Midterm elections are different from presidential elections because turnout is much lower. I’d argue that midterm elections can be further distinguished depending on whether they occur during a president’s second year in office or his sixth. Since World War Two ended, we have had four midterm elections in the sixth year of a presidency and two additional midterm elections that almost fit that model. In 1958, Eisenhower’s sixth year in office, the Republicans lost 48 seats in the House. In 1986, Reagan’s sixth year in office, the Republicans lost five seats in Congress. In 1998, Bill Clinton’s sixth year in office, the Democrats gained five seats in the House. In 2006, Bush’s sixth year in office, the Republicans lost 30 seats in the House. In 1974, Richard Nixon resigned in disgrace and the Republicans lost 48 seats. In 1966, which would have been JFK’s sixth year in office, the Democrats lost 48 seats.
This is a small data set, but the trend is pretty clear. The president’s party doesn’t do well in the sixth year of his presidency. In the post war era, the only time the president’s party gained seats in the sixth year was in 1998, and that was pretty clearly a backlash against the impeachment proceedings again President Clinton. Yet, if you look at all these examples, they all have unique circumstances that explain the results. In 1958, the booming economy we associate with that decade came to an abrupt halt and went into recession. In 1966, there was a backlash against the civil rights legislation of 1964 and 1965, and there was growing consternation about the war in Vietnam. In 1974, people were furious about President Ford’s pardon of Nixon. In 1986, there was the Iran-Contra scandal. In 2006, there was exhaustion with the war in Iraq.
I think it is a mistake to think that we can use these past sixth-year midterm elections to make confident predictions about what will happen in 2014. If the president is staggering through some massive scandal, his party will probably do badly unless (as in 1998) the public perceives the Republicans to be overreacting.
The truth is, no post-war president has had a good second term. The least bad second term was Eisenhower’s. He had a bad economy and got caught lying about the Gary Powers U-2 incident, but that scandal occurred after the midterms. I think we have to go back to FDR’s second term to find an example that is likely to fit Obama’s second term. But even that example won’t be helpful in predicting the congressional elections of 2014. In 1938, the Democrats lost 72 seats and still didn’t lose their majority. The last time a president went into their sixth-year midterms with a split Congress was in 1986. Reagan’s party lost control of the Senate.
I think the best guide for predicting the 2014 elections is to look at each individual seat and its likelihood of changing hands. From that standpoint, the House looks very stable. There is almost no chance of a wave election where 48 seats change hands. There are many more seats that the Democrats won narrowly than that the Republicans won narrowly, making it a little more likely for the Republicans to make significant gains. The Democrats will need to pick up 17 seats to regain control of the House. That doesn’t sound all that daunting, but it looks nearly impossible when you begin to look at the individual races.
We need to figure out a way to beat expectations or the House is going to be in Republican hands for the remainder of the decade.
