UPDATED 10/30/11 4:30 a.m.

WASHINGTON — Kirk Caldwell has continued his steady surge in the Honolulu mayor’s race, closing to within striking distance of Ben Cayetano just two weeks before Election Day, according to The Civil Beat Poll.

Cayetano led Caldwell, 50 percent to 45 percent, in a survey of 886 likely general election voters on Oahu.1 The poll, conducted on the nights of Oct. 24 to Oct. 26, after early voting had already begun, had a margin of error of 3.3 percent. Five percent were undecided.

Among the 30 percent who said they’d already voted, Cayetano, a 72-year-old retired two-term governor, was leading 54 percent to 45 percent. Early walk-in voting began Oct. 23, and absentee mail ballots were delivered to voters around the same time.

But the gap was much narrower among the 70 percent of respondents who said they hadn’t yet voted but definitely will. Caldwell was down just 48 percent to 46 percent. The race is also slightly tighter among those who say they’re locked in for their candidate, with Cayetano having more of the softer support among those who are leaning one way or another.

“A lot of it is going to come down to turnout,” said Matt Fitch, executive director of Merriman River Group, which partners with Civil Beat on its polls. “We know that Caldwell does better among the likeliest voters, and Cayetano’s best voting footprint is with a large turnout of casual voters who only vote sometimes.

“His peak is behind him, and it’s just a question of whether he can hang on another eight days.”

The five-point overall lead was Cayetano’s smallest in a Civil Beat Poll since he jumped into the race in January. Similarly, 45 percent is the high water mark this year for Caldwell, the 60-year-old former acting mayor. He’s gained in every survey of the race we’ve done.

Dates Cayetano Caldwell
Feb. 26-27 53 percent 17 percent
June 5-7 44 percent 23 percent
July 31-Aug. 2 51 percent 24 percent
Aug. 11 — Actual Primary Election Results 44 percent 29 percent
Sept. 26-28 51 percent 42 percent
Oct. 24-26 50 percent 45 percent

Source: The Civil Beat Poll

The Impact of PRP

The gradual erosion of Cayetano’s support comes amid an organized campaign of attack ads from Pacific Resource Partnership, a consortium of construction union members and contractors that support the Honolulu rail project.

But voters said the commercials have not had an effect.

Civil Beat asked: “During the campaign, a pro-Rail, pro-Caldwell political action committee has been airing commercials accusing Ben Cayetano of allowing ‘pay-to-play’ corruption while he was Hawaii’s governor, and of accepting illegal campaign contributions in the past. Have these commercials made you less likely to vote for Ben Cayetano in the election for Mayor?”

Just 15 percent of respondents said the ads made them less likely to vote for Cayetano, while 78 percent said the ads did not make them less likely to vote for Cayetano. Even most Caldwell supporters said the ads played no role in their decision; 26 percent of those respondents said the ads had an effect versus 68 percent who said it didn’t.

The Hawaii Poll, sponsored by the Honolulu Star-Advertiser and Hawaii News Now, on Sunday showed Caldwell up 53 percent to 42 percent. That poll was conducted mostly before Cayetano filed a defamation lawsuit against PRP on Oct. 22. The Hawaii Poll was in the field from Oct. 15 through Oct. 22, while the Civil Beat Poll was in the field in the days immediately after the lawsuit attracted widespread media attention.

That subtle difference in timing could be significant. Polls are snapshots in time, and neither the Hawaii Poll nor the Civil Beat Poll is intended to be a prediction of election results.

But despite the 16-point difference between the survey results, there are commonalities.

“We’re both showing the same trend, which is Caldwell is getting stronger,” Fitch said of the two polls. “Cayetano hasn’t been able to build on his primary election support. There are a lot of reasons to think Caldwell is going to close in the last week — he’s got the money, he’s got the organization, he’s got the outside groups.

“The difference between the two polls comes down to which is the more accurate demographic profile come Election Day,” he said. “But either way the momentum is certainly with Caldwell.”

Two Surveys, Two Samples

Because Civil Beat uses an automated dialing system, our sample is not allowed to include cell phones. The Hawaii Poll uses live callers and, according to the Star-Advertiser, features roughly a 50-50 mix of cell phones and landlines.

This means Civil Beat’s sample would tend to skew older and more conservative, because it’s young people that primarily have only a cell phone and no landline. To counteract that, Civil Beat weights its results to approximate a likely voter model based on who has actually voted historically. On Oahu, 41 percent of respondents in the new poll were 65 or older and another 33 percent were between 50 and 64 years old.

UPDATED Just 49 percent of the the Hawaii Poll sample on Oahu was 55 or older.

That age disparity could be significant.

In the Civil Beat Poll, Caldwell led 53 percent to 43 percent among voters between 18 and 29 years old, and 53 percent to 42 percent among voters in their 30s. But Cayetano’s biggest advantage was not the oldest voters who make up three-fourths of Civil Beat’s sample.

