The Iraq War, Partisanship, and Candidate Attributes: Variation in Partisan Swing in the 2006 U.S. House Elections

• Author: BRUCE I. OPPENHEIMER,CHRISTIAN R. GROSE
• Date: 01 November 2007
• DOI: http://doi.org/10.3162/036298007782398495
• Published date: 01 November 2007

531Iraq War, Partisanship, and Candidate Attributes

LEGISLATIVE STUDIES QUARTERLY, XXXII, 4, November 2007 531

CHRISTIAN R. GROSE

BRUCE I. OPPENHEIMER

Vanderbilt University

The Iraq War, Partisanship,

and Candidate Attributes:

Variation in Partisan Swing

in the 2006 U.S. House Elections

Although partisan swing is often assumed to be uniform across congressional

districts, our analysis of the 2006 House elections demonstrates that systematic

variation exists. In addition to incumbency status, partisanship, spending, and scandal,

variation in the local salience of national issues across districts affects vote shifts in

these districts. Notably, partisan swing in Republican districts proved highly sensitive

to the number of Iraq war deaths from that district and, to a lesser degree, to the roll-

call vote of Republican House members on the war resolution. These findings have

implications for theories of anticipatory representation, retrospective voting, and

electoral accountability.

Few political observers dispute that the Iraq war influenced the

outcome of the 2006 midterm elections. The perceived failure of U.S.

policy in Iraq contributed to President George W. Bush’s low approval

rating, which in turn disadvantaged Republican congressional candi-

dates. Pundits suggested that the Democratic wave was across the board,

contributing to the national partisan swing of the election. We analyzed

the partisan swing in U.S. House elections in 2006 and found that

three factors explain the variation in swing across House districts:

candidate attributes, the Iraq war, and the underlying district partisan-

ship. Regarding the Iraq war, we discovered that the number of district-

level “hometown” fatal casualties and legislators’ votes authorizing

the war can explain the variation in the partisan swing, although this

effect is contingent on the legislator’s party.

In 2006, Democrats regained control of the House of Represen-

tatives for the first time since the 1994 Republican landslide. The party

captured 30 seats, producing a 233-202 House majority at the start of

the 110th Congress. Nationally, Democrats won an estimated 54% of

532 Christian R. Grose and Bruce I. Oppenheimer

the two-party vote for the House, an increase of 5.4 percentage points

from the 48.6% they garnered in 2004.1 If that 5.4-point partisan swing

had been universal across all districts, then the Democrats would have

struggled to win control of the House and come nowhere near a 30-

seat pickup. Although it has been convenient for preelection forecast-

ers to think in terms of a national partisan swing with little variation

across congressional districts, that viewpoint is not valid. Partisan swing

may typically vary somewhat across districts, as the importance of

challenger quality, scandal, campaign spending, and incumbency status

of the seat differentially affect races. But the 2006 election was not

typical. In addition to these conventional factors, the salience of the

Iraq war may have led to larger partisan swings in some districts.

We investigated what factors influenced the amount of swing

between 2004 and 2006 across House districts and to what degree the

variation in swing was systematic. To do so, we accounted for some

variables that scholars have commonly included in the analysis of con-

gressional elections, such as incumbent versus open seat, quality of

challenger, and scandal. In addition, we examined the underlying par-

tisan composition of each district and the salience of a key national

issue. Aside from some results consistent with previous studies, we

found that the swing was larger in Republican districts, in districts

that were party competitive, and in districts where a key issue (the

Iraq war) was most salient. Voters held individual Republican mem-

bers of Congress accountable for the local impact of Iraq war deaths in

their congressional districts and also held GOP members accountable

for their roll-call votes on the Iraq war. Voters did not reward or punish

Democrats for their votes on the Iraq war or the number of war deaths

in their districts.

2006 House Elections: Expectations and Reality

As the 2006 elections approached, many scholars and pundits

hedged on the question of whether or not the Democrats would win

control of the House or Senate. In the latter case, reservations arose

from the fact that the Democrats were more exposed, holding 18 of the

33 seats to be contested, thus limiting the number of targets from which

they could make gains. For the House, however, attention focused on

a perceived small number of potentially competitive seats. As

Abramowitz (2006, 863) correctly noted in his article forecasting the

midterm elections, only 12 of the Republicans elected in 2004 won

with less than 55% of the two-party vote and only 16 represented