Public Choice Theory: An Introduction
By Julie Novak

Introduction

Traditionally, economists developed theories and advocated policies on the implicit assumption that the government implementing the policy was a ‘benevolent despot’ – a single entity that would, without question and in good faith, routinely implement policies that were in the public interest. This assumption was formalised in the 1950s by Samuelson and Bergson in the form of the ‘social welfare function’ (SWF). According to SWF theory, the government could implement policies in the public interest simply by plugging relevant variables into each individual’s utility function, then aggregating the utilities to derive the SWF. The policy that maximised the SWF was therefore the recommendation to be carried out by government.

However, over time, more and more economists found the SWF approach to be simply inconceivable in real-time politics. It seems impossible for governments to have such a level of cognitive knowledge as to possess files of each individual’s utilities, let alone be able to aggregate every constituent’s economic desires for the purpose of SWF calculation. In addition, questions have arisen as to whether the ‘benevolence’ attribute resembles realistic patterns of political behaviour. Economists increasingly felt that this duality assumption – where participants in private markets are assumed to be self-interested utility maximisers while political players are publicly-interested welfare maximisers – was a highly questionable proposition in real life situations.

The ‘public choice’ school emerged in the late 1940s and 1950s in response to this growing sense that orthodox economics, through its acceptance of governmental benevolence, neglected the importance of the political process in shaping and implementing policy. According to public choice theory, analysis of tax raising, public expenditures and other forms of collective action cannot be made without a more realistic consideration of the political process through which decisions are formulated.

The purpose of this Schools’ Brief is to recite some of the many developments in public choice theory and, in so doing, illustrate how economic tools can be utilised to research a range of complex issues overlooked by traditional economics.

Equilibrium in Majority Voting Systems

Under the majority voting system a measure requires at least 50% of the voting public’s approval before it is passed. It is for this simplicity that majority voting is widely used across the world in political elections, and as part of the decision-making process of companies and other organisations. However, the 18th century French social theorist Condorcet found that there may not be a unique equilibrium, and that the winning issue or candidate will be different depending on the order in which the choices are considered. This result was confirmed by the welfare economist Kenneth Arrow in 1951.

Consider Panel A in Figure 1, which relates to the voting behaviour of three hypothetical voters (x, y, z) over three alternative public good budgets (Large, Medium, Small). Voter x prefers Large to Medium to Small; voter y prefers Small to Large to Medium; whilst voter z prefers Medium to Small to Large. In an election between Large and Medium only, x and y would vote for Large, while z would vote for Medium. Therefore, Large would win by a majority of 2 to 1. Similarly, if an election were held between Medium and Small, Medium would win the majority decision. Logic would imply that the Large budget is preferred to the Small.

However, if an election were to be held between Large and Small, Small would win by a majority of 2 to 1! Here, y and z prefer Small, but only voter x chooses Large. In other words, the consistent notion that Large would be preferred to Small is rebuffed by community preferences. This quirk is known as the ‘majority voting paradox,’ and it implies that the ultimate majority outcome chosen by society depends on the order in which the votes are taken.

Duncan Black (1948) analysed this problem in the context of voting behaviour in small committees. He found that in order to avoid the paradox, voters’ preferences must be single-peaked (see Panel B in Figure 1). This means that, if a voter moves away from her/his most preferred outcome in any direction, utility will fall. If we took voter z’s situation in Panel B, satisfaction levels derived from Large and Small budgets are smaller than that derived from a community decision to select the Medium budget.

Another requirement for a stable majority voting outcome is that the issue being voted on must be on a single issue only (in other words ‘one-dimensional’). Our example above (i.e.the question put to vote ‘how much to spend on a public good’) clearly satisfies this condition.

Black also found that as long as voting preferences are single-peaked, voting reflects the preferences of the median voter. The median voter is the voter whose preferences lie in the middle of the set of all voters’ preferences – half the population votes for higher values and the other half want lower values. Looking at B again, while voter x prefers Large to Medium to Small (worst-case scenario) and y prefers Small to Medium to Large (worst-case scenario), they could both settle for Medium (voter z’s preference) if only to avoid their respective worst-case satisfaction scenarios. In other words, the median vote – indicated diagrammatically as the mid-point on the x-axis – wins, and majority voting finds a stable equilibrium.
 

