On the eve of the third and final presidential debate, it's a good time to take a step back and assess how the race has changed since very late September-that is, since right before the first debate.
At that time, Bush was running a consistent lead over Kerry though, as I argued repeatedly, the magnitude of that lead was likely fairly modest, despite the gaudy results obtained by some public polls.
Not only that, Bush was continuing to display a number of underlying weaknesses that made even that small lead quite vulnerable. As Guy Molyneux pointed out in his excellent article, "The Big Five-Oh,"
in incumbent elections, the incumbent's percentage of the vote is a far better indicator of the state of the race than the spread. In fact, the percentage of the vote an incumbent president receives in surveys is an extraordinarily accurate predictor of the percentage he will receive on election day-even though the survey results also include a pool of undecided voters. Hence the 50-percent rule: An incumbent who fails to poll above 50 percent is in grave jeopardy of losing his job.
And, before the debates, Bush was consistently averaging under 50 percent of the vote in trial heats. Not only that, but
polls in [the battleground] states actually reveal an even more precarious position for the president. Taken together, Bush receives a bit less support in these critical states than in the nation overall. In the latest NBC/WSJ poll, Bush receives 49 percent support nationally but only 47 percent in the battleground states, a typical finding. (Bush and Al Gore split the vote in these states evenly, 48 percent to 48 percent.)
More importantly, if we take an average of recent published polls of registered voters in individual states, Bush falls short of the 49-percent benchmark in nearly every one, including Ohio (47 percent), Florida (47 percent), and Pennsylvania (46 percent). Wisconsin (51 percent) is the only crucial battleground state in which Bush appears to have a fairly solid lead. Bush even fails to clear the 49-percent bar in such 2000 Bush states as West Virginia (47 percent), Missouri (49 percent), and Arkansas (48 percent).
The root of Bush's weak support in these terms was pretty simple: people still thought he was doing a lousy job running the country, especially in key areas like the economy, Iraq, and health care. These indicators stubbornly refused to budge during the entire time Bush was maintaining a lead.
In sum, Bush was ahead before the first debate not because his campaign had succeeded in convincing voters that Bush was doing a great job, but rather because his campaign had managed to shift a significant amount of attention away from Bush's poor performance and onto Kerry's alleged character flaws. Therefore, Bush's lead was likely to dissipate as soon as voters' attention was drawn back to his actual performance in office and the concrete policy alternatives proposed by Kerry.
That is, in fact, what has happened. The debates have allowed Kerry-Edwards to refocus the campaign around Bush's record and Kerry's alternatives, thereby taking advantage of the weaknesses Bush never managed to fix in August-September. This was particularly true of the first debate where Kerry's strong performance put Bush on the defensive in what was supposed to be his area of strength: foreign policy.
The immediate post-debate polls all showed Kerry a clear winner: 43 percent to 28 percent among uncommitted voters (Knowledge Networks for CBS News); 45 percent to 36 percent (ABC News); 45 percent to 32 percent (Democracy Corps); and 53 percent to 37 percent (CNN/USA Today/Gallup Poll). And, across the latter three polls, no matter Kerry's overall margin, he always won by more among undecided and independent voters. For example, in the ABC News poll, he won independents by twenty points, and in the DCorps survey, he won undecided voters by thirty-one points.
Moreover, by the weekend, when the debate had had a chance to "settle" in the public mind, Kerry's winning margin widened dramatically-to 57 percent to 25 percent in the Gallup Poll and a crushing 61 percent to 19 percent in the Newsweek poll.
The most important result of the subsequent vice-presidential debate was not who won overall (where Cheney had a slight edge) but the way that Edwards kept the focus relentlessly on Bush's record and Kerry's alternative proposals and repeated Kerry's success in reaching undecided voters.
The second presidential debate was another Kerry win, albeit less spectacular than the first: 44 percent to 41 percent (ABC News); 47 percent to 45 percent (CNN/USA Today/Gallup); and 45 percent to 37 percent (Democracy Corps).
As before, Kerry won easily in all these polls among independents and undecided voters. And, as before, his overall winning margin has widened as impressions of the debate have settled among voters (the latest Gallup poll, for example, has his winning margin rising to fifteen points from Gallup's initial debate night margin of just two points).
These debates, and the dynamic they set up, have transformed the race from a referendum on the challenger and his character to a referendum on the incumbent and his record. And, in the process, voters have received a lot of direct, unmediated exposure to John Kerry and his ideas that has been nothing but helpful to his candidacy, leading more voters to conclude that he is an acceptable alternative to a poorly performing incumbent.
