CONTACT: Stanford University News Service (650) 723-2558
Issues overpower incumbency in congressional elections
STANFORD -- Computer analysis at Stanford of congressional election results have ruled out the “Perot factor” and anti-incumbent fever as explanations for the stunning Republican victory on Nov. 8.
“There was no Perot effect, and it was not strictly an anti-incumbent election because Republican incumbents didn't lose,” said political scientist David Brady.
“If Perot made a difference you would expect his vote totals [within congressional districts in 1992] to predict a shift in the percent of votes or seats [in those districts in 1994], but for Democrats who won and those who lost, there was no difference in the 'Perot vote.' ”
But increasing voter attention to national policy issues - as opposed to local constituency service - may explain how the Republicans won control of both houses in the Nov. 8 election, said Brady and political scientist Douglas Rivers.
Based on their early computer analyses of election results, the two scholars said that there is some evidence that conservative voters' displeasure with their congressional representatives' votes on a variety of national issues may have overcome the usual advantage that incumbency offers.
“Even though Congress makes national policy, the bulk of congressional campaigns have been run on local issues” in the past, said Rivers, professor of political science. “This [election] was unusual in that we had more candidates running on national issues.”
Brady and Rivers, along with graduate students and Hoover Fellow John Cogan, are continuing to compare 1994 congressional district election returns to incumbents' voting records. Brady said the results so far suggest that President Clinton's party may have been punished by voters, particularly in the South, for incumbents' votes on national policy issues such as restricting gun ownership, the president's “don't ask, don't tell” policy for dealing with gays in the military, the 1993 budget and the North American Free Trade Agreement.
“The only policy conclusion I have, and I am not certain of this yet, is that I think the people who got hurt were the Democrats who were moderate in the South and perhaps the West,” said Brady, the Bowen H. and Janice Arthur McCoy Professor of Political Science, Business and the Changing Environment in the business school.
“Those who ran conservatively back home but basically were liberal in Washington got beat. There is no one [congressional] vote that stands out but incremental small differences for several votes may have made the difference,” he said.
Rivers pointed out that liberal Democrats, such as Ron Dellums of Oakland, won handily while more conservative Democrats such as Dave McCurdy in Oklahoma and Jim Cooper in Tennessee lost. McCurdy was the last hold-out among Democrats to vote for Clinton's 1993 budget. Cooper offered a health care bill alternative to Clinton's plan.
“I don't think it's because [particular] issues hurt them. I think they are out of step with their districts. At the presidential level, these are districts that have been voting mostly Republican a long time. These were moderate Democrats whose overall voting records caught up with them,” Brady said.
In the South, there was an 8.9 percent swing in votes from Democrats to Republicans, a larger margin even than the normal 6 to 7 percentage-point edge that incumbent candidates get on average. Nationally, the swing in votes from Democrats to Republicans averaged 5.8 percent.
There were more Republican congressional candidates for the first time this year than Democratic candidates, Rivers said. That reflects the fact that the South is no longer uncontested terrain for Democrats.
The only two Democratic presidents in recent years, Carter and Clinton, were Southerners who ran as moderates, Rivers said. “When confronted with a Democratic Congress, they turned out to be more liberal than their campaigns. I think this disappointed particularly white male Southern voters. It was enough to convince a number of them they weren't Democrats any longer and that has strong long-term implications for the Democratic Party.
“I'd say 1996 looks very bleak for Clinton in the South.”
Brady said that Clinton did better in 1992 than Michael Dukakis in 1988 “in places where people thought Clinton was a 'New Democrat,' and that's where Democrats did worst this time. I don't think it]'s [Clinton's] character but probably policy issues.”
While the size of the Republican victory surprised Brady and many others who use past data on election trends to predict results, he said the swing of votes from one party to another and the accompanying change in House seats is similar in size to swings that have occurred four times in the 24 previous congressional elections since World War II.
In the post-World War II period, other swings above 5 percent occurred in:
“Each one of these swings in the vote brought about 45- to 55-seat losses,” Brady said, “so vote swings of more than 5 percent are associated with a big change in seats.”
What makes 1994 stand out from the others, he said, is that a shift of a similar size changed control of the House of Representatives because the Democrats held a smaller majority going into the election.
A more precise explanation of the recent election will be available next April from results of a study funded by the National Science Foundation. It involves not just computer comparison of election results but re-interviewing a sample of voters who were interviewed in 1992, said Rivers, who helped design that study as part of the study's seven-member oversight board. The interviews are conducted by the Center for Political Studies at the University of Michigan.
This is an archived release.
This release is not available in any other form.
Images mentioned in this release are not available online.