Sabato's Crystal Ball

The Republicans’ Electoral College Newt-Mare

Larry J. Sabato, Kyle Kondik and Geoffrey Skelley, U.Va. Center for Politics January 26th, 2012

What can we expect from the Florida primary? What are the possible repercussions of a Gingrich victory versus those of a Romney win in the Sunshine State? Check out our latest Crystal Ball video by clicking on the picture. Below, we examine the Republican electoral situation and what a Gingrich candidacy might mean:

In the aftermath of voting in the deeply conservative South Carolina primary, one voter quoted by The Washington Post seemed to summarize what Republicans there were looking for in a candidate:

Many voters said they liked the scrap in the former House speaker’s personality — his willingness to seek confrontations with his GOP rivals and the news media. That, they said, was what the GOP would need in the race against President Obama this fall.

“I think Mitt Romney is a good man,” said Harold Wade, 85, leaving a polling place in this picturesque seaside suburb outside Charleston. “But I think we’ve reached a point where we need someone who’s mean.”

And Gingrich, he said, was the only one mean enough.

Republicans don’t just disagree with President Barack Obama; they viscerally dislike him. And, as Mr. Wade indicates, they don’t just want a candidate who can beat Obama. They are seeking someone who will unequivocally bash and bloody him.

As has been clear from recent debates, Mitt Romney is not that candidate, at least compared to the South Carolina winner and this week’s poll leader in Florida, Newt Gingrich. On the stages in South Carolina, Gingrich mastered both the moderators and the crowd, and Romney did himself no favors with his dissembling answers to repeated requests to release his tax returns. (Romney finally released his most recent tax returns on Tuesday.) By contrast, when Romney finally went on the offensive against Gingrich in the first Florida debate on January 23, his quizzing of Gingrich was mainly mild-mannered, as though he were building a genteel case to bring to his country club’s decorum committee.

The presumption among the press — and among us analysts — for months has been that Mitt Romney was far and away the most electable candidate in the fall against President Obama. But the actual voters, at least in South Carolina, felt differently: in exit polling, among the 45% who said that the most important quality in a nominee was the ability to defeat Obama, they preferred Gingrich 51% to 37% over Romney. Granted, considering how South Carolina Republicans broke for Gingrich, such a result from the exit polling wasn’t really surprising: when supporting a candidate, voters likely don’t think that their guy is a loser. Plus, we also wonder — just anecdotally from interviews aired after the election — whether voters were conflating Gingrich’s ability to win in November with his ability to win debates. Obviously, those are two vastly different things, and with only a few exceptions — 1960’s Nixon/Kennedy debates and 1976’s Gerald Ford gaffe on whether the Soviet Union controlled Eastern Europe — general election debates are not all that vital to the outcome anyway.

A tale of two maps

To be clear, President Obama is vulnerable. His approval rating – 46% approve, 48.6% disapprove, according to the RealClearPolitics approval average — is middling. And against Mitt Romney, Obama’s vulnerability shows: in Pollster.com’s average of national polling, Obama leads Romney just 46.5% to 45.3%. But against Gingrich, the average tells a very different story — Obama, at 49.9%, leads Gingrich (39.1%) by more than 10 points.

Granted, there are outliers: Gallup on Tuesday had both Romney and Gingrich running evenly with Obama, both down 50% to 48%. But the bulk of the polling data indicates that Romney, at least at this point, is the stronger candidate against Obama.

Why does Romney fare better? Well, the best candidate against Obama remains “generic Republican” — that make-believe person who leads Obama by a single slim point, according to RealClearPolitics. Let’s be honest here: Romney’s the closest thing out there to a generic Republican available. He is not going to steal the presidency away from the incumbent if Obama’s having a good year and the economy is solid. Rather, if the country is ready to make a change, then Romney would be a credible alternative. The national polling numbers bear that out; they also show that Gingrich, at least right now, is not seen as a similarly acceptable alternative.

Let’s look at this a different way. Chart 1 shows the Crystal Ball’s Electoral College map. At this point, it is effectively a map pitting Obama vs. a generic Republican; such a race, we believe, would be close, given the president’s relative weakness.

