海角大神

Lights, camera, faction! The fun part of Campaign 2016 is about to begin. That鈥檚 debate season, of course 鈥� that glorious early period when every candidate still has hope, Iowa and New Hampshire haven鈥檛 winnowed the obvious also-rans, and the gaffes flow as freely as lobbyist cash.

Henceforth the presidential hopefuls will receive yet more intense news coverage. Reporters will deconstruct every speech, question every ad, and turn each policy paper upside down and thump it on the bottom to see if bits of nonsense fall out. Stories will burrow into candidates鈥� biographies, their hairstyles, their donors, and their promises. Lots of coverage will compare and contrast: 鈥淒oes Trump鈥檚 bluster raise stakes for Christie?鈥� 鈥淲ill Sanders pull Clinton to the left?鈥� 鈥淜asich and Pataki: Here鈥檚 how to tell them apart.鈥�

It鈥檚 the clash of multiple personalities that creates much of the entertainment value of this stage of the race. Then the contenders in the field will dwindle, slowly at first and finally all in a rush, until only the party standard-bearers are left. The race will be reframed as a two-person conflict: a boxing match, a 100-yard dash, or Grand Slam tennis final. Finally a single individual will stand at the summit, having grasped, with his or her wit, toughness, and sagacity, the most powerful political position on the planet, the job that used to be called Leader of the Free World.

Unless that鈥檚 not how it really works. On the cusp of the 15-month cavalcade of bunting and banter that will be the presidential campaign, let鈥檚 stop and consider this question: Does it really matter who the parties choose to be their presidential candidates?

Maybe the contenders aren鈥檛 as important to the final outcome as the media portray them to be. Maybe they鈥檙e striding around speechifying while riding strong unseen forces that will largely determine the outcome in 2016.

That鈥檚 the way many political scientists look at the race. It鈥檚 not that they see candidates as ants on a floating stick who think they鈥檙e steering. Candidates do make some difference. To say otherwise would be to deny the obvious.

But the personalities involved probably have less to do with the election results than most voters believe, particularly when it comes to the big prize, the presidential general election.聽

Take the accepted analogy for much campaign coverage: the horse race. It implies a fast-moving contest that鈥檚 changing all the time because of candidate effort. Obama is ahead! Here comes Romney on the right flank! Oh, Romney鈥檚 tripped on a comment about the 鈥�47 percent鈥�; now it looks like he鈥檚 slipping behind. Too bad!

That鈥檚 not an accurate depiction of the dynamics of a presidential contest, according to some who study politics for a living. It鈥檚 much more stable than the horse race comparison implies. The better analogy might be a tug of war, in which two generally strong teams, with equal amounts of experience and cash, strain and strain to little apparent movement. The daily twitches of campaign politics, the gaffes and speeches and attack ads the media love so much, don鈥檛 really change where the candidates鈥� feet are planted.

鈥淚f, for some reason, one side let go 鈥� meaning they stopped campaigning 鈥� the other side would soon benefit. But of course the candidates do not let go and that makes it hard to see that their efforts are making a difference,鈥� write political scientists John Sides of George Washington University and Lynn Vavreck of the University of California at Los Angeles in 鈥淭he Gamble,鈥� their study of the 2012 election.

鈥� 聽 聽 鈥� 聽 聽 鈥⒙�

Is this just another story about how the state of the economy determines who wins the White House?

No. Well, maybe a little. Look, the fundamental state of the nation matters a lot when voters make their presidential choices. Perceived prosperity is a big part of that. History shows that it is very difficult to defeat the incumbent party when the economy is growing.聽

The important measure in this regard will be the change in US聽 gross domestic product from January to June 2016. If it鈥檚 going up, that will be good for the Democratic candidate, almost certainly Hillary Clinton. If it鈥檚 not, the GOP standard-bearer will benefit.

Throw in voter views as to whether the United States is becoming more or less entwined in foreign conflicts and you鈥檒l have a pretty good basis on which to predict the 2016 outcome, according to many political scientists.

