Contributing editor Norman Barry is professor of social and political theory at the University of Buckingham in the U.K. He is the author of An Introduction to Modern Political Theory (St. Martin’s Press) and Business Ethics (Macmillan).

It was once said that “democracy is the most promiscuous word in the language; she is everybody’s mistress.” Indeed, political regimes of widely differing institutional features label themselves democracies, as did totalitarian communist orders. Often, the best guide to a country’s democratic credentials was that it didn’t call itself democratic: compare West Germany’s Federal Republic with the East German Democratic Republic.

A particularly instructive example of the meaninglessness of the term was the election of the Marxist Salvador Allende in Chile in 1970. He has always been heralded as a democratic communist who was destroyed by America and international capitalism. Yet he got only 36 percent in the presidential polling and was faced with a majority opposition in Congress. This licensed him to socialize Chile under the aegis of democracy.

A key to unraveling the problem is to borrow a distinction from old-fashioned logical positivism: emotive versus descriptive meaning. Some words convey descriptive information about the world–like those used in the weather forecast–while others are designed not to tell us anything factually important but to act on our emotions and garner our support, such as advertising slogans and political words. This is true of democracy: If anybody confesses to being anti-democratic he is likely to be called a fascist. In its emotive sense, all sorts of good things, such as liberty, rights, majority rule, and the public interest, are bundled up and marketed under the label “democracy.”

A starting point then for extricating ourselves from this definitional quagmire is to put an adjective in front of “democracy.” In the postcommunist regimes and in the West it is perhaps best understood as “liberal” democracy, which does indicate that the regime is understood not entirely in majoritarian terms. It might be contrasted with “communitarian” democracy or industrial democracy.

Unlike today, democracy was not always greeted with unsullied admiration. In nineteenth-century Britain it was even respectable to be opposed to democracy precisely because if it were understood as untrammeled majority rule, which was its single indisputable feature in those days, it would be a threat to liberty, the rule of law, individual rights, property, and “civilization.” In John Stuart Mill’s Considerations on Representative Government, all sorts of checks on the will of the people are adumbrated. Indeed, the whole point of a constitution is to protect long-term, permanent values against the transient whims of the mob. Indeed, it was slightly encouraging that Al Gore’s defeat in the 2000 presidential election, despite his majority of the popular vote, was not met with outrage. Federalism is, of course, a major and welcome constraint on democracy.

A modern critique of democracy must proceed from these nineteenth-century skeptical insights-but with one crucially important qualification: the threat to civilization has not actually come from the unfettered mob but from the uncontrollable influence of pressure groups (Madison’s “factions”) on the social and economic process. They are much more dangerous than majoritarianism precisely because they can claim the imprimatur of liberal democracy for their anti-individualistic and anti-market effects.

Elitism and Democracy

It was the Italian theorists of elitism who first produced a powerful, theoretical critique of democracy, and some of their strictures are relevant to modern considerations.¹ The economist Vilfredo Pareto demonstrated the inevitability of elite rule, derived from a fundamental theory of human inequality, and showed that the word “democracy” could not be used to distinguish forms of government since all were based on differing types of irresponsible minority rule. Gaetano Mosca did the same thing, except that his elites derived from the bureaucratization of modern society. And, more interestingly, a study by Robert Michels of the German Social Democratic party revealed the “iron law” of oligarchy, which showed that, because of people’s varying propensity for political activism, a minority of enthusiasts would manipulate a formally democratic system. In modern language, people whose opportunity cost for politics is low will dominate the system whatever its rules.

However, it was Joseph Schumpeter who demonstrated that a liberal democratic political regime could be consistent with a type of elitism.² He also produced the beginnings of the modern economic theory of democracy. It was commendably realistic and remarkably prescient.

Schumpeter attacked what had become the orthodoxies of democratic thought: that majority voting somehow transmitted the will of the people to the government and that democratic government produced a better class of citizen (Mill’s major justification for his version of representative government). Schumpeter easily showed that there is no homogeneous will of the people, only a collection of competing wills, and that the “public interest” is an illusion believed in only by political philosophers. He effectively turned democratic theory on its head. It was not characterized by an upward flow of opinion from the people to government, but the reverse. Competing elites (political parties) offer themselves and their wares to the public just as entrepreneurs offer their goods to consumers.