Chart 1: Crystal Ball Electoral College map, Obama vs. generic Republican

Romney, were he the nominee, could potentially make this generic map even better for Republicans. He has no obvious weaknesses in any of the swing states, at least as measured by head to head polling, which we fully grant is not necessarily all that predictive at this stage. Also, he could potentially make some of the competitive states harder for Obama: he could mobilize the not-insignificant Mormon population in Nevada to help in that state and play off his relative home field advantages in Michigan (his father was governor of the state he grew up in) and New Hampshire (he served as governor of neighboring Massachusetts and has built deep bonds with the Granite State, which paid off in his recent primary romp there). In fact, if Romney were the nominee, we’d likely switch New Hampshire to “leans Republican” from toss-up. While this only represents four electoral votes, consider this: take another five Obama states that went for George W. Bush twice (Florida, Indiana, North Carolina, Ohio and Virginia) and put them in Romney’s column, and then add New Hampshire and all of John McCain’s states, and Romney is president.

Unlike Romney, we don’t see any competitive state where Gingrich would potentially perform better than a generic Republican. And, based on polling and our own thinking about the individual state-by-state contests, at this point we think he’d do worse than a generic Republican or Romney.

Again, polling at this early stage is not particularly trustworthy, but the differences between Romney and Gingrich head to head vs. Obama are telling. For instance, in Ohio, a recent Quinnipiac poll had Romney down only two — but Gingrich down by 14. In December in Colorado, Public Policy Polling showed Obama up only two on Romney, but up eight on Gingrich. A Quinnipiac survey of Virginia showed Romney up 2… and Gingrich down five.

Chart 2 shows the Obama v. Gingrich map, which is, needless to say, a nightmare for Republicans.

Chart 2: Crystal Ball Electoral College map, Obama vs. Gingrich

Notice, in particular, how this second map is much bluer in the crucial Midwestern states than the first map. Ohio, perhaps the nation’s key swing state, begins in the leans Democratic column. We also believe Gingrich would have little appeal to Hispanic voters, who are major voting blocs in a number of western swing states. Colorado, Nevada and New Mexico, thus, are more Democratic in this map.

Henry Barbour, nephew of the former Mississippi governor and Republican National Committee Chairman Haley Barbour, drew flak recently for saying that he couldn’t support Gingrich because he said he liked having a Republican speaker of the House; that is, Gingrich would do poorly enough so that Obama’s coattails would elect a Democratic House.

With this map, one can understand his fears. In several Midwestern swing states, Republicans gained many House seats in 2010 that, thanks to their control of redistricting, they sought to lock in through aggressive remapping. Under this map, all of those states — Michigan, Ohio, Pennsylvania and Wisconsin — would be tough territory for Gingrich. If his candidacy were a disaster, those new Republican gerrymanders could unravel. The close battle for the Senate could also be affected — Michigan, Missouri, Montana, Nevada, New Mexico, Ohio and Virginia all have competitive Senate races this year, and all of those states get bluer on our map under a hypothetical Gingrich candidacy.

Gingrich as Reagan?

During debates, Gingrich likes to say that Ronald Reagan was down 30 points to Jimmy Carter in early 1980. He’s correct about that, although Carter, at that point in the race, was getting a sizable poll bounce in the aftermath of the taking of American hostages by the Iranians. Think of it as something like the poll boost George W. Bush got after 9/11 — a rally around the flag effect. By the end of March, Carter’s approval rating was back under 40% after shooting up to nearly 60% — and he led Reagan 40% to 34% in the national Gallup poll, with independent Republican John Anderson taking 21%. As spring melted into summer, and summer into fall, Anderson fell off, and Carter and Reagan fell into essentially a dead heat. Then, when negotiations to free the hostages fell apart the weekend before the election, the bottom fell out for Carter, and Reagan won by 10 points.

Is Gingrich Reagan? Probably not. And the electorate is a lot different and more partisan than it was 30 years ago. And despite Obama’s troubles — he hasn’t had a real polling bounce since Osama bin Laden was killed in the spring — Obama leads Gingrich convincingly in many polls. Much more convincingly than he leads Romney, if he leads him at all.

Of course, this is all highly speculative right now. If circumstances permit — if some unexpected crisis damages Obama, or the economy dips back into recession — Newt Gingrich could be giving his inaugural address roughly one year from now. That’s how important external factors are in deciding the presidential contest.

But don’t doubt for one instant that top Republicans in Washington and across the country haven’t already gamed out their own maps in fearful anticipation of a Gingrich presidential candidacy. And, if Gingrich wins Florida, those same Republicans will be trying to revive Romney or even casting about for a new candidate, however unlikely that scenario may be.

Do Endorsements Matter?

Larry J. Sabato, Director, U.Va. Center for Politics January 26th, 2012

Note: This article originally appeared in the Jan. 24 edition of The Wall Street Journal.

