Who Does Edwards' Exit Help?

With the news breaking today that Edwards will be leaving the race, one of the questions raised is how this will affect what will now be a two-person race. There are two schools of thought on this. On one hand, some may argue that Edwards and Obama were splitting the anti-Clinton vote and that this vote will now coalesce behind Obama. On the other hand, the constituencies Edwards was appealing to (blue collar workers, for example) seem to line up more with the groups who have been supporting Clinton in the early primaries.

I did a little checking to see if any polls out there could offer a clue as to which candidate this might help, if any. Many pollsters (Pew, for example) have actually asked respondents who their second choice was. However, I couldn't find information immediately available concerning which candidate was the preferred second choice among just Edwards supporters.

However, compliments of pollingreport.com comes information on a national NBC News/Wall Street Journal poll (1,008 respondents, January 20-22) that asks who voters would prefer if Obama and Clinton were their only choices. This was the actual question:

"Suppose the choice for the Democratic nomination came down to Hillary Clinton and Barack Obama. For whom would you vote?"

Clinton received 53% and Obama received 37% when the question was asked this way. In comparison, when Edwards was included in the question, Clinton received 47% and Obama 32%.

Based on this information, it would appear that Edwards supporters divide roughly equally between Clinton and Obama, with Clinton picking up 6% and Obama increasing his total by 5%. Thus, this poll suggests that Edwards' exit may not help either candidate.

Of course, it is important to be cautious with this interpretation. First, the poll was conducted January 20-22, which was before Obama's win in South Carolina and the endorsement from Ted Kennedy. The latter, in particular, may change how Edwards' supporters view the choice between Clinton and Obama. In addition, the survey pushed respondents who said they were unsure toward one candidate or the other by asking which way they leaned. Finally, this was a national survey, and the dynamics may be quite a bit different in some states relative to others.

I'm sure pollsters will release more information like this from their recent surveys throughout the next few days, but this is a reasonable "first cut" at answering the question until more information is available.