In an earlier post, I explained why the results of U.S. Presidential elections since 1980 closely parallel the results of the elections from 1932 through the '60s. Based on those parallels, I was wondering whether they say anything about midterm elections -- and, come to think of it, there are some parallels here too.
For example, I mentioned in my earlier post that not only are Eisenhower and Clinton similar Presidents, but their first mid-term elections had similar results (Eisenhower's party lost both houses of Congress in 1954, as did Clinton's in 1994). And there's a strong parallel between the 1962 midterms, during Kennedy's Presidency, and the 2002 midterms, during Bush II. In both cases, you had a President who had gotten into office by a very slim margin, and a Congress that seemed to be potentially in play for the opposition party. Under normal circumstances, the President's own party might have been expected to lose in the midterms.
Instead, the Democrats in 1962 picked up several Senate seats and only lost two House seats (astonishingly few at the time; in the '30s, '40s and '50s the President's party usually lost 20+ seats in a midterm), while the Republicans in 2002 gained a few seats in both Houses. There were various reasons for these results, some of them a bit unsavory (quite a few scandals of today are linked to stuff Abramoff, DeLay et al. did in 2002 to help win the midterms), but the main reason was that the President and his party had been given a boost by a crisis that had emerged the previous year -- the building of the Berlin Wall for Kennedy; 9/11 for Bush -- and had a big foreign-policy issue to take directly into the midterms: Kennedy had the Cuban missile crisis in October 1962, while Bush had Iraq and the impending war. This produced, not a big win, but a disappointing result for the other party that solidified the President's party's control over the Congress.
The 1966 midterms were a different story, and not just because there was a different President: there was widespread discontent with the President's domestic agenda, with the war abroad, and with the President's party in general. This led to a big win for the Republicans in the midterms: they didn't manage to take back either house of Congress, but they won a large number of seats in both Houses, effectively crippling Johnson's ability to get anything passed. Now, the parallels are never perfect, of course. A lot of discontent with Johnson's domestic policy revolved around the best things he'd done, like his anti-segregation policies. The discontent with the current administration, on the other hand, is more akin to what happened during the Carter administration: the perception that the President can't get anything worthwhile done even though his party controls Congress. Still, the parallels are there. And while discontent with the war hadn't reached the boiling point in 1966, it was certainly there, to the extent that HUAC opened an investigation into anti-war protesters in the summer of 1966.
So if the parallel holds, and I'm not saying it definitely will, the Republicans would be expected to sustain losses in both Houses while not losing control altogether. That seems like a reasonable prediction, and close to what the professional forecasters are forecasting. The one thing to note is that the margins of control are slimmer now than they were in the '60s, so a relatively small swing of seats (by historical standards) could swing control; however, large swings of seats are relatively uncommon now.
I will add again that if the parallel holds this year, I'm going to be betting real money an Al Gore presidency for 2009.