If, as seems likely but not quite certain, Barack Obama is elected next Tuesday, a key question for public policymaking will be how many Democrats are elected to the Senate. Currently, there are 51 Democrats there, including Joe Lieberman, but Democrats are seriously contesting 11 Republican-held seats, and there is a by-no-means-trivial chance that they could win each one. Meanwhile, Republicans are seriously contesting either zero Democratic-held seats, or only one, that of Mary Landrieu in Louisiana. The only public polls there since July are from Rasmussen, and the latest shows Landrieu ahead of Democrat-turned-Republican John Kennedy by 53 percent to 43 percent. So, Landrieu, a narrow winner in 1996 and 2002, seems headed to a third term.
Two and probably three Republican seats seem certain to be gained by Democrats: Virginia, where Mark Warner is way ahead of his predecessor as governor, Jim Gilmore; New Mexico, where Tom Udall, a reluctant candidate at first, is way ahead of his House colleague Steve Pearce; and Colorado, where Bob Schaffer has not been able to overcome the lead of Mark Udall. I have a lot of respect for Schaffer's campaign manager (and state Republican chairman) Dick Wadhams, but I don't see how he pulls this one off. Four-dollar-a-gallon gasoline gave Schaffer a good issue over the summer, but it has clearly waned in importance. And Colorado, like New Mexico and Virginia, is a Bush '04 state where Barack Obama has had consistent and often statistical leads in the polls since the financial crisis hit on September 15.
A fourth seat most observers expect Republicans to lose is New Hampshire, where Sen. John Sununu has been trailing the Democrat he beat, former Gov. Jeanne Shaheen, back in 2002. Sununu did not lead, or lead by much, in polls six years ago; he is behind in public polls this year, with the most recent RealClearPolitics.com average at 49 percent to 42 percent. Sununu has always said he would save up his campaign treasury until the end and close strong, and I suppose that's still possible. But the overall environment is very different from 2002. John McCain, winner of New Hampshire presidential primaries in 2000 and 2008, trails Barack Obama in the state by 53 percent to 40 percent in the latest RealClearPolitics.com average.
Then, there's Alaska, a state the McCain-Palin ticket will carry. The polls don't close there until 1 a.m. Eastern time, so it's the last state to report on election night. But Sen. Ted Stevens's conviction this week on seven counts of failing to report gifts looks like it will be fatal to his campaign. The only postconviction poll, from Rasmussen, shows Democrat Mark Begich leading, 52 percent to 44 percent. I think Stevens could save the seat for his party by promising he would resign if his conviction is upheld on appeal to allow a new senator to be chosen in a special election. (Alaska law is not clear on whether the governor can appoint a successor to a vacancy.) But I doubt that will happen.
If Democrats win all these five seats, they'll be up to 56. There are five more races in which no candidate leads by as much as 4 percentage points in the RealClearPolitics.com averages. Democrats have to win all but one of them to get to 60 and, in Mississippi, Roger Wicker, who was appointed to succeed Trent Lott, has run up double-digit leads in the last two polls and seems highly likely to beat former Gov. Ronnie Musgrove.
In Oregon, Gordon Smith has trailed Democrat Jeff Merkley in every poll taken since the collapse of Lehman Brothers on September 15, except for one Rasmussen poll that showed the race tied. Before September 15, almost every poll showed Smith ahead. It looks like he could be carried down by the undertow of the financial crisis.
In North Carolina, Elizabeth Dole trails Kay Hagan, 46 percent to 44 percent. Hagan made much of the fact that Dole did not spend much time in the state in 2005 and 2006, when she chaired the Senate Republicans' campaign committee, and Dole has not led in any poll since early October. But neither candidate has been ahead by 7 points in any poll since July. This looks like a real nail-biter.
In Georgia, incumbent Saxby Chambliss, who seemed utterly safe at the beginning of the cycle, is currently averaging a 46 percent to 43 percent lead against Democrat Jim Martin. There has been heavy black turnout in early voting—not a good sign for Chambliss. However, Martin has not been ahead in any public poll.
In Minnesota, DFLer Al Franken has led incumbent Norm Coleman in most October polls after some fuss over who paid for Coleman's clothes at Neiman Marcus surfaced. Independent Dean Barkley, appointed for a brief time in the Senate by then Gov. Jesse Ventura after the death of Paul Wellstone, has been getting double digits in the polls. The latest Rasmussen poll has Coleman ahead, 43 percent to 39 percent to 14 percent, a turnaround from the week before when Rasmussen had Franken ahead, 41 percent to 37 percent to 17 percent. Coleman reportedly scored well in debate this week, as he did six years ago against Walter Mondale after Wellstone's death.
In Kentucky, Senate minority leader Mitch McConnell faces a tough challenge from self-financer Bruce Lunsford. Recent polling shows McConnell leading, 48 percent to 45 percent, and Republicans say their internal polling has him doing better in a state that John McCain seems sure to carry by a comfortable margin.
What's my bottom line? If I had to bet $1,000 on each of these races, I would bet on Smith and Dole to lose, and Coleman, Chambliss, and McConnell to win. That, assuming Sununu doesn't somehow pull it out, would leave the Democrats with 58 seats. (But I could easily be wrong on any or all of these races, and I reserve the right to change my prediction before Tuesday.) Fifty-eight Democrats would be enough to stop filibusters if they can get a couple of Republicans (and not drop any Democrats) on an issue, but not enough to run the table.
It's a little scary to think that major differences in public policy can be settled by the outcomes in just a few close Senate races. But then, major differences in public policy were settled by George W. Bush's paper-thin victory in 2000. Our representative democracy gives both parties huge incentives to squeeze just a few more votes out, because they can make a huge difference in the long run.