Voting is the commonest and most basic way of participating in a democracy, but far too many citizens do not exercise their right to vote, especially in the United States. In the 1988 and 1992 Presidential elections, the turnout of registered voters was only 50 and 55 per cent, respectively, and in the midterm Congressional elections in 1990 and 1994, it was only 33 and 36 per cent.
This is a serious problem for two reasons. One is democratic legitimacy: Can a government that has gained power in a low-turnout election really claim to be a representative government? For instance, some Americans questioned President Clinton’s mandate because he received only 43 per cent of the votes cast and because only 55 per cent of those registered to vote actually did so -- which meant that he received the support of fewer than 25 per cent of all eligible voters in 1992. The other, even more serious problem is that low turnout almost inevitably means that certain groups vote in greater numbers than other groups and hence gain disproportionate influence on the government and its policies.
The only way to solve these problems is to maximize turnout. It may not be realistic to expect everyone to vote, but a turnout of, say, 90 per cent is a feasible goal, as the experience of quite a few democracies shows.
On the basis of studies ranging from the 1920s work of Harold F. Gosnell at the University of Chicago to the 1990s research of Robert W. Jackman of the University of California at Davis and Mark N. Franklin of the University of Houston, we know a great deal about the institutional mechanisms that can increase turnout. They include voter-friendly registration procedures; voting on the weekend instead of during the week; easy access to absentee ballots; proportional representation, with multiple lawmakers representing electoral districts instead of the current U.S. system of winner-takes-all elections; and scheduling as many elections as possible -- national, state, and local -- on the same day.
The evidence suggests that using all of these measures together can produce a voter turnout of around 90 per cent. But adopting all of them is a tall order. Only a handful of states have even managed to introduce the minor reform of allowing citizens to register to vote on the same day as the election.
Fortunately, one other reform, by itself, can maximize turnout as effectively as all of the other methods combined: compulsory voting. In Australia, Belgium, Brazil, Greece, Italy, Venezuela, and several other Latin American democracies, mandatory voting has produced near-universal voter turnout.
It is somewhat surprising that making voting compulsory is so effective, because the penalties for failing to vote are typically minor, usually involving a fine roughly equal to that for a parking violation. Moreover, enforcement tends to be very lax; because of the large numbers of people involved, compulsory voting simply cannot be strictly enforced. (Parking rules tend to be enforced much more strictly.)
For instance, with 10 million eligible voters in Australia, even a typical turnout of 95 per cent means that half a million people did not vote, and it obviously is not practical to issue such a large number of fines. Australia is actually among the strictest enforcers of compulsory voting, but even there, only about 4 per cent of non-voters end up having to pay the small fines. In Belgium, fewer than one-fourth of 1 per cent of non-voters are fined.
Mandatory-voting requirements produce large turnouts, however, even though a government technically cannot compel an actual vote. A government can require citizens to show up at the polls, or even to accept a ballot and then drop it into the ballot box, but it cannot require its citizens to cast a valid vote; secret ballots mean that nobody can be prevented from casting an invalid or blank one.
It is worth emphasizing why low voter turnout is such a serious problem for democracies -- one that deserves our attention. Low turnout typically means that privileged citizens (those with better education and greater wealth) vote in significantly larger numbers than less-privileged citizens. This introduces a systematic bias in favor of well-off citizens, because, as the old adage has it,"If you don’t vote, you don’t count.” The already-privileged citizens who vote are further rewarded with government policies favoring their interests.
The socio-economic bias in voter turnout is an especially strong pattern in the United States, where turout is extremely low. In Presidential elections from 1952 to 1988, turnout among the college-educated was 26 percentage points higher than that among the population as a whole; the turnout for people without a high-school diploma was 16 percentage points lower. Unless turnout is very high -- about 90 per cent -- socio-economic biases in voting tend to be a major problem. For instance, low and unequal voter turnout is a major reason why politicians find it so much easier to reduce government aid to the poor than to cut entitlement programs that chiefly benefit the middle class.
The low levels of voter turnout in the United States are often contrasted with turnouts as high as 95 per cent in a few other countries. But when we measure turnout in other democracies in the way we usually measure it in the United States -- as a percentage of the voting-age population, rather than as a percentage of the registered electorate -- we find very few countries with turnouts above 90 per cent, and most of those nations have compulsory voting. According to a study by G. Bingham Powell of the University of Rochester, half of the world’s democracies have turnout levels below about 75 per cent of the voting-age population. This half includes most of the larger democracies: not only the United States, but also Britain, France, Japan, and India, none of which require citizens to vote.
