Archives - Crunching the South Carolina Numbers
January 2008
2008 Race for the White House: Crunching the South Carolina Numbers
Search for:

Home

"The Clinton camp would like to brush off the weekend primary in South Carolina as not really that representative because so many African Americans voted. That in essence was what former president Bill Clinton was saying when he dismissed Illinois Sen. Barack Obama's victory over New York Sen. Hillary Rodham Clinton by comparing it to Jesse Jackson's victories in South Carolina during his runs for the presidency in the 1980s.

But in fact, the numbers don't bear that out. If half as many black voters turned out, Obama still would have beaten Clinton, albeit not by as large a margin. Let's say, for instance, that African American voters represented just 20 percent of the total electorate in South Carolina instead of the 55 percent they did -- closer to, say, their share of the New York Democratic electorate. Assuming each candidate won the same proportionate support within each racial group, then Obama still would have won the primary with 34.6 percent of the overall vote to 32.6 percent for Clinton and 32.4 percent for former North Carolina senator John Edwards.

The reason is that Obama did pretty well among white voters as well. He received 24 percent of the white vote compared with 36 percent for Clinton, according to exit polls sponsored by The Washington Post and other news organizations. So even if the white vote were much larger, as in our scenario above, he would be close enough to Clinton that his overwhelming advantage among African Americans (he won 78 percent of the black vote to her 19 percent) would still be enough to put him over the top.

Now there are all sorts of reasons this back-of-the-envelope math doesn't necessarily mean that much. The dynamics that played out in South Carolina may be much different in other states. Each state has its own particular economic, cultural, political and demographic characteristics, and the campaign dialogue may play out differently elsewhere. Plus, other upcoming states--particularly California and Florida--have large Hispanic populations and Obama saw in Nevada that Hispanic voters broke strongly for Clinton.

But the point is that Obama did not win South Carolina solely because the electorate there was disproportionately African American. Even with some of the most racially charged discussion of the campaign so far, he still extended his appeal across racial lines, at least enough to make a difference. And since he proved in Iowa that he can win even in predominantly white states, it's fair to assume he is not a candidate relegated to one demographic -- at least if he does not let the Clintons marginalize him by making him out to be the Jesse Jackson of 2008.

What is less knowable is what would have happened had John Edwards not been in the race. The South Carolina native won 40 percent of the white vote. Did he draw white votes that would have gone to Clinton had it been a two-person contest with Obama? Did he draw the "change voters" hoping to shake up Washington away from the other perceived change agent, Obama? Or did simply draw those voters who were turned off by the acrid discourse between Clinton and Obama, the "grown-up wing of the Democratic Party," as Edwards tried to term it? Hard to say, of course. The punditocracy would probably lean more toward the idea that Edwards split the white vote and therefore helped Obama. At the moment, both camps are talking with Edwards people in the expectation that he may drop out. Yet he has vowed to stay in through the convention at least to play some sort of kingmaker role.

Looking ahead to Super Tuesday a week from tomorrow, Obama obviously has to be able to win in different kinds of states with different kinds of demographics to beat Clinton for the nomination. Only in Georgia and Alabama will he find dynamics similar to South Carolina's. But he obviously can win his home state, Illinois. Now, with an endorsement coming today from Sen. Edward Kennedy, he has a good chance of winning Massachusetts as well. He could be competitive in places such as Arkansas and Tennessee. He has set up organizations in a lot of the small states that have caucuses, hoping to replicate the strategy that worked for him in Iowa. Clinton presumably wins New York and New Jersey easily enough.

A big test will be California, the motherlode of delegates. Suffice it to say, California is not South Carolina any way you play with the numbers." (Peter Baker, The Washington Post, January 26, 2008)


Comments? Questions? Write me at george@loper.org.