Political Competition, Market-Style

In a forerunner to his seminal book, An Economic Theory of Democracy, Anthony Downs (1957) extended the median voter model by examining the roles of the politician and voter in a simple two-party representative democracy. To begin, the major axioms used by Downs were:

From here, Downs hypothesised that politicians act in rational self-interest in the sense that they attempt to formulate policies to maximise votes, rather than win the election to formulate important policies later. On the other hand, rational voters will cast their votes for those political representatives who will best represent their fiscal interests. Specifically, the voter’s objective is to maximise the excess of benefits which she/he derives from government expenditures over the voter’s tax costs.

However, there are reasons to expect that there are significant costs associated with voting, which a rational individual would tend to avoid. The factors that encourage the phenomenon of ‘rational ignorance’ among voters are as follows:

How do vote-maximising parties respond to the rational ignorance condition? Instead of selling a confusing number of policies to an apathetic electorate, parties can simplify and stimulate electoral competition by offering alternative ideologies. If a voter detects a stable relation between a party’s ideological stance and its policies, she/he can rationally vote on an ideological basis instead of on policy platforms.

In a special case, Downs shows that the party ideology most preferred by the median voter will win the election. That is, if rival parties aim to get elected into office, there will be incentives to introduce similar policies in order to attract the median voter – consensus politics will prevail.

In the Downsian model of politics, the politicians effectively play the entrepreneur selling policies in return for the votes of the citizenry. That is, there is a close analogy between the firm’s competition for consumers in private markets and the politician’s competition for voters. In addition, apathy among citizens towards elections and ignorance of political issues can be rationalised as efficient reactions in a large democracy.

The Calculus of Consent: The Virtues of Unanimity

In 1896, the Swedish economist Knut Wicksell developed his ‘theory of just taxation,’ which sought to specify the institutional framework necessary if efficient budgets are to emerge from political processes. According to Wicksell, efficient budgets can only emerge under a rule of unanimous voting (not majority voting). This ensures that the preferences of minority groupings are sufficiently recognised when deciding on the efficient level of public good expenditure (along with an associated tax package to cover its cost). While Wicksell later relaxed the 100% unanimity rule in favour of near-unanimity (e.g. 90%, 75% approval), the concept of protecting political minorities from abuse by instituting more inclusive voting structures remained.

Buchanan and Tullock’s classic work of 1962, The Calculus of Consent, examined the issue of what would constitute the most efficient voting rule from a transaction costs perspective (see Figure 2). Following Wicksell’s insights, as the voting rule becomes more inclusive to the point of unanimity, the expected cost of budgetary inefficiency falls (Panel A), since more of the voters have their preferences considered. On the other hand, the costs of securing consent rise the closer the voting rule moves to unanimity (Panel B) – it is easier to secure some sort of decision in a simple majority (51%) than under the unanimity rule (100%).

The optimal size voting rule is that percentage of the population at which these two costs are together minimised. This occurs at K (in Panel C), where the vertical addition of the two curves reaches a minimum. Therefore, the optimal majority to pass the issue at stake, given the inefficiency and decision cost curves, is K/N. At this percentage, the expected gain in utility from refining a proposal to gain one more supporter just equals the expected loss in time from doing so.

The important implication from this cost analysis is that no special significance is attached to majority rule – majority voting is efficient only by accident. Indeed, K/N could well be that percentage of population representing near-unanimity voting. Furthermore, Buchanan and Tullock agree with Wicksell that unanimity, not majority consent, is the appropriate benchmark. Taking the example of voting on the level of public goods expenditure and their associated tax costs, under the unanimity concept the minority needn’t have to submit to a majority consumption pattern (i.e. mix of private and public goods) which is not to their liking.

Whether it be 100% unanimity, or near-unanimity, this system is preferred by Buchanan and Tullock to majority voting on welfare maximisation grounds. In the private market, Pareto optimality occurs when a move from a given state of resource allocation couldn’t possibly improve the welfare of one or more individuals without simultaneously eroding the welfare of another person. Similarly, once a unanimous decision has been made through the political process, no new proposal could possibly be preferred. When fiscal propositions command unanimous consent no individual can be forced to accept public sector projects which she or he does not value. Therefore, unanimous political consent is the analogue of Paretian market efficiency.
 

