GRABBING THOSE COATTAILS

Which Senate Seats Might Turn on the Presidential Outcome in 2012?

Facebooktwitterredditpinterestmail

It is already obvious that control of the Senate will be up for grabs in 2012, with Republicans needing just 3 or 4 seats to take control (depending on whether the GOP wins the presidency and, along with it, the vice president’s tie-breaking vote).

How much do Senate races in presidential years depend upon which White House ticket carries a state?

Without a doubt, it is the tight Senate contests that can be most affected by the top of the ticket. So let’s take a look back at all Senate races that were decided by 53-47% or less in the last eight presidential match-ups (1980-2008).

Competitive Presidential-Year Senate Races, 1980-2008

*: Under Georgia election law, Senate candidates must capture a majority of the vote to win on Election Day, otherwise the top two candidates proceed to a runoff. In 1992, Wyche Fowler captured first place in the initial Election Day balloting, but lost a later runoff.

Overall, a healthy 58% of the 66 truly competitive Senate battles have been won by the candidate of the presidential nominee that carried the state.

Every election is different, of course, and coattails appear longer in some years (1980-Reagan vs. Carter and 2004-Bush vs. Kerry) than others (1988-Bush vs. Dukakis and 1992-Clinton vs. Bush vs. Perot).

In 2004, Bush’s robust margins in Florida, Kentucky, North Carolina, and South Dakota may well have pushed, respectively, Republicans Mel Martinez, Jim Bunning, Richard Burr, and John Thune over the top. In 2008, Democrats Al Franken of Minnesota and Jeff Merkley of Oregon won wafer-thin victories while Barack Obama was sweeping their states. It is doubtful either Franken or Merkley would be in the Senate today had Obama not led their ticket so handsomely.

The 58% total actually understates the effect of presidential coattails. Look at 1980, for example. Surely, Sens. Gary Hart (D-CO) and Thomas Eagleton (D-MO) would not have had close scrapes in winning reelection had Ronald Reagan not been securing landslides in Colorado and Missouri. Even in his relatively coattail-less 1984 reelection triumph, Reagan surely made the difference for Mitch McConnell in Kentucky and Jesse Helms in North Carolina in their difficult campaigns.

There are remarkable exceptions along the way, too. Ben Nelson (D-NE) had a weak opponent in 2000, but it is still noteworthy that he won with 51% while George W. Bush captured a commanding 62% of the statewide vote in Nebraska. Will Nelson be able to repeat his feat in 2012, when any mainstream Republican nominee for the White House will probably trounce President Obama in the Cornhusker State?

Other than Nelson in Nebraska, there are quite a few 2012 contests almost certain to be substantially influenced by the downdraft from the presidential race, including but not restricted to Florida (Sen. Bill Nelson-D), Massachusetts (Sen. Scott Brown-R), Missouri (Sen. Claire McCaskill-D), Montana (Sen. Jon Tester-D), Nevada (Sen. John Ensign-R), North Dakota (open seat of Sen. Kent Conrad-D), Ohio (Sen. Sherrod Brown-D), and Virginia (open seat of Sen. Jim Webb-D).

When you watch these nine seats, seven of them currently held by Democrats, remember to include the presidential factor in your calculus. You can be certain that some lucky candidates in these states will be grabbing for their White House nominee’s coattails on Election Day.


For further reading: Political scientists have long been fascinated with the relationship between presidential votes and the outcome of Senate races. Readers who want more historical depth should take a look at:

Presidential Coattails in Senate Elections
James E. Campbell and Joe A. Sumners
The American Political Science Review
Vol. 84, No. 2 (Jun., 1990), pp. 513-524

Examining presidential election year Senate races from 1972 to 1988, Campbell and Sumners found that presidential and Senate votes were positively correlated, cautioning that the correlation was “not strong,” but “also not inconsequential.” “A 10-percentage-point gain in a party’s presidential vote in a state, according to this estimate, adds about two percentage points to the vote for its Senate candidate.”

Politics, Parties, and Pressure Groups
V.O. Key, Jr.
Thomas E. Crowell, New York (4th Edition, 1958), pp. 592-599.

In a particularly early examination of the coattail effect in Senate races, Key found in 1958 that parties do in fact win more seats in states in which the presidential candidate ran well.

Incumbency and the Presidential Vote in Senate Elections: Defining Parameters of Subpresidential Voting
Barbara Hinckley
The American Political Science Review
Vol. 64, No. 3 (Sep., 1970), pp. 836-842

Hinckley’s examination of Senate races in competitive states during the presidential election years between 1956 and 1964 showed a positive correlation (r = .70) between “the presidential voting deviation for the base party vote” in each state and the “deviation for affiliated Senate candidates.” “The stronger the presidential candidate ran in a state, the stronger the affiliated Senate candidate ran…”