Instead, Cayetano dominated among voters in their 40s; he led 58 percent to 31 percent there. Caldwell led by a narrow 49 percent to 48 percent among voters between 50 and 64, and Cayetano led 52 percent to 45 percent among voters 65 or older.

That’s fairly similar to how the groups felt about the race when Civil Beat last polled in late September, with Caldwell making gains among voters in their 30s.

UPDATED Another difference between the Hawaii Poll and the Civil Beat Poll is ethnicity.

The Hawaii Poll ethnicity breakdown was 25 percent Japanese, 22 percent Caucasian, 17 percent Hawaiian and 16 percent Filipino. Civil Beat’s new survey of Oahu voters, again based on a likely voter model, had 36 percent Japanese, 36 percent Caucasian, 9 percent Chinese (not included as an option in the Hawaii Poll), 5 percent Hawaiian and 5 percent Filipino.

Again, that relative oversampling of Caucasians could be significant. But because of the groups that were comparatively undersampled it’s not clear how.

In the new Civil Beat Poll, Cayetano holds a major edge among Caucasians, leading 61 percent to 36 percent. But he actually leads by more among Filipinos (57 percent to 29 percent) and almost as much among Hawaiians (57 percent to 41 percent). So replacing Caucasian voters with Filipinos and Hawaiians wouldn’t seem to give Caldwell a major edge.

Instead, Caldwell’s strength is the Japanese community, which backs him 57 percent to 39 percent. Civil Beat actually had more Japanese voters in its new sample than were in the Hawaii Poll.

The ethnic demographic crosstabs are largely unchanged in the last month for the Civil Beat Poll. Cayetano has expanded his edge among Filipinos and Caldwell has expanded his edge among voters of Japanese ancestry.

The Role Of Rail

The dramatic turnaround for Caldwell in the Hawaii Poll also showed up in the poll’s measure of rail support.

The Hawaii Poll showed more voters now support rail than oppose it, a flip from its last survey in July. The Civil Beat Poll found growing support for rail but that it had not yet grown into a majority. Forty-two percent of Civil Beat respondents said they generally support the project, the highest mark in a Civil Beat Poll this year.

Dates Sample Surveyed In Favor Opposed Unsure
Feb. 26-27 Likely general election voters 34% 55% 10%
April 15-17, 22 Registered voters 36% 53% 10%
June 5-7 Likely primary election voters 40% 53% 7%
July 31-Aug. 2 Likely primary election voters 39% 55% 6%
Sept. 26-28 Likely general election voters 38% 53% 10%
Oct. 24-26 Likely general election voters 42% 51% 7%

Note: Results might not sum to 100 percent due to rounding.

Rail remains a major factor in the mayor’s race, with Cayetano pledging to kill the project if elected and Caldwell suggesting the project should be completed. Of the voters who support rail 88 percent back Caldwell and just 9 percent back Cayetano. Of those who oppose rail, 86 percent back Cayetano and just 11 percent back Caldwell.

But not all voters care about rail equally.

Fifty-seven percent of respondents said rail is very important in their choice for mayor, 28 percent said it’s somewhat important and 13 percent said it’s not important. Cayetano led 58 percent to 39 percent among the first group, but Caldwell led 54 percent to 40 percent among those who said rail is only somewhat important and 53 percent to 38 percent among those who said rail is not important.

With a presidential race at the top of the ballot, the general election could attract casual and occasional voters who are less politically active and less ideological on the issue of rail than the single-minded voters in the Aug. 11 primary election.

In that contest, voters backed Cayetano 44 percent to 29 percent for Caldwell and 25 percent for incumbent Mayor Peter Carlisle.

Do Endorsements Help?

Since the primary, Carlisle has encouraged his backers to support Caldwell, and most of them have.

Asked who they supported in the August primary, respondents in the new poll approximated the actual election results. Forty-five percent said they’d voted for Cayetano, 25 percent said they’d voted for Carlisle and 23 percent said they’d voted for Caldwell. Eight percent either said they didn’t vote in August or they’re unsure who they voted for.

Of those who said they voted for Carlisle in August, 78 percent are now supporting Caldwell versus 17 percent who’ve chosen to back Cayetano. Most of those who supported Cayetano (92 percent) and Caldwell (90 percent) in August have not changed their minds.

Carlisle isn’t the only politician to throw his support to Caldwell. U.S. Sen. Daniel K. Inouye made an endorsement and cut advertisements on Caldwell’s behalf.

But it’s not clear if that endorsement had the desired effect. Far more voters (37 percent) say Inouye’s endorsement didn’t change their likelihood of voting for Caldwell than said it made them more likely to vote for Caldwell (21 percent). In fact, the most popular response, at 38 percent, was that Inouye’s endorsement made them less likely to back Caldwell.

Most of those people were Cayetano supporters. In fact, 60 percent of Cayetano backers said Inouye’s endorsement made them less likely to support Caldwell versus 32 percent who said it had no impact. Even among Caldwell supporters, 39 percent said it made them more likely to back him versus 44 percent who said it had no impact. And 15 percent of Caldwell supporters back him despite Inouye’s endorsement, which they said was a negative.

About the Author