The Theory of Rent-Seeking

In his influential 1967 paper, ‘The Welfare Costs of Tariffs, Monopolies, and Theft,’ Gordon Tullock provided a number of illustrative examples to show that individuals may use the political process to redistribute wealth from others to themselves. Tullock suggested that if this rent-seeking behaviour existed, then the welfare costs of monopolies and tariffs may be significantly greater than that empirically estimated by economists.

Tullock’s tariff example is illustrated in Figure 3. Taking the case of, say, automobiles, the foreign automobile supply curve is initially P0, with the domestic automobile supply curve upward sloping. Without the tariff, the price for automobiles is P0 (the prevailing world price), domestic producers produce D0 units and Q0 - D0 units are imported to make up the national automobile consumption of Q0.

Suppose that a automobile tariff of P0P1 per unit is imposed. In this case, the price of imports will rise to P1. Domestic production would expand to D1 and imports would shrink accordingly (to Q1 - D1). The dotted triangle represents the losses associated with a relatively inefficient domestic car industry producing more automobiles, while the shaded triangle represents the loss of consumer surplus due to the tariff’s impact on price. Government tariff revenue would be represented by rectangle marked Ta, and the owners of the resources in the domestic industry would receive an amount of resources equal to the trapezoid Tr. Before Tullock’s analysis, areas Ta and Tr were not regarded as social losses.

However, according to Tullock, there is an additional loss to society which depends on the size of transfer Tr. In our example, domestic automobile manufacturers would be willing to expend resources up to the value of Tr to convince the government to levy the tariff. Thus Tr represents a waste of society’s resources because expenses are being incurred not in an effort to increase wealth, but in an attempt to transfer it.

Tullock’s insight showed that if an industry or social lobby group can succeed in currying favour with the government of the day then it stands to receive substantial benefits, while the cost associated with the successful implementation/maintenance of the program will be thinly spread over society in the form of increases in current taxes (or increases in taxation in the future as a result of an immediate increase in public sector debt).
 

The Economics of Public Administration

An important element of the governmental apparatus is the government department, which exists on the exchange relation of offering a total output in exchange for a budget to carry out the department’s functions. While the traditional view of bureaucracy was that bureaucrats implemented policies in the name of the public interest, public choice theorists, particularly William Niskanen (1973), instead view bureaucrats as self-interested utility maximisers.

Niskanen hypothesised that bureaucrats seek to enhance personal utility by maximising the budgets of their respective departments, since it is expected that their personal incomes and power status (through increased promotional opportunities) would be increased. In this ‘empire building’ scenario, the bureaucrat will use the department’s monopoly position as sole supplier of the public good to submit to the sponsor (ie. government minister) a budget request which either :

Therefore, according to Niskanen, there is an inherent tendency by bureaucrats to inflate public sector budgets.

In order to avert the resulting overexpansion of the public sector, Niskanen suggested changes in the nature and use of the bureaucracy in order to maximise its responsiveness to the public interest. For example, appropriate public sector reforms could include greater competition between government departments in providing services to the sponsor and hence the community, greater parliamentary oversight of departmental operations (e.g. parliamentary estimates committees), structures to improve the productivity of personnel such as performance agreements and flatter departmental structures, contracting out government activities, or privatisation.

The Repudiation of Keynesian Economics

Public choice experts have also shown that macroeconomic theories can be used in politically expedient ways. For example, Buchanan and Wagner (1977) found that the Keynesian macroeconomic model – using the government budget to maintain high employment and output, and matching budget surpluses in boom periods with budget deficits in recession – ignores the realities of the incentives that drive the democratic political system.

In this context, elected politicians enjoy spending public monies on projects that yield ‘demonstrable benefits,’ however defined, for their constituents. Politicians do not enjoy imposing taxes on these same constituents – even deferring current taxation through increased public sector borrowing is a preferred option. On the other hand, constituents will tend to oppose fiscal consolidation because it will involve an increase in real tax rates or a decrease in real rates of public expenditure, or a combination of the two.

According to Buchanan and Wagner, the underlying tendency of governments to prefer spending over taxing to win electoral support has been encouraged by Keynesian ‘pump priming’ theories. However, this has meant that Keynesianism has introduced the basis for unsustainable public sector growth and continual inflationary pressure into modern economic systems. In other words, Keynes effectively created a political climate in which budget deficits are more ‘legitimate’ even in periods not marked by economic depression.

