Romney now holds a slim, seven-point edge among registered voters when it comes to handling the economy, even as there is also limited faith that things would quickly get better if he was to win. Obama counters with an equivalent advantage over Romney when it comes to who is seen as having a better understanding the financial problems people are facing.
Obama’s campaign has spent freely to portray Romney as a businessman more interested in profits than creating jobs, and one who enriched himself even as his company, Bain Capital, was closing companies and shedding jobs. In the new poll, nearly six in 10 voters say Romney, as president, would do more to help the wealthy than the middle class. For Obama the numbers are basically reversed, with about six in 10 saying he does more to help the middle class.
The president’s advantage in helping the middle class is key and gives him some push-back when it comes to voters’ views on the size of government. By 54 percent to 36 percent, voters are more apt to say that unfairness in the economic system is a bigger problem than government overregulation of the private sector. By a similar split, voters prefer a smaller government to a larger one and overwhelmingly see Obama as favoring a more robust role for Washington.
On 13 issues and attributes tested in the poll, Obama has double-digit leads among voters on three: social issues, women’s issues and being more friendly and likable. On that last attribute, likability, Obama’s lead is a still-dominant 34 percentage points — 61 percent to 27 percent.
In addition to his narrow edge on the economy, Romney has a clear edge when it comes to dealing with the federal budget deficit. On all the others, the two are about even or one is ahead by an insignificant margin.
Another indicator of why the race for the White House remains so close is the public’s judgment of Obama’s overall performance. In this latest poll, his approval rating is 50 percent positive, 46 percent negative among all Americans. Among registered voters, 47 percent saying they approve and 50 percent say they disapprove.
The near-even appraisal continues to put Obama in a tenuous position for reelection — neither strong enough to virtually ensure victory nor low enough to presage a loss.
Obama’s predecessor, George W. Bush also had a 50 percent approval rating at this time in 2004. And it is Bush who provides a boost to Obama now — more than half of all voters see Bush, not Obama, as the one more responsible for the economic challenges the country faces.
The president has other advantages, too. On the choice of whether to create jobs by cutting taxes or enacting more government spending programs, spending holds a 14-point advantage among voters. Even among those who would prefer a smaller government, more than a third say spending is a better way for the government to try to spur job growth, as opposed to tax cuts.
One advantage that Obama held in previous polls — a big lead on enthusiasm — is no longer evident. In the new poll, 48 percent of Obama’s backers say they are “very enthusiastic” about his candidacy, compared with 42 percent of Romney backers who feel that way about the Republican’s campaign — a difference that is not statistically significant. Obama had a 13-point edge in July in this measure; his lead was 25 points in May.
Newfound enthusiasm for Romney probably stems in part from his selection of Ryan as his running mate. Very conservative voters overwhelmingly approve of his pick and are now more likely to be “definite” GOP voters. For the first time, enthusiasm among conservative voters who back Romney rivals that of liberals for Obama.
The latest poll points to demographic divisions that have defined the Obama-vs.-Romney contest all year. Prominent among them are a gender gap, with 49 percent of female voters backing Obama and 43 percent supporting Romney, while men break 51 percent to 42 percent in favor of Romney. White voters break for Romney by 18 percentage points; non-whites go for Obama by better than three to one.
The telephone poll was conducted Wednesday through Saturday among a random national sample of 1,002 adults, including 857 registered voters. The margin of sampling error for both samples is plus or minus four percentage points.
Peyton M. Craighill and Scott Clement contributed to this report.