• Cover Story
• Table of
  Contents
• Contents By
  Topic
• Columns
  • Brownstein
  • Cook
  • Crook
  • Powers
  • Rauch
  • Stokes
  • Schneider
  • Taylor Jr.
• Regular
  Features
  • Hotline Extra
  • Inside Washington
  • Insiders Poll
  • K Street Corridor
  • People
  • The Week on the Hill

After Hillary Rodham Clinton's decisive win in last week's Pennsylvania primary, Barack Obama and his advisers quickly offered a series of explanations for her resounding advantage among working-class white voters there. In rapid fire, Obama and his team insisted that he had carried those voters in many other states, was improving his performance among them, and did not need them to win a general election; to the extent he faced a problem at all, Obama declared, the difficulty was age and not class.

But exit polls from this year's Democratic primaries show that almost all of those assertions are debatable and some are flat-out wrong. Together, the arguments from Obama and his aides raise questions about whether his campaign is honestly confronting the challenge he is facing with working-class whites--or whether he is in some measure of denial.

Since early 2007, Obama has consistently run better in polls among white voters with college and (especially) post-graduate degrees, while Clinton has led among white voters without advanced education. That division follows an upscale-downscale--or "wine track" and "beer track"--pattern that has frequently characterized Democratic presidential primaries since 1968.

Centered on blue-collar whites, Latinos, and seniors, Clinton's coalition resembles those of previous "beer track" candidates such as Walter Mondale and her husband, Bill Clinton. Obama's coalition of young voters, independents, and well-educated whites resembles the following of earlier "wine track" candidates such as Gary Hart and Bill Bradley, with one key difference. While previous "wine track" candidates struggled with African-Americans, Obama is dominating among black voters.

Obama's new coalition has allowed him to seize the advantage over Clinton in the nomination race. But the durability and magnitude of Clinton's edge among working-class white voters grounds her argument to Democratic superdelegates that she would be a stronger general election candidate.

Obama added fuel to this debate last week by defending his performance among working-class whites during the primaries with three principal arguments; his chief strategist, David Axelrod, then contributed a fourth argument relating to the general election. What follows is a close look at each of those claims.

Obama has carried these voters elsewhere. On the morning after Pennsylvania, Obama told radio host Roland Martin, "We have won the white blue-collar vote in a whole bunch of states." But that's not true, no matter how the white blue-collar vote is defined.

Most analysts define the working class as those voters without a four-year college degree. The Edison/Mitofsky National Election Pool has conducted exit polls in 29 primaries this year, and Clinton has outpolled Obama among white voters without a college degree in 26 of them, according to a compilation provided to National Journal by the NBC News elections unit. Obama has carried white noncollege voters only in Wisconsin, Vermont, and Utah; and Clinton didn't compete in the latter two. Obama has failed to attract even 40 percent of those voters in every other state except New Mexico, Virginia, and his home state of Illinois (where Clinton still narrowly carried noncollege whites).

Obama's campaign prefers to look at voters by income, noting that he has carried voters earning less than $50,000 in more than a dozen states. But those results include his strong performance among African-Americans; his strength among recent college graduates just beginning their careers at lower pay also helps him here. Even so, according to figures provided by NBC News, Obama has carried white voters earning $50,000 or less annually in just four states: the same three where he carried noncollege whites plus Illinois.

Obama's performance with these voters is improving. The candidate declared last week that among groups like "white working-class folks, [we] did better in Pennsylvania than we did in Ohio, so we're continually making progress." But in Pennsylvania, Clinton carried white voters without college degrees by 70 percent to 29 percent, almost exactly her advantage of 71-27 in Ohio, according to the exit polls. In any case, Ohio is a questionable yardstick because it represented one of Obama's weakest performances among noncollege white voters outside the South; his showing with those voters in Ohio and Pennsylvania lagged behind his results in earlier contests such as New Hampshire, California, Connecticut, Missouri, Maryland, New Mexico, and even New York (Clinton's home state). In all the states with exit polls until Pennsylvania, Obama had won a cumulative 31 percent of the vote among whites without college educations, compared with 60 percent for Clinton, according to recent calculations by Gary Langer, polling director at ABC News. Those numbers don't support the argument that Pennsylvania represented progress for Obama.

The problem is age, not class. Obama made that case to reporters the day after Pennsylvania when he said, "Our problem has less to do with white working-class voters" than with the fact "that older voters are very loyal to Senator Clinton." There's a germ of truth in that contention, but only that. In another recent analysis of all the exit polls, Langer found that Obama holds a cumulative 53 percent to 43 percent lead over Clinton among white noncollege voters under 30. But Clinton leads Obama by 19 percentage points among noncollege whites ages 30 to 44 and by 33 percentage points among those 45 to 64. (White working-class seniors preferred her by over 3-to-1.) Age compounds Obama's class problem but doesn't create it.

These voters are already gone. Axelrod, Obama's chief strategist, offered that argument when he told National Public Radio that the Democrats "don't rely solely" on white working-class voters to win anymore because they have "gone to the Republican nominee for many elections." That's largely, though not entirely, correct. Democratic public opinion analyst Ruy Teixeira notes that in the five presidential elections since 1988, Democrats have won only between 38 percent and 44 percent of the vote among whites without a college education (Bill Clinton did twice carry a slim plurality of those voters because of Ross Perot's candidacy).

To win this year, Democrats don't need a majority of those voters, but they probably need to reach the higher end of that recent range, calculates Teixeira, a visiting fellow at the Brookings Institution who has written extensively on the white working class. Teixeira rejects the idea that Obama's weak performance with those voters in the primary season means he couldn't hit that target in a general election against presumptive GOP nominee John McCain. "These downscale whites prefer Hillary," Teixeira says. "Does that mean they are all going to vote for McCain in the fall? I doubt it."

Other Democratic-leaning analysts are more concerned that Obama's coalition may depend too heavily on affluent liberals and lower-income minorities, with a dangerous hole in the middle-income center. "If you look at Obama's vote in Pennsylvania, you begin to see the outlines of the old George McGovern coalition that haunted the Democrats during the '70s and '80s," veteran liberal journalist John Judis wrote in a much-discussed New Republic article last week.

Indeed, some polling has found Obama losing more working-class white Democrats to McCain than does Clinton: A Pew Research Center survey released on May 1, for instance, found McCain winning nearly one-fourth of noncollege white Democrats against Obama. In the survey, Clinton loses about one-sixth of those voters to McCain--roughly the same share that John Kerry lost to President Bush in 2004. Fully one-third of noncollege white Democrats expressed an unfavorable opinion of Obama in the Pew poll.

Obama supporters argue that if he's nominated, he has plenty of time to repair that damage. Even if noncollege whites' hostility persists, they say, he could overcome it with a strong performance among independents, especially those with an advanced education. If anything, Clinton's struggles with black Democratic voters might foreshadow greater risks in November, Obama partisans insist.

From both directions, these contending arguments really raise the same larger question: Who is more likely to recapture in the fall the voters who have eluded them in the spring--Obama or Clinton? Even after the longest nomination contest in a quarter-century, the answer to that is still far from clear.