Americans have long taken great pride in their presidential elections. With a proud history extending back over 200 years, the selection of a U.S. president has long been held up to emerging democracies as an example to be emulated. Power has been transferred peacefully and with widespread agreement on the legitimacy of the winner's electoral mandate. Of course, some elements of the election process were not as fair as they could be--particularly the Electoral College--but the overall sense of fairness is something most analysts of U.S. presidential elections have long taken for granted. The disputed outcome of the 2000 presidential contest between Bush and Gore changed all that.
In 1996, the American National Election Studies (ANES) asked about the fairness of the election only because this question was part of a module being administered in many countries as part of the Comparative Study of Electoral Systems (CSES). Several days after the 2000 election, the ANES realized that in light of the controversy in Florida, it would be interesting to repeat this question in their post-election questionnaire. Two years later, when the ANES reinterviewed respondents from the 2000 study, they decided to see whether the controversy still simmered in the public mind.
The findings from 1996 indicate that even in the face of a normal undisputed presidential election, Americans did not rate the fairness of their election particularly high compared to citizens of other established democracies. And in the shadow of the contested outcome of the 2000 American presidential race, complaints about the fairness of the election were widespread. Only the Peruvian election of 2001 was worse in this respect among the 36 national elections covered in the CSES study. One might expect that by 2002, few Americans would continue to question the fairness of Bush's election given the rise in patriotism spurred by 9/11 and the war with Iraq. However, lingering bitterness about the election remained strong, and the patterns of opinion on this question had hardened along party lines. This research note reviews and expands upon these findings.
A Comparative Perspective on Election Fairness
As part of the CSES, random samples of populations in various countries were asked the following question soon after a national election: "In some countries, people believe their elections are conducted fairly. In other countries, people believe that their elections are conducted unfairly. Thinking of the election we've just had, do you believe it was very fair, somewhat fair, neither fair nor unfair, somewhat unfair, or very unfair?" (1) Table 1 displays the results of how citizens responded to this question in a wide range of both established and developing democracies.
Ranking the countries according to the percentage that responded that their election was very fair, the U.S. election of 1996 finishes in 15th place out of 36. The perceived fairness of the Clinton-Dole contest of 1996 pales compared to that observed in a number of parliamentary elections conducted under rules of proportional representation (PR). All of the countries in which elections were rated as very fair by at least two thirds of survey respondents employed PR. Given that the aim of PR is to distribute political power fairly according to votes received, it makes perfect sense that citizens are most likely to think an election is fair when PR is used. A presidential election, however, cannot use PR, as there is only one office to allocate. Hence, winner-take-all presidential elections are at a comparative disadvantage when it comes to being perceived as fair. Third-party voters in 1996 were the least likely to rate the election as very fair, with only 40 percent saying this, compared to 52 percent among Dole voters and 59 percent among Clinton voters. The fact that a typical Scandinavian voter was more likely to see their election as fair than even …