This is the second part of a two-part analysis of the 2002 midterm election results.
President Bush campaigned actively before this year's elections, doing everything possible to leverage his popularity among the Republican base into votes on Election Day. Gallup data showed a significant increase in the percentage of Republicans -- compared to Democrats -- who said they were giving a great deal of thought to the election as it approached.
It's most likely that this shift reflected, at least in part, Bush's campaign efforts. On the eve of the elections, Gallup asked voters about Bush's likely impact on their votes. Thirty-five percent of likely voters said they were sending a message of support for Bush with their vote, and 18% said they were sending a message of opposition. The rest said Bush did not affect their votes either way. Among Republican likely voters, 64% said that their vote was one of support for the president. Among Democrats, only 36% said that their vote was one of opposition to Bush.
Terrorism
A second way that Bush presumably helped the Republicans this year was through his administration's relentless focus on the war on terrorism. It's hard to discern the exact effect of terrorism on the election results, but our final analysis suggests that it ended up playing to the GOP's advantage.
Terrorism was not the dominant issue to voters by any means. September and November Gallup Polls showed the economy atop Gallup's monthly measure of what Americans see as the country's most important problem. Terrorism was the most important problem in an October poll. The two issues were far and away the primary ones in Americans' eyes all year, and both probably figured heavily in voters' minds.
More directly, election-eve data showed that terrorism was also not the most important issue among likely voters. We gave voters a choice of six issues and asked them to name the one that was most important to their congressional vote. The economy was chosen by 28% of likely voters -- by far the No. 1 issue. The possibility of war with Iraq was second, with 18%. Terrorism, at 14%, was essentially tied for third place with education (13%) and Social Security/Medicare (13%).
Despite its lower position in this importance hierarchy, the threat of terrorism appears to have been a key factor for Republicans. The reason? The Republicans had an enormous positioning advantage over the Democrats on the issue, and that advantage was not countered by a Democratic advantage on the economy, the issue that resonated most with voters.
Here are the numbers from Gallup's election-eve survey:
- A full 81% of those who said that terrorism was the No. 1 issue driving their vote indicated they would vote for the Republican congressional candidate in their district.
- Republicans had a one-point advantage, 49% to 48%, among those who said that the economy was their top issue.
- Republicans were roughly even with Democrats among those who said that Iraq or healthcare was their top issue.
- Democrats had a strong advantage among those who said that education or Social Security/Medicare was their top issue.
Taken as a whole, these results yield a GOP advantage. Republicans' strong position on terrorism more than made up for the Democratic advantage on education and Social Security/Medicare. There was an essential tie for the two parties among those whose votes were driven by other issues.
So despite conventional wisdom, large numbers of voters did not enter the booths this year with terrorism on their minds. More were concerned about the economy and the prospect of war with Iraq. But the minority who did have terrorism on their minds was overwhelmingly Republican, and try as they might, the Democrats were not able to establish positioning on enough of the other issues to counter this strong GOP advantage.
Looking Ahead
Do the GOP's 2002 victories and Bush's robust job approval ratings predict Republican success in the 2004 presidential election? Not necessarily. Three incumbent presidents since World War II have lost their bids for re-election: Gerald Ford (1976), Jimmy Carter (1980), and George H.W. Bush (1992).
Ford's Republican Party lost 48 seats in 1974 and Ford's job approval rating just before that election was 54%. (This was a highly unusual situation, in that Ford had just moved into the job a few months before the 1974 elections, replacing Richard Nixon who resigned in August of that year.) Carter's Democratic Party lost 15 seats in 1978, and his job approval rating at that time was 49%. The elder Bush's Republican Party lost eight seats in 1990, when his job approval rating was 58%.
The younger Bush's party just won six House seats and his job approval rating before the elections was 63%. Thus, Bush would appear to be in a somewhat better position vis-à-vis his forthcoming run for re-election than any of the three presidents who ran and lost.
But the relationship among midterm election results, the president's approval rating, and success at the polls in the next presidential election is far from straightforward.
Bill Clinton provides the most interesting example of this uncertain relationship. The Democrats lost 52 House seats in the famous "Contract With America" election of 1994, when Clinton's approval rating was a weak 46%. Yet Clinton came back convincingly in 1996 and retained the presidency. On the other hand, Clinton's party lost the White House in 2000, despite the fact that the Democrats won seats in 1998, and Clinton's approval rating was a high 66% just before the 1998 elections.
The Democrats lost seats in 1962, and they won the next presidential election in 1964 by an overwhelming margin. The Republicans lost seats in 1954 but retained the presidency in 1956. The Republicans lost seats in 1986 and won the presidential election in 1988. Nixon's party lost 12 House seats in 1970, yet Nixon won 49 of 50 states in the 1972 presidential election, sweeping back into office with one of the biggest popular vote margins in recent decades.
The long and the short of this exercise in history: there is no highly predictive pattern. Certainly, George W. Bush would rather be coming off of a GOP midterm victory with high job approval ratings than in the position of Carter, Ford, or Harry Truman, whose party lost 29 seats in 1950 when his job approval rating was an anemic 39%. (The Democrats lost the presidency in 1952.) But the Republican gains in 2002 by no means ensure that the Republicans (presumably with Bush running for re-election) will win in 2004.
