Midterm Elections Look Dicey for GOP

GALLUP NEWS SERVICE

PRINCETON, NJ -- The latest CNN/USA Today/Gallup poll presents a fairly grim picture for the Republican Party one year out from the 2006 midterm elections. Several indicators reveal broad public dissatisfaction with the Republican-led Congress -- dissatisfaction levels not seen since 1994. Additionally, given President Bush's low approval scores and that voters are more likely to say they'll vote for congressional candidates who oppose Bush than vote for those who support him, Bush could be the Republicans' liability-in-chief for retaining control of Congress.

Democrats Lead on Generic Ballot

The Oct. 21-23 survey finds Republicans trailing Democrats by seven points on the generic ballot. Half of registered voters now say they will vote for the Democratic congressional candidate on the ballot in their districts next fall; 43% say they will vote for the Republican.

That in itself is not a large deficit for Republicans on this measure. In the past two midterm elections (1998 and 2002), Republicans were down by nine points and five points, respectively, on the generic ballot among registered voters in Gallup's final pre-election surveys. By virtue of Republicans' higher turnout rates, the Republicans still went on to win a slim majority of seats in Congress.

The generic ballot result from late August was more worrisome for Republicans. At that time, Gallup found the Democrats leading by 12 points, 53% vs. 41% (a result mirrored in polls conducted by other organizations around the same time).

That 12-point Democratic advantage was a wider lead for the Democrats than anything Gallup has seen for most of the past decade that Republicans have been in power. The norm is for Republicans to be trailing by about five points among all registered voters on this measure, which converts to a slight lead among likely voters (that is, the smaller subset of voters who are most likely to go to the polls on Election Day).

This tendency is illustrated by the following table showing the projected percentage voting Republican, according to Gallup surveys, and by the actual results for the past three midterm elections.

Given the precision with which the generic ballot (based on likely voters) typically matches the actual two-party vote for Congress, this is therefore an indicator that bears close monitoring over the next year. (Gallup will not institute its "likely voter" model until closer to the election, so until that time, it is necessary to interpret registered voter numbers in the context of their historical relationship to "likely voter" numbers.)

Mostly Dismal Ratings of Congress

Besides the generic ballot, there are some stronger indications that the Republican majority in Congress may be in trouble. Chiefly, Americans' overall approval rating of Congress is, according to Gallup's Oct. 13-16 poll, just 29%. That compares with 50% approval for Congress in October 2002 and 44% in October 1998. The last time congressional approval fell below 30% was in 1994 -- the year the previously entrenched Democratic majority was ousted by a Republican tidal wave.

Also, the percentage of registered voters who believe that most members of Congress deserve to be re-elected has fallen below 50% for the first time since 1994. Today, just 46% believe most members of Congress deserve another term, while 44% disagree. While not as low as the 38% found just before the 1994 elections, the 46% today is substantially lower than the 57%-58% recorded before the past two midterm elections.

One bright spot for Republicans is that the percentage of voters saying their own member of Congress deserves re-election is holding on at the two-thirds level, similar to where it was in 1998 and 2002, and higher than what was seen in 1994.

Bush Not a Draw With Voters

The impact Bush will have on the congressional elections is unclear, but in principle, his low approval ratings cannot help the Republican Party. If his ratings continue to dip into the low 40s, as they have for the past two months, he could be a greater liability to Republican candidates than was Bill Clinton in 1994.

This finding is underscored by a separate question asking voters what impact a candidate's relationship with Bush will have on their vote for that candidate. By a 55% to 39% margin, a majority of voters say they would be more likely to vote for a congressional candidate who opposes Bush than for a candidate who supports him. Only 6% say it would make no difference.

This is notably more negative than what Gallup found in 1994 and 1998 in reaction to Clinton. In neither case did a majority of voters say they would be more likely to vote for a candidate who opposed Clinton.

A follow-up question, asking voters how strongly they feel about a candidate's support for Bush, reveals an even more dramatic difference in Bush's potential impact on the election. Close to half of all voters today (47%) feel very strongly about voting for a candidate who opposes Bush. Even in 1994, when Clinton's approval rating was similar to Bush's current rating, a much lower percentage (36%) expressed this level of animosity toward Clinton.

Results for this question are highly partisan, but Democrats are more unified in their support of candidates who oppose Bush than Republicans are in their support of candidates who support him (85% vs. 70%).

In 1994, Republicans and Democrats were less strongly partisan in their responses with respect to Clinton. Then, two-thirds (68%) of Republicans felt strongly about picking a candidate who opposed Clinton. This contrasts with the 85% of Democrats today who feel strongly about choosing a candidate who opposes Bush.

Similarly, in 1994, 57% of Democrats felt strongly about backing a candidate who supported Clinton. This contrasts with 70% of Republicans today who feel strongly about voting for a candidate who supports Bush.

Survey Methods

These results are based on telephone interviews with a randomly selected national sample of 1,008 adults, aged 18 and older, conducted Oct. 21-23, 2005. For results based on this sample, one can say with 95% confidence that the maximum error attributable to sampling and other random effects is ±3 percentage points. In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of public opinion polls.

Thinking for a moment about the elections for Congress next year,

2. If the elections for Congress were being held today, which party's candidate would you vote for in your Congressional district -- [ROTATED: the Democratic Party's candidate or the Republican Party's candidate]?

2A. As of today, do you lean more toward -- [ROTATE: the Democratic Party's candidate or the Republican Party's candidate]?

6. Please tell me whether you think each of the following political office-holders deserves to be reelected, or not. How about -- [ROTATED]?

A. The U.S. Representative in your congressional District

Q.6 CONTINUED

B. Most members of Congress

9. Would you be more likely to vote for a candidate for Congress who supports President Bush, or more likely to vote for a candidate who opposes President Bush?

Trends for Comparison: Bill Clinton

Thinking about your vote for Congress, would you be more likely to vote for a candidate for Congress who supports President Clinton, or more likely to vote for a candidate who opposes President Clinton?

9a. (If have preference:) Do you feel very strongly about this, or not so strongly?

COMBINED RESULTS: Q9/9A

Trends for Comparison: Bill Clinton