Do endorsements matter? Politicians certainly think so, and they spend loads of time courting party elites and opinion-makers. So far, though, 2012 has shown how the politics of anointment and appointment can fail.

South Carolina Gov. Nikki Haley flopped mightily in trying to deliver her state for Mitt Romney. Evangelical leaders held a summit to get the Palmetto State to back their new choice, Rick Santorum, but he fared much worse than Romney. Newt Gingrich knows the feeling — New Hampshire’s supposedly dominant Union Leader newspaper huffed and puffed for Newt and got him less than 10% of the vote.

But it’s easy to cherry-pick examples to prove the folly of endorsements. In some circumstances, they can make a substantial difference.

Throughout American history, presidencies have been created by the laying on of incumbent hands. Thomas Jefferson effectively passed the presidency to his friend and confidant, James Madison. Andrew Jackson handed his populist democracy off to an unlikely dandy, Martin Van Buren, in 1836. Few would have imagined the studious and portly William Howard Taft as president until Theodore Roosevelt picked him in 1908. More recently, George H.W. Bush might not have been elected president without Ronald Reagan’s blessing. Madison, Van Buren, Taft and Bush all got their predecessor’s third term — when popular, presidents have extraordinary powers.

What about little-known state legislators and local sheriffs? Even low-level backing can attract the cameras and generate a positive story. But this can backfire if candidates overplay their hands, as Jon Huntsman did when his campaign hinted at a “major” announcement in Florida. Speculation naturally centered on former Gov. Jeb Bush. Not quite. Huntsman got only his son, Jeb Jr., and the media’s letdown showed in the coverage.

Then there’s Hollywood glitz and glamour — good for attracting donors to fund-raisers, but risky if the stars outshine the candidates in public. When celebrities draw more adoration from the crowd than the fellow on the ballot, the candidate is diminished, not enhanced. Better to have one enthusiastic pop icon in your entourage than a tour bus full of hangers-on. Chuck Norris did wonders for Mike Huckabee in 2008, and now he’s with Newt. As they say, Norris doesn’t endorse — he tells America how it’s going to be.

Candidates try to use endorsements to convey a sense of inevitability. In the days leading up to South Carolina, Romney rolled out endorsements from Ohio Sen. Rob Portman and Virginia Gov. Bob McDonnell. Odds are, the vast majority of South Carolinians had never heard of either. These vice presidential possibilities probably impressed Romney a great deal more than they did Palmetto Republicans.

Even more than Democrats, Republicans typically nominate a candidate that party elites support. In “The Party Decides,” political scientists Marty Cohen, David Karol, Hans Noel and John Zaller analyzed endorsements made prior to the Iowa caucuses in presidential primary contests from 1980 to 2004. They found that the candidate who had won the biggest share of endorsements won the eventual nomination in nine of 10 competitive contests (the exception was Democrat John Kerry in 2004). On the GOP side, the eventual nominees all won a strong plurality of endorsements.

Not surprisingly, given Romney’s position as front-runner and the fear that many Republican officeholders have of sharing a ballot with Gingrich or Ron Paul, the former Massachusetts governor has a long lead in endorsements from elected officials. According to the Capitol Hill newspaper Roll Call, Romney has the backing of 72 members of Congress, versus a combined 17 for the other candidates.

This is good news for Romney. Gingrich is attempting to stir the populism of the GOP base by railing against “elites,” but many voters welcome guidance in intra-party contests. In a general election, voters have the invaluable shorthand cue of the party label. But in a nominating contest, all candidates have the same party label. How to choose just one? Differences in personality, background and policy help, but so does a candidate’s association with other well-known party figures. People want to puzzle out which candidate comes closest to their kind of Republican or Democrat.

Non-endorsements can send powerful signals to voters as well. For decades, leading Southern Democrats practiced “golden silence” in presidential years, refusing to endorse their party’s presidential nominees. This was a green light to voters that it was acceptable to support a Republican for the White House. In 1960, President Eisenhower wanted Vice President Nixon to succeed him, but he damaged Nixon’s campaign when asked what major decisions in his administration Nixon had influenced. “If you give me a week, I might think of one,” said Ike. The comment ended up in one of John Kennedy’s TV ads.

Could non-endorsements end up mattering in 2012, too? Despite decades on Capitol Hill and four years as speaker of the House, Gingrich has only 11 congressional endorsements, five of them from Georgians. Will rank-and-file Republicans see that as a warning about Gingrich’s volatility and management style — or as a badge of honor indicating the anti-establishment, transformational credentials they seek? The answer may determine this election.

