Saxby Chambliss has won the Georgia runoff by a 57.4 percent-to-42.6 percent margin with 97 percent of precincts reporting. That's a margin of 14.8 percentage points, far greater than the 49.8 percent-to-46.8 percent margin that Chambliss led by in the November 4 voting, and it's well above the 53 percent to 46 percent that was projected for the runoff on pollster.com.
Chambliss's victory over Jim Martin means that the Democrats will not get 60 seats in the Senate, even if Al Franken somehow manages to overcome Norm Coleman's circa 300-vote lead in the Minnesota recount. Franken's only apparent recourse is to the courts or to the full Senate; I doubt he'll get anywhere in the courts, and I doubt that Barack Obama will want the Democrats to take on a bruising partisan fight to get a 59th seat in the Senate (though labor leaders, eager to pass the card check bill and knowing that Arlen Specter voted to cut off the filibuster against it in the outgoing Congress, may press for that).
What are the other implications? For that, let me take a closer look at the election results and compare the results of the December 2 runoff as reported on the Georgia Secretary of State website with the results of the November 4 election on Dave Leip's election website. There are 159 counties in Georgia, so I'm going to show numbers only for counties that cast more than 20,000 votes for senator on November 4. For each of them, I'll show the increase in Chambliss's percentage in the runoff (it didn't decrease in any of them) and the percentage of November 4 turnout that turned out on December 2. I'm grouping them by region and then appending notes for each region.
|Chambliss (% pt.)||Turnout (%)|
|Atlanta metro central city (black majority)|
DeKalb has the most affluent black-majority suburbs in metro Atlanta; northern Fulton County, both Buckhead in Atlanta and the suburbs to the north, is heavily Republican. The evidence here is that black voter turnout remained pretty high relative to white voter turnout, in areas where black political organization has been strong for many years.
|Atlanta metro suburbs (increasing black population)|
The evidence here suggests that Republicans did a better job of turning out voters for the runoffs than Democrats. These counties have less history of organized black politics (because they didn't have many black residents 10 years ago), and the Obama campaign did only a fair job of ginning out turnout (best in Rockdale County, just southwest of DeKalb County). Northern and western Cobb County remain heavily white, very affluent, and very Republican, and voters here, whether because of Republican organization or because of their own motivation, evidently turned out in relatively large numbers.
|Atlanta metro suburbs (small black population)|
(31 of 33 precincts reporting)
What's striking here is the large increase in Chambliss's percentage in almost every county. Turnout, as measured relative to November 4, was highest in very heavily Republican counties: Fayette (69.2 percent for Chambliss on December 2), Walton (79.9 percent), Forsyth (84.9 percent), Cherokee (81.8 percent). Gwinnett County has a large and fast-growing Hispanic population; my hunch is that Hispanics turned out (or the Obama campaign did a good job of getting them to turn out) in big numbers on November 4, but they just stayed home on December 2.
|Northern Georgia (typically very low black percentages)|
|Clarke (Athens, college)||3.0||56|
|Columbia (Augusta sub.)||7.6||58|
(14 of 15 precincts reporting)
Turnout in most of these counties was lower relative to November 4 than in most metro Atlanta counties. Hypothesis: The Obama campaign's success in getting black voters to vote early or absentee for November 4 was not replicated for the Martin campaign. The result was big increases in the Chambliss percentages in many counties, with few Democratic voters showing up at all. Clarke County is a university town county, one of the few in Georgia in my estimate in which a substantial majority of whites vote Democratic. Next-door Oconee County (not on the list because it had fewer than 20,000 voters November 4) is much more Republican, just as heavily white Columbia County is much more Republican than its black-majority next-door neighbor, Richmond County (Augusta is the one metro area in Georgia with a newspaper with a robustly conservative editorial page).
|Southern Georgia (varying racial percentages)|
|Houston (Macon sub.)||7.8||54|
|Bulloch (Savannah sub.)||10.7||50|
(15 of 16 precincts reporting)
As in northern Georgia, turnout was relatively anemic compared with November 4, with the conspicuous exception of Dougherty County—evidence of well-organized black voters, perhaps going back to the 1960s, when Albany was a focus of the civil rights movement. Similarly, turnout in black-majority Macon and Savannah was better, relative to November 4, than in adjacent suburban counties. Not much evidence at all of serious Republican organizational efforts here.
The bottom line: The Obama campaign did a magnificent job of turning out black voters in rural and small-town counties in Virginia, North Carolina, and Georgia for the November 4 election. But it was not able to replicate those results in the Georgia runoff. Black turnout pretty much matched white turnout in the inner Atlanta area, where black political organizations have been active for many years, but it failed to do so in the outer suburbs with increasing black majorities and in North Georgia counties with few blacks. Black turnout did match statewide levels in black-majority cities in southern Georgia, but not enough to outweigh similar white turnout in adjacent suburban counties. As the analysts at NBC News suggest, Obama coattails that were helpful to many newly elected Democrats in the South in November 2008 may not be so helpful to them in 2010 and any special elections that occur between now and then.
That suggests another hypothesis: that the Obama turnout effort among blacks may not be replicable. You can only vote to elect the first black president once.
In contrast, Republicans were able to produce good turnout in affluent suburban Atlanta counties, both those with few blacks and those with growing black populations. This is a countertrend to Obama's good showings in affluent suburban counties in November—showings often far better than any previous Democrat has done since 1964. This occurred even despite Obama's relatively moderate choices for top economic policy positions and his hints that he won't seek tax increases on high earners anytime soon. To be sure, you won't find any suburban counties outside the South that are as heavily Republican as some of the metro Atlanta counties. Forsyth and Cherokee counties voted more than 80 percent for Chambliss. I'm not aware of any suburban counties outside the South that vote 80 percent Republican, and even in 2004 George W. Bush did not win more than 80 percent of the vote in any congressional district in the nation. But the results here do suggest that other Democrats will have a hard time duplicating Obama's percentages in affluent suburban counties. Note that this runoff took place when opinion is very favorable to Obama and when he has been getting credit for bipartisan or at least nonpartisan appointments (Robert Gates, Timothy Geithner).
For the past two years, I have been writing that we are in a period of open-field politics, in which many results are possible. The Georgia runoff result—and the sharp increase in Chambliss's percentage in a four-week period—are evidence that we are still in such a period.