Home | Opinion | Columnists | Mark Mellman

Splitting the ticket not so rare

With a tight race expected for the White House, control of both chambers of Congress at stake and many of the marquee House and Senate contests taking places in traditionally red presidential states, discussions of split-ticket voting are everywhere. The conventional wisdom: Ticket-splitting is rare. 

Not so fast. 

ADVERTISEMENT
One approach to the data analyzes the extent to which states produce different outcomes in races for president and Senate (or House). A study of 800 presidential-year Senate contests during the last century, undertaken by the Humphrey School of Public Affairs, revealed that on average, 30 percent of races produced winners of opposite parties.

Some states are much more apt to split than others. Montana, where one of the premier Senate races of the cycle pits Sen. Jon Tester (D) against Rep. Denny Rehberg (R), produced split decisions over half the time. North Dakota, the scene of another hot race, between Heidi Heitkamp (D) and Rick Berg (R), splits its decisions exactly 50 percent of time. Missouri, where Sen. Claire McCaskill (D) faces a showdown contest, splits more than 40 percent of the time. History suggests that what happens at the top of the ticket in these states will tell us precious little about the outcome of these Senate races. 

By contrast, Kansas, Utah, Wyoming and North Carolina have only split once each over the last century.

Of course, 100 years is an exceptionally long time. Has the trend altered or abated? In the 2008 cycle, only 20 percent of races produced split decisions, as did 21 percent in 2004. Those years alone might suggest ticket splitting is in decline from the 30 percent average over the century. In 2000, though, the rate was an above-average 32 percent. Two data points do make a line (as we all learned in high school geometry), but they don’t necessarily constitute a trend. 

Ticket-splitting is actually a process that takes place at the level of the individual voter, so survey data constitute a more appropriate test. Has the number of ticket-splitters changed? According to the American National Election Studies, not very much. In 2008, ANES found 17 percent of voters splitting their tickets between president and House (the chamber they track). That is down from a high of 30 percent in 1972, when Dixiecrats still abounded and George McGovern topped the ticket. But the 2008 level matched that in 2004 and 2000 and was higher than in any year from 1952 (the start of the time series) through 1968. No evidence of a clear, sustained downward trend.

Voters proved quite adept at splitting their tickets in individual races. In split-happy Montana, exit polls tell us Sen. Max Baucus (D) garnered the votes of 94 percent of those who cast ballots for President Obama but also won a 52 percent majority of those who voted for John McCain. In 2004, Byron Dorgan (D), running in North Dakota, the queen of split decisions, accomplished a similar feat, capturing 98 percent of Kerry voters and 52 percent of Bush’s. An unfair test, you say — longtime entrenched incumbents.

In Louisiana, one of the most conservative states in the country, Sen. Mary Landrieu (D) took 96 percent of Obama voters and 24 percent of McCain supporters, enough to win. In an open-seat contest in Colorado, Ken Salazar (D) piled up 93 percent of Kerry voters on top of 11 percent of Bush supporters to eke out a victory.

Ticket splitting is not a one-way street. Even as Obama was cruising to victory in Maine, Sen. Susan Collins (R) picked up 40 percent of his voters while holding 94 percent of McCain’s.

Partisanship remains the central organizing principle of politics, so most voters vote with their party, most of the time, yielding lots of straight ticket voting. However, even in an era when partisan tides run strong, voters seem perfectly capable of distinguishing the candidates they like from those they don’t, regardless of how they vote for president.

Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982. Current clients include the Majority Leader of the Senate and the Democratic Whip in the House.