Romney’s Coronation: Just Delayed, or Gone Awry?

Larry J. Sabato and Kyle Kondik, U.Va. Center for Politics January 21st, 2012

This much is obvious: South Carolina has proven to be a disaster for Mitt Romney. The size of his defeat by Newt Gingrich — a 12%+ landslide in a four-way race — is virtually a repudiation of his candidacy in a state that has prided itself on picking the eventual nominee for 32 years. And we suspect Romney will have several more nights of heartburn, much like this one, as the nomination process unfolds.

The surprise for some may be that Mitt Romney remains a strong, though no longer prohibitive, favorite for the Republican presidential nomination. He continues to benefit from an unorganized, inadequately funded, polarizing opposition. The weak Republican field has aided Romney from the start. Still, if there is another South Carolina-style collapse in Florida for Romney, it will have devastating consequences for the frontrunner — and may even bring calls for a new candidate to enter the race. The pressure is on for Romney in the Sunshine State.

Concerning South Carolina, one has to give Gingrich credit for remarkable resiliency. He has risen from the dead not once but twice in this campaign: First, after his campaign imploded in the middle of last year and now, after poor performances in the first two nominating contests. During Gingrich’s first shipwreck in May, his then-spokesman put out a statement saying that “…out of the billowing smoke and dust of tweets and trivia emerged Gingrich, once again ready to lead those who won’t be intimated by the political elite and are ready to take on the challenges America faces.” It was an over-the-top statement, but now seems strangely appropriate after such a wild week of developments in South Carolina.

Granted, this was favorable territory for Gingrich as a conservative, southern candidate from neighboring Georgia, but Romney, as recently as Wednesday, led by 8% in the polls according to the RealClearPolitics polling average. For Gingrich to snap back in such spectacular fashion is a testament to the way he performed in two crucial debates, and in how he reacted to the controversy over his ex-wife’s accusations that he asked her for an “open marriage” before they got divorced. Gingrich handled the accusations about as well as he could have, turning CNN’s John King into a punching bag at the start of Thursday night’s debate and, more than likely, turning a negative into a positive with voters. Conservative Republicans detest most of the news media and rally to support any GOP target of press attacks. In South Carolina, Gingrich received a boost from his framing of the scandal coverage not dissimilar from the polling bounce Gingrich’s one-time nemesis, Bill Clinton, got during his impeachment battle over his affair with Monica Lewinsky. News media “feeding frenzies” can have unintended consequences like this.

The South Carolina exit poll shows Gingrich doing surprisingly well everywhere, with every group, including women who didn’t appear put off by the adultery and open marriage charges. Romney’s team had hoped Newt’s baggage would weigh him down, and tried to add to it by jumping on Gingrich for his “anti-capitalist” targeting of Romney’s work at Bain Capital. Many conservatives around the nation joined in condemning Gingrich’s tactics, and 64% of Palmetto State exit poll respondents said they held a positive view of Romney’s background as an investor. Yet Gingrich’s attacks did not seem to backfire: 31% said Romney ran the most unfair campaign, compared to 25% for Gingrich.

Gingrich’s win brings to a halt any prospect Romney had for ending this contest with a win in Florida on Jan. 31. What Gingrich now needs is for Rick Santorum to drop out of the race so the conservative vote can consolidate. The writing is in capital letters on the wall for the ex-Pennsylvania senator, who has lost out to Gingrich as the leading “anti-Romney” candidate. Santorum may stay in the race until Florida, but he has no path left to the nomination. Just as Gingrich is eyeing Santorum’s anti-Romney voters in Florida, Romney hopes Santorum will survive through the voting there — and won’t become the next Rick Perry, dropping out before Election Day.

Ron Paul, meanwhile, also has no path to the nomination, and his showing in South Carolina was poor. His views on defense, foreign policy and social issues did not sell in this deeply conservative state. Still, Paul won’t drop out soon — or perhaps at all, prior to the convention. As long as he can win delegates — particularly in future caucus states — and be given a national forum for his insurgent campaign through free press and debates, he is likely to continue on. His realistic goals are to influence the GOP platform and secure a primetime speaking slot.

In the first three states we have seen that the proportion of evangelical Christians in the electorate is revealing about, and somewhat predictive of, Romney’s performance. In Iowa, nearly three in five caucus goers were evangelical/born-again Christians (57%) according to entrance polls. In South Carolina, that number exceeded Iowa — 65%, according to exit polls. New Hampshire, where Romney romped, has a much more moderate profile — only about one in five voters there were evangelical/born-again. To use sports parlance, Romney has played two of his first three games in hostile environments.