Schumpeter countered Pareto by seeing democracy as a competition (in this sense it does not matter how wide the electorate is as long as there is some freedom to compete): democracy is that “institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people’s vote.”³ It is purely instrumental, a method, like monarchy or dictatorship, for choosing government. It does not necessarily produce some desirable end state, and it is consistent with any outcome. (We all know that Hitler won an election in 1933.)

Although Schumpeter had the key to a wholesale critique of democracy, he still supported it. Democracy could work, he thought, if a society were reasonably homogeneous, had a reliable bureaucracy (he wrote before Public Choice and its explanation of “rent-seeking” by officials had been invented), and not too many affairs were subject to political, as opposed to private economic, decision-making. He thought that the level of rationality fell as soon as people left the marketplace and played politics, either as voters or activists. This seems to be true. Just watch the supreme rationality of the housewife quickly responding to price changes at Wal-Mart, compared to her ignorance of the policy proposals of political parties at an election.

But Schumpeter didn’t quite get it right. The problem is not that the housewife’s rationality disappears. The problem is that it is not in her interest to be well-informed about politics. It is simply in no one’s rational self-interest to be informed about what is in the “public interest.” Least of all is it in anyone’s interest to sacrifice his well-being for the “common good.” Democratic theorists have never solved the problem of why rational people vote at all, given the nugatory effect a single vote can have on the result of an election.

The Logic of Democracy

A coherent critique of democracy requires things of which Schumpeter never dreamt: first, a logical explanation of why the public good cannot often be transmitted through the voting mechanism (not merely the casual observation that it rarely happens) and, second, a theory of why, in practice, democratic politics degenerates into a squabble over benefits among rival interest groups.

There is a famous paradox of democracy, first identified by the Enlightenment thinker the Marquis de Condorcet (who died in a French prison during the Revolution) and proved by Kenneth Arrow in the twentieth century.⁴ Superficially this looks like a purely logical or mathematical game. However, it has serious implications for the normative theory of democracy.

The paradox arises when there are at least three possible decisions and three choosers. Imagine that individuals A, B, and C have preferences for options x, y, and z. A prefers x to y and y to z (therefore x to z), B prefers y to z and z to x (therefore y to x), and C prefers z to x and x to y (therefore z to y). In separate one-on-one votes (x versus y, x versus z, and y versus z) each voter would triumph. The rotating majority would prefer x to y, y to z, and z to x. Thus there is no single “people’s will.” Only in special circumstances will a coherent result be produced.⁵

Of course, in regular elections with only two parties no paradox occurs since there is only one vote and therefore only one winner. But look at Great Britain, which has three parties. With only one vote taken, the party that garners the most votes wins, and that is almost always a party with less than a majority. Tony Blair’s Labour party was elected in 2001 with 44 percent of the popular vote. If three choices had been given and people asked to rank their preferences, the election either would have produced an Arrow paradox or the smallest party, the Liberal Democrats, probably would have won. When Ross Perot was a serious presidential candidate in 1992, a paradox could have occurred in America if voters had been asked to rank him, George H. W. Bush, and Bill Clinton, and separate votes had been taken.

The significance of this for democratic theory is that it is almost impossible to design a system that produces “the people’s” verdict. Of course, even if a result occurred that avoided the Arrow problem, there is no guarantee that it would have been morally right. It all depends on what the spread of opinion is. In a racist society people’s opinions are likely to be consistently ordered in degrees of nastiness. Arrow problems are likely to occur in free and open societies where there are varieties of opinion.

Individualists say that the voting problem would be circumvented if decisions were taken by individuals privately. But presumably there would be some collective decisions, regarding national defense for example, so Arrow problems could still occur.

What Is Bad About Democracy?

Let’s look now at the economic analysis of how conventional democracy works. The relevant question is whether it is an efficient mechanism for delivering genuine public goods, even for the minimal state. We should refer to Schumpeter’s point that politicians are entrepreneurs interested in either power or the non-market income that office can bring. Is there an “invisible hand” in democratic politics, like the one in economics, which ensures that rational self-interest, via market exchange, will produce the public good?

Theory and evidence strongly suggests there is not. Public spending in a democracy is higher than people would actually prefer, as are the taxes imposed to finance it. And outbreaks of inflation regularly occur in democracies because governments have removed the constraint on money creation produced by the gold standard. Moreover, myriad policies would not be pursued were it not for the vote motive.

