Politics: Don't Be Certain About Senate

TORONTO, ONTARIO - The prevailing wisdom on the upcoming Federal elections next Tuesday in the United States is that the Republicans will take over control of the House of Representatives, and the Democrats will retain control of the Senate. While many responsible observers emphasize the volatility of the situation--the Republicans could pick up more than 70 seats in the house, or fewer than 40 such that the Democrats maintain control--the election is getting close enough now that big movement is unlikely. Looking at the discrete races involved in the Senate, it strikes me that scenarios in which the Republicans take over are being discounted--because of potential defections.

The election website that I pay the most attention to, Nate Silver's FiveThirtyEight.com, now hosted by the New York Times, currently places the chances of a Republican takeover of the Senate at 12%, compared with 83% in the House. Certainly, 12% is far from zero. But, in his extensive model, Silver only accounts for the elections themselves, in which he projects 52.1 Democrats and 47.8 Republicans as having seats. The model doesn't account for what happens with some sitting senators.

After Scott Brown's win in the Massachusetts special election earlier this year, the Republicans have held 41 seats in the Senate to the Democrats' 59. To take control of the chamber, they need 51, as the tiebreaker, vice president Joe Biden, is a Democrat. Most analysts have seats in North Dakota, Arkansas, and Indiana as certain takeovers for the Republicans. That brings them to 44. If the election were held today according to FiveThirtyEight's models, Republicans would also gain seats in Wisconsin, Pennsylvania, Nevada, Illinois, and Colorado. That takes them to 49. Assuming that Democrats indeed hang on to contested seats in West Virginia, Washington, and California and the rest of the "safe" seats don't result in any surprises, the Republicans would be two seats short of control.

I'm not certain that, in such a scenario, the Democrats would actually hang on to their majority. Democrat Ben Nelson of Nebraska has a very conservative voting record, and there is persistent speculation that he could be convinced to switch parties. He may face a challenge in the Democratic primary in the next election cycle (though considering the TEA party movement he might well face one in a Republican primary as well--there's little room for centrists anywhere, as Arlen Specter learned in Pennsylvania). Considering that he was accused of holding health care legislation hostage for special preference for his state, it does not seem difficult to imagine that he could be enticed to the Republican side if it were the majority.

However, that only gets the Republicans to 50. They would still need one more defection, and it's pretty obvious who that might be. "Jolly Joe" Lieberman of Connecticut has already shown that he is willing to ignore the Democratic party, running as an Independent (later "Independent Democrat") in his last election after losing in the Democratic primary. He almost became the Republican nominee for vice president with John McCain in 2008--most reports say McCain wanted him and Lieberman was willing to do it, but the Republican establishment wouldn't accept it. Some incentive concerning the pharmaceutical industry in Connecticut, and it would be easy to believe that Lieberman would caucus with the Republicans instead of the Democrats, and with both Nelson and Lieberman, the Republicans would have control.

This scenario seems entirely plausible to me. The Democrats can still maintain control by winning more seats than currently predicted (as finding more than two senators who would be willing to change parties seems far-fetched), but if things play out and eight senate seats change party hands, I strongly suspect it will result in Republican control of the Senate, not the slim Democratic majority it would seem on the surface.