At first glance, North Carolina’s apparent competitiveness seems surprising. Even though Obama won North Carolina by just 14,000 votes in an election he won by more than 7 percent nationally, both campaigns treated North Carolina as a battleground state in 2012, an election which promises to be far closer than 2008. And despite a tied national race, Obama remains within striking distance in North Carolina, where post-debate polls show Romney leading by just 3 points. The three polls with pre-debate counterparts (PPP, YouGov/Economist, and Rasmussen) only show Romney gaining an average of one point, providing added confidence that the race might remain relatively tight.
How could Obama lose 7 points nationally, yet remain close in a state that he won by just 14,000 votes in a historic election? The answer lies in the resilience of Obama’s diverse coalition and the changing composition of North Carolina’s electorate.
For the most part, Obama’s biggest losses have come from predominantly white states where Obama won plenty of moderate, former Bush voters—like Wisconsin, Montana, and Indiana. North Carolina is relatively insulated from Obama’s losses, since most of Obama's gains came from young voters, college educated whites, and African Americans, three groups where Obama's support has remained relatively resilient. Indeed, nearly half of Obama’s ’08 voters in North Carolina were non-white—more than any other battleground state. From this perspective, North Carolina has not moved toward the Democrats, but the rest of the country, where white working class voters play a more central role in the Democratic coalition, has just moved away from Obama at a faster pace. But even if Obama’s support is resilient, it’s not completely impervious to the huge national swing in Romney’s direction. Obama won by just 14,000 votes in 2008, so Romney could win the state if he could peel off even a few of Obama’s supporters.
However, the changing composition of North Carolina’s electorate allows Obama to compensate for at least some of Obama’s losses since 2008, even if it only leaves the president with a narrow path to victory. According to the most recent voter registration data from the North Carolina State Board of Elections, the white share of registered voters has declined to 71.5 percent, from 73.4 percent on Election Day 2008 and 74.2 percent at this time four years ago. Although the data from November 2008 and the most recent report are internally inconsistent (they actually don’t add up, so this could be off*) there are probably around 193,000 more non-white registered voters than there were at this time in 2008, compared to 87,000 new white voters. If these voters are allocated according to the 2008 exit polls, then Obama could gain a net of approximately 100,000 new voters. If these new white voters are primarily young or northern transplants, this estimate could be low. In part as a result, registered Democrats currently hold the same 800,000-vote edge over Republicans that they did at a similar point four years ago (although smaller in percentage terms, 45-32 in ’08 to 43-31 today).
And North Carolina's one-stop voting, which allows new voters to register and vote on the same day, might allow the Democrats to make additional gains. In 2008, the white share of registered voters actually fell from 74.3 percent on October 11 to 73.4 percent by Election Day. Given Obama’s slim 14,000 vote victory, one-stop voting probably provided Obama’s margin of victory—more than 72,000 black voters registered over the final four weeks of the campaign. Perhaps Obama can't expect similarly dramatic gains over the final month, but the North Carolina electorate is likely to get whiter and more Democratic. So far this year, Democratic gains have tracked their improvements from 2008: nearly 50,000 new African American voters registered to vote since August 25--the same increase that Democrats saw between August 30 and October 11 of 2008.
But let’s say that same-day voting and less than 100 percent turnout from newly registered voters cancels out and Obama gets an advantage of 100,000 votes from new voters. Would it give him much breathing room? Yes, but it only provides Obama with a narrow route to victory. Here’s one way to think about it: North Carolina might behave more like a state that voted for Obama by about 2.5 points in 2008 if this year’s new voters had registered in ’08 and voted according to the exit polls. That's about as much as Obama won in Florida—but Obama would still need to run the table.
In 2008, Obama’s North Carolina victory was more dependent on turnout from young or black than any other battleground state. Any meaningful reduction in turnout among either group, and perhaps especially among young voters who appear far less enthusiastic than they were in ’08, would quickly cause much of Obama’s newfound 2.5 point lead to vanish. Romney could then take the lead as old Obama ’08 supporters switch back to their Republican habits. After all, Obama's '08 performance was also bolstered by largely uncontested advertisements that allowed Obama to perform relatively well in rural North Carolina. And this doesn't even account for surging Republican enthusiasm or a far more prepared GOP ground operation.
The polls suggest that this might be Obama's fate in the Tar Heel state. The increasing non-white share of registered voters will allow Obama to compensate for a slight decline in Democratic enthusiasm or slight losses among swing voters, but it probably can’t overcome both—let alone if “slight” becomes “modest.” Whether Obama can overcome his deficit in the polls hinges on the possibility that Obama’s strong ground operation and same-day voting combine to defy predictions of lower turnout. With early voting now underway, we might soon have a better idea of whether that's a realistic possibility. And if Obama makes a stop in Charlotte or Raleigh, that might be a sign that they continue to see a narrow path to victory.