But in semi-socialist democratic regimes, like Great Britain and the rest of Europe, many services, like health and pensions, meet with regular disapproval precisely because they are under-financed. The classic example is Britain’s socialist medicine on which the state spends a derisory 7 percent of GDP. If people could spend their money privately they would exceed this level. Thus democracy produces the paradox of vastly excess state spending in general while the government underspends on things that are actually wanted: there is no political mechanism that accurately reveals people’s choices.

The explanation is that in a democracy the government tends to be a coalition of interests (factions), and in a system subject to little constraint the key to success lies not in promoting some “public good,” but in rewarding with privileges these electorally significant groups. In most cases, the public good would consist in the government’s getting out of the activity. But there are no votes in that direction because almost everybody gains from some governmental activity. Thus there are subsidies to farmers and, now, trade privileges (import controls) for the steel industry. In European systems, which have proportional representation, the bargains are struck at the parliamentary level. In arrangements where the winner requires only a bare plurality, the rivalrous parties themselves consist of rent-seeking coalitions.

In America, public spending is greatly increased by logrolling. For one state to get a majority in Congress for some federal benefit it has to support another state in its demands, which leads to greater spending all round. Also, the practice of passing “omnibus” bills severely weakens the opposition of the president (the line-item veto was defeated in the Supreme Court) since he might welcome parts of the bill on “public interest” grounds. In fact, a democracy would work better if the people voted directly on separate issues rather than having their representatives vote on bundles produced by the parties. Contra conventional conservatism, direct democracy is actually better than representative government.

A further point favoring government growth is that benefits are narrowly concentrated while costs are widely diffused. Not many people notice the minor increase in taxes used to finance a privilege, yet the effect on the beneficiaries is immediate and obvious. What also encourages the emergence of special-interest groups is asymmetric information. Their members will have more knowledge of the particular issues that interest them than nonmembers and therefore will be in a strong position to influence government.

All of this makes everybody worse off, but each person is in a prisoner’s dilemma: he has no rational incentive to break out of the system from which he gets some benefit because he cannot guarantee that others will be so enlightened. Things, then, might have to get worse before they get better. This happened in Australia in the 1980s. Protection was so costly that few benefited from it and there was an incentive to get rid of most restrictions in one go. A Labour government there did precisely that.

Is There a Way Out?

Because of its emotive appeal, selling an anti-democratic idea is politically difficult. Various alternatives have been suggested, but most are infeasible whatever their internal logic. As suggested above, an effective approach, paradoxically, might be to demand more democracy, with choices put to the people rather than to their elected representatives. It is clear that the erosion of the market has come about through the adoption of a democracy that imposes few constraints on government. No doubt a presidential candidate wouldn’t believe it, but at the moment it is rather easy to get political power: you only have to win about 50 percent of the voters, not of the electorate, in a two-horse race for the office. It is even easier in multiparty parliamentary regimes.

It is true that America has a Constitution, but its original constraints have virtually disappeared. Besides, one wonders how “parchment protections” could ever be effective in an age of mass democracy. Most damaging of all has been the loss of genuine federalism. The Tenth Amendment, which was designed to preserve the integrity of the states, is senescent, and Washington has taken responsibility for things undreamt of by the Founding Fathers.

Still, it is worthwhile speculating on the kind of institutional reforms that might protect the individual and the market. Switzerland has managed to resist some of the excesses of rent-seeking because it has made use of its constitution. Amendments to it have to be approved by a majority of the cantons and a majority of the voters; citizens may demand a referendum on any legislation passed by the federal parliament; and there are other devices to preserve liberty. At the moment the cantons still spend significantly more than the federal government and the voters regularly resist their rulers by voting against any European Union connections. Their rulers favor the Union precisely because it is a rent-seeker’s paradise.

All of this is rather tame for an anti-democrat. Even the Swiss constraints are not insurmountable; they have failed to resist some advances of centralized government. But they do constitute a model from which further dents in the edifice of conventional majority rule and almost unlimited sovereignty might be made.

1. See Norman Barry, An Introduction to Modern Political Theory (London: Macmillan, 2000), pp. 284-87.
2. Joseph Schumpeter, Capitalism, Socialism and Democracy (London: Allen and Unwin, 1954).
3. Ibid., p. 229.
4. Kenneth Arrow, Social Choice and Individual Values (New York: Wiley, 1963).
5. See Norman Barry, “The Stakeholder Concept of Corporate Control Is Illogical and Impractical,” The Independent Review, Spring 2002, pp. 541-54.