Clinton, Giuliani, and the Polls

John Harwood's story in the Wall Street Journal is headlined "Poll Suggests Clinton Is Vulnerable." The key finding is that although adults want a Democratic president rather than a Republican president by a margin of 50 to 35 percent, they favor Hillary Clinton over Rudy Giuliani by a statistically insignificant 46 to 45 percent. Clinton's lead over Giuliani is down from her previous leads in NBC/Wall Street Journal polls of 49 to 42 percent in September, 47 to 41 percent in July, and 48 to 43 percent in June.

This improvement in Giuliani's standing versus Clinton's is reflected by a similar improvement in some, but not all, other polls recently: Rasmussen, ABC/Washington Post, Fox News. Why does Clinton run so far behind the Democratic vote?

While a 51 percent majority gives her high marks for being "knowledgeable and experienced enough to handle the presidency," pluralities rate her negatively on honesty, likability, and sharing their positions on the issues.

On honesty, only 34 percent rate her positively, and 43 percent rate her negatively.

Other Democrats, it appears, have different problems as general election candidates. Harwood does not disclose the numbers but reports:

John Edwards and Barack Obama both run even against Giuliani, too, matching Clinton's standing even though they aren't as well known as she is. But Obama would enter a general election with serious vulnerabilities of his own, since just 30 percent of Americans rate him positively on having enough experience for the presidency, and just 29 percent rate him positively on "being a good commander in chief."

The NBC/Wall Street Journal poll has Clinton doing well nationally in the Democratic primary race. She leads Obama 47 to 25 percent, with 11 percent for Edwards (his worst showing in this poll this cycle). Similarly, pollster Scott Rasmussen shows no change in the national numbers for the Democratic candidates since the October 30 debate in which Clinton wobbled on driver's licenses for illegal immigrants. A Marist national poll taken both before and after the October 30 debate found that Clinton's support was 52 percent before the debate and 43 percent after the debate, but the sample size of each group was small, and there may be no statistically significant difference between the groups.

However, Rasmussen's post-debate poll of New Hampshire Democrats shows Clinton leading Obama in New Hampshire by only 34 to 24 percent, with 15 percent for John Edwards. Clinton led Obama 38 to 22 percent in late October before the debate, with 14 percent for Edwards, and in late September she led Obama 40 to 17 percent, with 14 percent for Edwards. Rasmussen also found that only 19 percent of New Hampshire Democrats favor driver's licenses for illegal immigrants—the position taken ultimately by Clinton and by Obama and Edwards as well—while 66 percent are opposed.

This fortifies my conclusion in my Creators Syndicate column that this issue could hurt Clinton, or another Democrat, in the general election. More evidence comes from the Quinnipiac polls of Pennsylvania and Connecticut released today. In Pennsylvania, Clinton leads Giuliani 45 to 43 percent, down from 48 to 42 percent in an October 10 poll. In addition, Quinnipiac found that 81 percent of respondents oppose driver's licenses for illegal immigrants and that—by a margin of 54 percent to 3 percent— voters would be less inclined, rather than more inclined, to vote for a candidate who favored them. In Connecticut, Quinnipiac has Clinton leading Giuliani by just 45 to 44 percent. John Kerry carried Pennsylvania by 51 to 48 percent and Connecticut by 54 to 44 percent in 2004; Giuliani against Clinton seems to be running at the Bush 2004 level in Pennsylvania and significantly ahead of it in Connecticut.

All of which leads me to think that Scenario A for the Democratic race is somewhat less likely and Scenario B is somewhat more likely. Scenario A is that Clinton wins a narrow victory in Iowa on January 3, as suggested by recent polls there, and then wins a stomping big victory in New Hampshire on (presumably) January 8 and is effectively the Democratic nominee on the morning of January 9.

But Iowa polls are particularly dicey, since it's hard to get a sample that represents the relatively small percentage of the electorate that will turn out for the precinct caucuses. And much of Clinton's appeal to Democrats is based on the notion that she is a strong general election candidate, the notion challenged by John Harwood's story. In Scenario B, Obama, who has more organizers on the ground in Iowa now than Clinton does, wins in Iowa and gets a bounce and wins in New Hampshire. The drop in Clinton's current margin in New Hampshire suggests her support is not rock-solid there. Scenario B gives us a two-candidate, Clinton-Obama race, continuing at least through the February 5 primaries, by which time half the nation will have had a chance to vote, with the winner, in my view, by no means clear. As the NBC/Wall Street Journal poll suggests, both have weaknesses as general election candidates, weaknesses that may trouble Democratic voters who are eager to win in November.

Last June, I wrote that the Republican primary electorate is fluid and the Democratic primary electorate is viscous. This week's poll numbers have made me think that the Democratic primary electorate may be more fluid than I thought.