鈥淭he fundamentals are most of it,鈥� says Scott Lemieux, a professor of political science at The College of Saint Rose in Albany, N.Y., and frequent contributor to the political and cultural blog 鈥淟awyers, Guns & Money.鈥�

But 鈥渕ost of it鈥� isn鈥檛 the same as 鈥渁ll.鈥� There鈥檚 still some room for other things to influence who gets to choose the Oval Office furnishings. Recent presidential elections have been close enough for even marginal effects to make a difference. Just ask President Al Gore.

And there are other underlying factors that stabilize and influence presidential elections besides the economy and war or peace. There鈥檚 the nature of the electorate, for one thing. Simply put, it鈥檚 more predictable than its constituent voters may think. Consider, for instance, these axioms, to the extent that there are axioms in politics:

鈥淚ndependent鈥� voters often aren鈥檛. Heard the one about the growing number of US voters who say they鈥檙e independent, beholden to no party? While there鈥檚 some truth behind that, generally speaking it does not mean there is a large pool of persuadable voters waiting to be wooed by one candidate or the other.

Let鈥檚 split up the electorate into chunks so you can see what this means. About one-third of US voters are Democrats and one-third are Republicans, points out Jennifer N. Victor, an assistant professor of political science at George Mason University in Fairfax, Va. One-third identify as independents.

But many of those self-described 鈥渋ndependents鈥� are independent in name only. About one-third of them are secret Republicans, and one-third secret Democrats. Push them and you鈥檒l find that they actually vote like declared partisans.

The remaining one-third of the one-third is indeed composed of independents. 鈥淭hese independents amount to 10 to 12 percent of the voting population, which is much smaller than most people think,鈥� writes Ms. Victor in a post on the 鈥淢ischiefs of Faction鈥� political blog.

Wait, there鈥檚 more. About half of the actual independents don鈥檛 make it to the polls. That鈥檚 right 鈥� they don鈥檛 vote, according to research conducted by other political scientists. The portion that does vote generally splits its ballots between Democrats and Republicans.聽

聽And these final true swing voters tend to be the least-informed adults in the US electorate. That makes sense if you think about it 鈥� they鈥檙e the folks who don鈥檛 follow political news and thus haven鈥檛 formed opinions about candidates and issues they don鈥檛 know exist.

Parties unite. Here鈥檚 a corollary to the decline in true independents: Party identification is a big factor 鈥� probably the biggest 鈥� in determining how people vote.聽

You鈥檒l see lots of stories in coming months about Democrats and/or Republicans being internally divided because of one contretemps or another. Don鈥檛 believe them. Or, rather, don鈥檛 take them too seriously. In a narrow sense, it鈥檚 true that Donald Trump鈥檚 views on immigration are dividing the GOP and Vermont Sen. Bernie Sanders is more appealing to liberals than Mrs. Clinton. It鈥檚 just that this stuff, in the end, will have little effect on how partisans vote.

The fractious primary period quickly fades away. After parties nominate their standard-bearer, their supporters almost uniformly rally to the cause. Absent a Ross Perot-level third-party candidate, around 90 percent of Republicans, including Republican-leaning independents, will cast ballots for the GOP candidate. Ninety percent of Democrats will vote for their nominee, write Mr. Sides and Ms. Vavreck in 鈥淭he Gamble.鈥�

鈥淚n presidential elections, partisanship reigns, and predictably so,鈥� they note.

This process isn鈥檛 necessarily automatic. Presidential campaigns can strengthen voters鈥� predispositions toward one party or the other. But the point here is that tough primary battles don鈥檛 appear to disillusion partisans and cause them to switch parties or stay home on Election Day. Take the 2008 election, when Clinton and Barack Obama slugged it out for months. In the November general election, Mr. Obama actually did better in states where the Democratic primary was particularly competitive.

Gaffes don鈥檛 matter much. The way the press depicts it, campaigns and candidates are likely to implode at any moment. The long, tiring march to November provides plenty of opportunities for gaffes 鈥� poorly phrased assertions, secret videos, and other mistakes that can damage electoral prospects in an instant. The weary contenders have to guard against these missteps as best they can.

Here鈥檚 the real story: Few of these blunders actually move voters. Those that do change minds often produce only temporary effects.

鈥淒o gaffes matter? It鈥檚 not clear there鈥檚 systematic evidence for that,鈥� says Julia Azari, an assistant professor of political science at Marquette University in Milwaukee and frequent 鈥淢ischiefs of Faction鈥� contributor.

