THE AMERICAN PRESIDENTIAL ELECTION JUST HELD APPEARS to have been a momentous event. It was the greatest repudiation of an incumbent president and party in America since that of Hoover and the Republicans in 1931. Predicted by no one — by neither political scientist, pollster nor politician — this was as much a surprise to joyful Republicans as to luckless Democrats. The voters had not discussed the matter much, except in deprecatory terms borrowed from the media, and on the morning after they too looked around among themselves with wonder and satisfaction.

It could hardly be believed, as Carter's pollster asserted, that everyone turned to Reagan on the last day when the hostages held in Iran failed to appear. Something so considerable must have been meditated, or at least prepared; and this misfortune for Carter does not explain why the Democratic Senators, who were not involved with Iran, were voted out. Perhaps the large number of ‘undecideds’ were Reagan voters, too embarrassed to confess their decision to callers they suspected were unsympathetic. Or perhaps — a wishful fantasy of mine — the voters deliberately did not disclose their intention to the pollsters since they knew that sovereignty stays alive only with an occasional surprise. The voters may have sensed that they can be exploited if they can be predicted.

ON TUESDAY, 5 NOVEMBER 1996 BILL CLINTON WAS REelected President of the United States by a comfortable margin. In addition, the Republicans retained control of both the House of Representatives and the Senate. Although these results were generally unsurprising, the election was an important one for students of American electoral behaviour. In particular, the results were widely expected to cast new light on two distinctive strands of thought in the literature: claims that we are in the midst of a realignment of the electorate towards the Republicans; and claims that American voters now consciously choose divided government so as to limit the scope of government and avoid the extremes in both parties. The second part of this article will address these questions and will discuss their implications for President Clinton's second term. First, however, it will be useful to examine the results in more detail.

ONE OF THE MORE INTERESTING ASPECTS OF THE 1996 US presidential election was the evident inverse relationship between the results that year and those of the election contested exactly one hundred years earlier. Most of the states won by Bill Clinton were carried by the Republican William McKinley in 1896, while nearly all the states won by Bob Dole were those carried by William Jennings Bryan, the Democratic candidate. In 1896 there were 45 states, and of them 28 were won by the Democrats and 17 by the Republicans in the 1996 presidential election. Twenty-one of the 28 Democratic states in 1996 (or 75 per cent of the total) had been Republican a hundred years earlier; 15 of the 17 Republican states in 1996 (or 88 per cent of the total) went for the Democrats in 1896. This impression of the two election years being mirror images of each other, at least so far as the pattern of presidential results is concerned, is even more evident when looking at the distribution of Electoral College (EC) votes. (The five states created since 1896 are omitted from the entire discussion.) Of the 358 EC votes the 282 that went to the Democrats were from ‘McKinley states’, and they constitute 78 per cent of the Democrat total. In the case of the Republicans, 133 out of 148 EC votes were from ‘Bryanite states’ - that is 90 per cent of the 1996 Republican EC vote.

The American Elections This Year Overturned conventional expectations. In March 1991 President Bush — leader of the free world during the fall of the Berlin wall and winner of the Gulf war — seemed invincible. With the troops home in victory parades, Bush's approval rating stood at 88 per cent in Gallup polls. Few thought he could lose against Governor Clinton, dogged by questions about the draft and marital fidelity, perceived as a second-rank Democratic contender from a small southern state. Yet President Bush won 37.7 per cent, the lowest share of the popular vote of any incumbent president since William Taft in 1912. Against most predictions, 19 per cent of the vote went to Ross Perot, the best result for a third party candidate since Teddy Roosevelt. Governor Clinton enjoyed a comfortable victory, with 43.2 per cent of the popular vote and 370 electoral college votes in 32 states. The result for Clinton was not a landslide, indeed his share of the popular vote was similar to Mondale's in 1984 and Carter's in 1980, and less than Dukakis's in 1988. But questions about a popular mandate are academic: Democrats now control both ends of Pennsylvania Avenue, and almost two-thirds of governorships and state legislative chambers.

It is widely assumed that a representative democracy requires an enlightened citizenry in order to function properly. The competence of citizens has been studied extensively and the sociodemographic determinants of political sophistication are particularly well known. Much less is known about whether and how citizen competence affects electoral behaviour and outcomes. This article reviews the existing literature on these topics. Despite the widespread consensus that, generally speaking, citizen competence matters for electoral outcomes, the review produced a mixed result: some studies suggest that the political left would benefit from a better-informed electorate, while other studies suggest the opposite. Although the majoritarian electoral context is overrepresented in the evidence, the review shows that at the individual level, political knowledge greatly increases a person’s ability to match personal preferences with the right candidate or party in an election. The article also identifies several gaps in existing knowledge, thereby suggesting future research questions.

For more than 40 years, studies trying to explain macro-level electoral turnout have been one of the pillars of political behavioural research. From January 2004 to December 2013 alone, more than 130 articles were published in peer-reviewed journals using turnout at the national, regional or local level as the dependent variable. This meta-analysis tries to synthesize the results of these studies. I find there is a strong consensus in the literature that turnout is higher under compulsory voting, if the election is important, and if it is held in a small country. I also find that the influence of most other predictor variables, including the type of electoral system, the number of parties, development, income inequalities and electoral closeness is inconclusive at best. These results hint at the fact that the determinants of turnout might be more complex than the current theory suggests and is rather more context dependent.