Even these figures cast turnouts in a deceptively favorable light, because they measure voting in what political scientists call first-order elections -- that is, national-level parliamentary or presidential elections. But the vast majority of elections are second-order elections -- for lesser posts -- which attract less attention from citizens and lower turnouts. In the United States, only Presidential elections produce turnouts of more than 50 per cent of the voting-age population; turnout in midterm Congressional elections has been only about 35 per cent in recent years, and in local elections is closer to 25 per cent.
Low turnout is typical for second-order elections in other countries, too. For local elections in Britain, it is only about 40 per cent. Even in Australia, it is only about 35 per cent, because voting at the local level is not mandatory, as it is for national elections. In the 1994 elections for the European Parliament, another example of a second-order contest, the average turnout in the 12 nations of the European Union was 58 per cent. The power of mandatory voting is highlighted by the fact that when it is applied to local elections -- as it is in all nations with compulsory voting except Australia -- turnout levels are almost the same as those for presidential and parliamentary contests.
It is time that we paid more attention to the issue of voter turnout, because the already low levels of voting in many countries around the world are declining even more. In the United States, voting in Presidential elections has fallen to 50 to 55 per cent of the voting-age population in the 1980s and ‘90s, from 60 to 65 per cent during the 1950s and ‘60s. In the Presidential contest on November 5, turnout is likely to be around 50 per cent or even lower.
The biggest advantage of compulsory voting is that, by enhancing voter turnout, it equalizes participation and removes much of the bias against less-privileged citizens. It also has two other significant advantages. One is that mandatory voting can reduce the role of money in politics, since it does away with the need for candidates and political parties to spend large sums on getting voters to the polls. Second, it reduces the incentives for negative advertising.
As the political scientists Stephen Ansolabehere of the Massachusetts Institute of Technology and Shanto Iyengar of the University of California at Los Angeles have shown in Going Negative: How Attack Ads Shrink and Polarize the Electorate (Free Press, 1995), attack ads work -- indeed, they work all too well. They are effective not because they persuade people to vote for the candidate making the attack and against the candidate attacked in the ads, but because they raise enough doubts in voters’ minds that they decide not to vote at all. So the candidate making the attack has lowered his or her opponent’s total vote.
Moreover, attack ads breed general distrust of politicians and cynicism about politics and government. Under mandatory voting, it would be so much harder for attack ads to depress turnout that I believe they would no longer be worth the effort.
The main objection to compulsory voting is that it violates the individual’s freedom -- the freedom not to vote. This was the main reason it was abolished in the Netherlands in 1970, for example. It is unlikely, however, that the Dutch would have made this decision had they foreseen the disastrous plunge in their voter turnouts, from about 90 per cent in all elections to only 50 per cent and 36 per cent, respectively, in the most recent elections for provincial offices and for seats in the European Parliament.
In any case, the indiidual-freedom argument is extremely weak, because -- as I’ve noted -- compulsory voting does not actually require a citizen to cast a valid ballot. Besides, mandatory voting entails an extremely small decrease in freedom compared with many other, more onerous tasks that democracies require their citizens to perform, such as serving on juries, paying taxes, and serving in the military.
Some scholars argue that U.S. courts might rule compulsory voting unconstitutional because it restricts individual freedom. Richard L. Hasen, of the Chicago-Kent College of Law at the Illinois Institute of Technology, recently has argued, in"Voting Without Law?” (University of Pennsylvania Law Review, May 1996), that the only plausible ground for such a ruling would be the First Amendment’s guarantee of freedom of speech. But the Supreme Court has explicitly rejected the notion that voting can be regarded as a form of speech. For instance, in 1992, in Burdick v. Takushi, the Court upheld Hawaii’s ban on write-in votes, ruling against a voter’s claim that the ban deprived him of the right to cast a protest vote for Donald Duck. The Court said an election is about choosing representatives, not about expressing oneself. Of course, even if mandatory voting were to be found unconstitutional, a constitutional amendment permitting it could be adopted -- a difficult, but not impossible, prospect.
Probably the most important practical obstacle to compulsory voting in countries that do not have it is the opposition of conservative parties, like the Republican Party in the United States. High turnout is clearly not in their partisan self-interest, because unequal turnout favors privileged voters, who tend to be conservative. But conservative parties generally were also opposed to universal suffrage, which eventually was accepted by all democracies, because it was recognized to be a basic democratic principle. Compulsory voting should be seen as an extension of universal suffrage -- which we now all take for granted.
Arend Lijphart is a professor of political science at the University of California at San Diego. This article is adapted from his presidential address to the American Political Science Association.