four conditions are relaxed, a consistent ordering may not be obtained. The most famous invention is due to Black (1948). He found that if it is possible to array the alternatives in such a way that each voter’s preferences exhibit single-peakedness, then cycles could be avoided and in fact the median voter decides for the whole community (median voter theorem).
Is Black’s condition realistic? Single- peakedness seems plausible if the community has to decide on whether to have one, two or three football fields. The football fans, on the one hand, are confronted with the football grumpies, on the other. But single-peakedness is much less plausible when the community has to decide how to use the last plot of land in a local community. Should the plot be used for a football field, an ecological park or a building site? There is no reason whatsoever that the voters have single peaked preferences (Mueller 2003).Gerald Kramer (1973) asked the question: how much homogeneity of preferences is necessary to avoid a cycle. He found that only when individuals have non-crossing personal indifference curves can a cycle be avoided. This is a very strong condition. For non-crossing can be expected for one individual, but not among several individuals. Non-crossing between several individuals in fact implies unanimity between these individuals. However, with unanimity, preferences are simply juxtaposed and not really aggregated. An aggregation of preferences leading us to a social welfare function as aimed at by Abram Bergson (1938) or Paul A. Samuelson (1947), indicating what is good and what is bad for the society, is not possible. This again shows the importance of Arrow’s impossibility theorem. An implication of these results is that we should not rely too much on political and the rationality of collective choice. In fact, we should rather avoid voting processes.
Black has shown what single majority rule cannot produce. But what can single majority rule do? Kenneth O. May (1952) found that, under two alternatives, the sum of yes (+1) and no (-1) will generate a group decision function which is a simple majority rule if and only if it satisfies the following four conditions:
1. Decisiveness (that is, no ambiguity);
2. positive responsiveness (of the outcome when either the pros or the cons increase);
3. anonymity (among ballots of voters); and
4.
neutrality (on particular issues) (Mueller 2003: 133-6).If we add transitivity as an additional condition it follows from May’s theorem that no voting rule will satisfy all five conditions, not even the simple majority rule.
Neutrality means that all issues are of the same importance, anonymity that all individuals are equally affected. Dropping these two conditions opens a multidimensional issue space and hence opportunities for trading votes (logrolling) between issues of unequal importance and between individuals with different intensities of preferences. Decisiveness will break down and majority rule will lead to cycles. A further consequence of this analysis is that logrolling opens no escape from Arrow’s impossibility theorem (Bernholz 1978).
Under these new conditions another set of very strict conditions has to be met to restore a stable outcome of majority rule. Charles Plott (1967) has shown that a majority rule equilibrium exists if it is a maximum for one and only one voter while the other voters are evenly paired off so that any change improving the position for one individual is balanced by another individual being made worse off (see Mueller 2003).