According to exit polls from 2008, Romney is heading to more favorable ground. Chart 1 shows the exit/entrance polling data from the first three GOP contests, plus information from four years ago about some upcoming states. As is clear from the chart, these upcoming contests are less conservative than Iowa and South Carolina, if not New Hampshire.

Chart 1: Relative conservatism of early caucus/primary states

Notes: *Polls of caucuses are called entrance polls, because they are taken before a caucus participant enters the caucus. Exit polls are taken after a voter votes.

**Other contests from now until the end of February are caucuses in Maine (Feb. 4-11), Colorado (Feb. 7) and Minnesota (Feb. 7). Exit/entrance polling data are not available from these states from 2008; the other contest is the Missouri primary (Feb. 7), but that contest does not award delegates and Newt Gingrich did not qualify for the ballot there.

And, beyond these numbers, Romney has built-in advantages in these upcoming states. He is the only candidate — at the moment — who has the resources to saturate the heavily populated Sunshine State with advertisements. Florida is a much bigger state than the ones that have voted thus far — total turnout in Florida was roughly 2 million in 2008’s Republican primary; that’s more than double the votes cast in the first three contests combined. Indeed, as of Saturday, many more absentee/early votes have already been submitted in Florida (about 197,000) than votes cast in Iowa (about 121,500). And that number presumably will shortly eclipse New Hampshire’s total GOP primary vote (about 248,000). Florida is a mega-state battle in a war thus far dominated by small-state skirmishes.

After South Carolina, though, there are no guarantees for Romney anywhere; he will have to work hard to beat Gingrich in the Sunshine State. Romney is leading by nearly 19 percentage points there according to the RealClearPolitics average, but because of Gingrich’s Palmetto win that lead will almost certainly be cut dramatically this week. Remember: Gingrich led in that very same poll average in Florida just two weeks ago (see the graph at the link).

Going into February, Romney — on paper — would seem to have many advantages. About a quarter of caucus-goers in Nevada in 2008 were Mormon; Romney should do well there, and the other candidates, save Paul, might not even bother contesting it. The major contests looming at the end of the month are primaries in Arizona and Michigan. Arizona is not as conservative as one might think. As the chart shows, its percentage of evangelical and “very conservative” voters (according to 2008’s exit polls) was almost identical to Florida’s — somewhat surprising demographics from Barry Goldwater’s home state. Meanwhile, Romney has deep Michigan roots; his father was governor and he grew up there.

At the very least, there’s every reason to suspect that this contest will go at least into March’s Super Tuesday, and perhaps much longer. We also cannot eliminate the possibility, however faint, that this field is not yet complete, especially if Gingrich wins Florida. The Crystal Ball’s Rhodes Cook laid out the possibilities last month, and his analysis still holds. Establishment Republicans do not want Gingrich as their nominee, period. Privately, some senior GOP members of Congress have predicted that if Newt is the Republican nominee, the House will go Democratic in November.

Nonetheless, Gingrich will fight hard, using elite opposition to him as a battle cry that has resonance with rank-and-file party members still smarting over the establishment’s “imposition” of John McCain in 2008. That makes the White House happy, even though it is not obvious that a long campaign is automatically bad for Romney. After Mitt’s self-inflicted wounds over releasing his tax returns during last week’s debates and stumbling answers on some other topics, it’s clear that Romney needs plenty of training before any fall showdown with Barack Obama. Hillary Clinton made Obama a better candidate during the 2008 slog; Gingrich might do the same for Romney. Or Romney and Gingrich might destroy each other, with Paul and Santorum, if he remains in the race, inflicting additional wounds on the eventual nominee. Arguably, there are no more presidential tries left for any of the Final Four — it’s now or never. That means no one will hold back.

Tonight’s result reminds us that the past does not always predict the future: This is the first time since the South Carolina primary’s inception in 1980 that Iowa, New Hampshire and South Carolina have picked three different Republican winners: Santorum in Iowa, Romney in New Hampshire and now Gingrich in South Carolina. Streaks are meant to be broken, and in all likelihood South Carolina’s perfect record in picking Republican presidential nominees will be broken as well.

Moreover, South Carolina has often signaled the beginning of the end of the nominating process. Instead, by its shocker landslide for Newt Gingrich, South Carolina may be just the end of the beginning in a long, winding road to August’s Tampa convention.