Buchanan and Wagner point out that before Keynes gained prominence, the prevailing moral was that governments should always maintain a sound budget balance policy. Therefore, to restore moral constraints on politicians to rein in budgets, Buchanan and Wagner argued for the re-establishment of a responsible peacetime ‘macroeconomic constitution.’ This may involve elements such as the formal imposition of balancing outlays with receipts, automatically adjusting budget surpluses and deficits back into balance, and limits on spending and tax rates. Monetary stability could also be provided by a constitutional rule directing the central bank to increase the monetary base at a rate equivalent to the real growth rate for the national economy.

Conclusion

The need for collective action, and hence the organisation of government, stems from the realisation that without collective enforcement of capitalist institutions, such as property rights, the anarchy which is likely to result would only serve to consign human life to one of nastiness, brutishness and, ultimately, short-termism. As Friedman (1962) demonstrated, in a world inhabited by imperfect people, collective organisation is required to prevent ‘bad’ people from doing harm as much as enabling ‘good’ people to do good (noting, of course, that the ‘good’ and ‘bad’ people may be the same people).

Whilst good (limited) governance is intimately linked with liberal values, the existence of government is itself not sufficient to assure the delivery and maintenance of the public good. If it is assumed that the players of the ‘game of politics’ are the same as their fellow human beings, in terms of being self-seeking and opportunistic, then constraints would be required (either electoral, constitutional, legal or other) to ensure that the political process is not used for exploitative purposes. As noted by David Hume in 1741:

This summation is supported by public choice experts who argue that methodological consistency, not to mention the experience of history, implies that the benevolent despot model of government should be rejected as the normal course of events. Therefore, it can be reasonably argued that the deeper understanding of political institutions and behaviour that is public choice theory represents a valuable tool in developing and maintaining government systems that are consistent with the liberal order.

References

Black, Duncan 1948, ‘On the Rationale of Group Decision Making,’ in C.K. Rowley (ed.) 1993, Social Choice Theory I, Edward Elgar, Aldershot.

Buchanan, James 1987, ‘The Constitution of Economic Policy,’ American Economic Review, 77:243-250.

Buchanan, James 1991, ‘Constitutional Economics,’ in J. Eatwell, M. Milgate and P. Newman (eds.), The New Palgrave Dictionary of Economics, Macmillan, London.

Buchanan, James and Gordon Tullock 1962, The Calculus of Consent, University of Michigan Press, Ann Arbor.

Buchanan, James and Robert Wagner 1977, Democracy in Deficit: The Political Legacy of Lord Keynes, Academic Press, New York.

Downs, Anthony 1957, ‘An Economic Theory of Political Action in a Democracy,’ Journal of Political Economy, 65: 135-150.

Friedman, Milton 1962, Capitalism and Freedom, University of Chicago Press, Chicago.

Hume, David [1741] 1985, Essays, Moral, Political and Literary, Liberty Classics, Indianapolis.

Musgrave, R. A. and P.B. 1989, Public Finance in Theory and Practice, McGraw-Hill, New York.

Niskanen, William A. 1973, Bureaucracy – Servant or Master: Lessons from America, Institute of Economic Affairs, London.

Tullock, Gordon 1967, ‘The Welfare Costs of Tariffs, Monopolies and Theft,’ in J.M. Buchanan, R.D. Tollison and G. Tullock (eds.) 1980, Towards a Theory of the Rent-Seeking Society, Texas A&M University Press, College Station.

Wagner, Robert E. 1973, The Public Economy, Markham, Chicago.

Julie Novak currently works as a research officer within the ACT Department of Business, the Arts, Sport and Tourism. She attended a CIS Liberty and Society Seminar in 1996. The viewpoints expressed in this paper are not necessarily those of the Department. 

Policy is the quarterly review of The Centre for Independent Studies. For more information on subscribing to Policy, click HERE

If you are interested in the Centre's activities and publications, why not subscribe to e-PreCIS, our regular email update on the latest news and events.

(e-PreCIS requires html capable email facilities, such as Microsoft Outlook Express or Netscape Messenger)