Here are a number of indicators of how much the race has changed:
1. The horse race. Even if one accepts all current polls at face value-that is, making no adjustments for any possible over-representation of Republicans-polls taken in the last week indicate, on average, a dead heat (using 2-way registered voter (RV) data where possible; 3-way and/or likely voter (LV) data where not). In other words, Bush's pre-debate lead has been completely eliminated. Moreover, Bush's average support level since the first debate has only been running at 47 percent, a very bad sign for an incumbent (see the "50 percent rule," discussed above).
And we find particularly sharp swings toward Kerry among the very public polls that had given Bush his largest pre-debate leads: Gallup shows a thirteen-point swing toward Kerry among RVs; Ipsos-AP, a nine-point swing toward Kerry among RVs; and Newsweek, CBS News and ABC News, seven-point swings toward Kerry among RVs.
2. Favorability ratings. Kerry now leads Bush in favorability in most recent polls. For example, Newsweek has Kerry at 52 favorable/40 unfavorable, compared to 49 percent /46 percent for Bush. Similarly, Time has Kerry at 50 percent /34 percent, compared to 48 percent /42 percent for Bush, while Gallup has Kerry at 52 percent /44 percent and Bush at 51 percent /46 percent.
3. Job ratings. Bush's job ratings, never very impressive, even when he was leading, now are sinking further. The most recent Gallup poll has his approval rating at 47 percent, his worst rating in that poll since July and Newsweek has it at 46 percent, also his worst rating since July. His ratings on Iraq, the economy, and foreign policy are also headed south.
4. Bush versus Kerry on the issues. Across polls, Kerry shows substantial gains on every issue. Where he was leading before the debates, he now leads by more. And where he was losing to Bush, he is now losing by less. For example, in the Newsweek poll, Kerry is now favored by thirteen points on the economy, but was only favored by two points before the first debate and is now favored by twenty-two points on health care/Medicare, compared to ten points before the first debate. And while Bush is still favored in this poll by five points on Iraq and by two points on foreign policy, before the first debate he was leading Kerry by fifteen and sixteen points, respectively.
Of course, domestic issue areas are Kerry's strength and Gallup's latest poll provides some eye-opening data on the extent of that strength. Here is Kerry versus Bush in ten domestic issue areas polled by Gallup (thanks to Alan Abramowitz for bringing these data to my attention):
The environment: Kerry 60 percent, Bush 31 percent
Stem cell research: Kerry 53 percent, Bush 33 percent
Health care: Kerry 56 percent, Bush 37 percent
Medicare: Kerry 53 percent, Bush 38 percent
Federal budget deficit: Kerry 53 percent, Bush 40 percent
Social Security: Kerry 50 percent, Bush 41 percent
Education: Kerry 50 percent, Bush 43 percent
The economy: Kerry 49 percent, Bush 45 percent
Abortion: Kerry 46 percent, Bush 42 percent
Taxes: Kerry 44 percent, Bush 51 percent
Average of all ten domestic issues: Kerry 51.4 percent, Bush 40.1 percent
5. Bush versus Kerry on candidate attributes. Same story: substantial gains for Kerry on every attribute, widening his lead on attributes where he was already leading and cutting his deficit where he has been trailing Bush. In the Time poll, Kerry widened his lead to nine points on understanding people's needs (up from a four-point lead), tied Bush on honest and trustworthy (up from an teight-point deficit to Bush) and took a five-point lead on having good judgement (up from a four-point deficit to Bush). Kerry has even opened up a five-point lead on being "likeable" (up from a four-point deficit to Bush).
And, critically, both the Time poll and the new Gallup poll now show Kerry leading Bush-albeit by slim one- to two-point margins-on who has clear plans to solve the nation's problems.
The task for Kerry seems clear as he heads into the third debate and the final weeks of the campaign: keep the heat on Bush's terrible record and keep telling voters-in the simplest possible terms-how he would do a better job. The voters, it would seem, are starting to listen.
A Note on the Washington Post/ABC News Tracking Poll
Alan Abramowitz points out:
<http://en.groundspring.org/EmailNow/pub.php?module=URLTracker&cmd=track&j=10391676&u=90428>The Washington Post tracking poll seems to be suffering from the same ailment that afflicted the Gallup tracking poll four years ago, albeit on a smaller scale so far. In the past few days we have seen Bush's lead among registered voters shrinking while his lead among likely voters has increased. This means that the likely voters and the unlikely voters are moving in the opposite direction, just as they frequently did in the Gallup tracking poll four years ago. This makes no sense, of course. With the WP tracking poll, as with the Gallup tracking poll, the registered voter results are probably a better indicator of the actual standing of the race.
Well said. It's also worth noting that, in 2000, the ABC/Washington Post tracking poll missed the final vote pretty badly, having Bush up by three points at the very end and three to four points up on every night of the final week. Looks like they're poised to repeat their fine 2000 performance.
Ruy Teixeira is a Senior Fellow at The Century Foundation and a Fellow at The Center for American Progress.