Would Hillary Have Won Iowa Without Edwards?
A close look at the numbers.
August 13, 2008 - 1:42 am
Former Clinton campaign spokesman Howard Wolfson told ABC News Monday that he believes John Edwards’ decisions may have cost Hillary Clinton the Democratic nomination.
His premise: had John Edwards dropped out of the Presidential race in October, 2007, when allegations of his affair with campaign aide Rielle Hunter first surfaced, it is likely Hillary would have won Iowa and ultimately become the nominee.
Bloggers and journalists reacted swiftly to Wolfson’s allegation, almost all of them refuting it, including Chris Bowers at Open Left, Nate Silver at 538, Markos at Daily Kos, the Boston Globe, CNN, the Washington Post and the Los Angeles Times.
Most of them place heavy emphasis on the final polling data available on the eve of the January 3 caucuses which found greater support for Obama than Hillary as a second choice candidate.
Some also rely on caucus entrance polling data and others refer to later election results in other states.
These writers ignore the key premise in Wolfson’s statement. He is addressing what would have happened had Edwards withdrawn from the race when the allegations about his affair with Rielle Hunter first became public on October 10, 2007.
What if rather than making a public denial of the charges the following day, Edwards instead admitted the affair and dropped out, leaving only Clinton and Obama as top-tier candidates for the ensuing two and one-half months? According to Wolfson, that scenario would have resulted in Hillary winning Iowa by the time caucus day came around.
In attempting to assess the validity of Wolfson’s belief, I don’t find the final pre-caucus polling to be that helpful. There are other relevant factors, beginning with the Iowa electoral landscape as it existed in October, 2007.
Edwards announced his candidacy for President in New Orleans on December 28, 2006, by which time he had been a frequent visitor to Iowa for more than year. He made four trips in 2005 and eleven in 2006 .
By contrast, Obama’s first appearance in Iowa was not until September, 2006. Hillary Clinton didn’t make her first trip to Iowa until the end of January, 2007.
In 2007, Obama and Clinton stepped up their visits to Iowa. By caucus day in 2008, Edwards had made 47 trips to Iowa compared to Obama’s 44,and Clinton’s 35.
Had Edwards dropped out after October 10, his early supporters would have had two-and-a-half months to rearrange their thinking, decide on an alternate first-choice candidate and become committed to that candidate. Determining who their second choice would be on January 3 with Edwards still in the race and still their first-choice, may not be an relevant measure.
More significantly, if Edwards dropped out in October, he would not have made his final 11 trips to Iowa. How does one gauge how many supporters he earned as a result of those post mid-October visits? Or who those late supporters have chosen instead of Edwards had he not been in the race?
Another question: Who would the labor unions that endorsed Edwards in mid-October, 2007 have chosen if he left the race that week? Hillary and Edwards at that time had greater union support than Sen. Obama.
Voter registration statistics and the final caucus results must be considered. As to the first, 239,000 Iowans attended Democratic caucuses, compared to 2004 total of 124,000. Many were first time voters in the Democratic caucuses.
Independents and Republicans may have propelled Obama to victory. Contrary to the Rush Limbaugh scenario that pushed Republicans to vote for Hillary, many Republicans switched to vote for Obama because they wanted to defeat Hillary.
Edwards’ early departure from the race would not have resulted in more votes for Hillary among these voters. As to election results the final numbers in Iowa were: Senator Barack Obama: 37.58% Senator John Edwards : 29.75% Senator Hillary Clinton : 29.47%
The county caucus results show Obama did great in the more populated urban areas like Des Moines, Cedar Rapids and Davenport, but together, Hillary and Edwards trounced him in the rural counties and counties with higher percentages of elderly voters.
Obama’s win in Iowa, which is 95% white, have led many to suggest that race wasn’t a factor in the caucuses. But there are indications that in some rural counties, Edwards supporters would have turned into Hillary supporters.
Obama got his biggest numbers in the counties with larger (less rural) populations that have more minorities and fewer elderly than the state wide average. Two examples:
- In Johnson County with 116,000 people , Obama got 52% of the vote to Hillary’s 20% (Edwards got 24%.) Only 8.5% of the population is over 65 (compared to 14.6% state wide), the white population is 89.9% (compared to 94.6% state wide) and there are higher percentages of African-Americans and Asians than state-wide.
- In Polk County, with 409,000 residents , only 11% are over 65 (14.6% state wide), the white population is 89.9% (compared to 94.6% state wide) and the African American population is 5.2% (compared to 2.5% state wide.) It’s Asian population is 3.1% (compared to 1.6% state wide) and its Hispanic population is 6.2% (compared to 3.8% state wide.)
There are similar demographic variations in Scott County, which Obama also won by a large margin. won big.
Finally, examining the entrance polls from the night of the caucuses, these numbers stick out:
- Among Democrats, Edwards was the least favored candidate: 31% voted for Hillary, 32% for Obama and 23% for Edwards.
- 40% of those who identified themselves as conservative voted for Edwards. With Obama being viewed as more liberal than Hillary, these voters might have chosen Hillary over Obama.
- Only 19% had decided to support Edwards more than a month before the caucuses while 32% had decided to support Hillary and 38% had decided to support Obama.
The bottom line is that there is no way to know for sure how the race would have turned out if Edwards dropped out in October.
There are far too many variables and they go both ways — those who argue confidently that it would not have made a difference in the Obama-Hillary battle are as off-base as Wolfson was when he declared that Edwards was responsible for Hillary losing the nomination.