To see this effect, or lack thereof, let鈥檚 look at 2012鈥檚 best-known gaffe 鈥� the 鈥�47 percent鈥� tape. That was the secretly recorded video of Mitt Romney telling a roomful of wealthy supporters that 47 percent of the US population believe they鈥檙e entitled to government aid and that it wasn鈥檛 his (Mr. Romney鈥檚) job to try and make these spongers take personal responsibility for their lives.

Asked directly what they thought of this comment, many voters said they didn鈥檛 like it. A Gallup poll taken shortly after the tape was made public by Mother Jones found that 36 percent of respondents said the remark would make them 鈥渓ess likely鈥� to vote for Romney.聽

But few people actually changed their allegiance. Gallup鈥檚 horse race polls of voter preference stayed fairly constant in the wake of the tape鈥檚 release, varying a point or two one way or the other. President Obama maintained a slim lead.聽

Averages of major polls showed the same generally stable trend line. Sides and Vavreck found that Romney鈥檚 numbers did deteriorate in detailed polling undertaken by YouGov
聽.com. But the change there was small, too, with a 49 percent to 44 percent Obama lead moving to 49 percent to 41 percent. Romney lost three percentage points 鈥� and those voters moved to 鈥渦ndecided,鈥� not 鈥淥bama.鈥� That implied they were wavering Romney backers who could be won back to the fold. (See 鈥淧arties unite,鈥� above.)

And gaffes aren鈥檛 the only game changers that often aren鈥檛. Campaign ads have only a minimal effect, for instance. At the presidential level they can move opinion a bit in markets where they鈥檙e aired, but only for a short period of time. Then their effects wear off, particularly if there鈥檚 an ad counterattack from the other side.聽

The same is true of debates. Even if one candidate obviously bests the other (think Romney鈥檚 adroit performance in the first 2012 face-off) the poll-o-meter just jiggles. Then the next debate comes up and the victor finds it hard to repeat the performance. The polls jiggle the other way.聽

Party conventions are usually nonevents as well. It鈥檚 been decades since the major parties actually picked a nominee at their conventions. Now the meetings are nothing but tightly controlled, weeklong commercials. Yes, they can result in significant poll gains, but ... OK, maybe this is where the no-effect theory breaks down. The national conventions can be true turning points in White House campaigns. If successful, they can result in significant poll gains that can affect the relative standing of candidates in November.

Perhaps the importance of the conventions is an exception that proves the rule: Getting through to US voters requires a massive effort that walls off a large portion of a major American city while filling hotel rooms for three states and attracting tens of thousands of journalists, each of whom will expect free food and a souvenir tote bag.聽

This does not mean that voters don鈥檛 care. It doesn鈥檛 imply that the masses are too busy watching every cat Vine video to consider national issues. It just means that voter choices are stable. These choices are heavily influenced by partisan affinity and the state of the nation writ large. People are busy; the campaign is just a background murmur until the election is only weeks away.

鈥淰oters are heavily influenced by the world in which they are living 鈥� more so than any campaign stunt,鈥� wrote Hans Noel, an assistant professor at Georgetown University in Washington, in an influential 2010 journal article, 鈥淭en Things Political Scientists Know that You Don鈥檛.鈥�

聽鈥� 聽 聽 鈥� 聽 聽 鈥�

In this context, what difference can a candidate 鈥� an individual, one person 鈥� make in a contest largely driven by fundamental forces?

We鈥檙e talking about the presidency here, remember. At the state level and below, candidates can matter a lot. That鈥檚 because in races for lesser offices, candidates have more contact with voters. It鈥檚 also more likely for political opponents and/or their campaigns to be unequally matched.聽

Remember the Delaware special Senate race in 2010? Republican tea party activist Christine O鈥橠onnell managed to upset Republican Rep. Mike Castle, a former governor, in a low-turnout primary. Statewide, Mr. Castle was one of the state鈥檚 most popular politicians, and might have easily beaten Democratic nominee Chris Coons. Ms. O鈥橠onnell was a much weaker choice, in part because she had dabbled in the occult when young. (She infamously began a campaign ad with the phrase 鈥淚鈥檓 not a witch.鈥�) Mr. Coons won and then won reelection to a full term in 2014. To this day, national Republicans grumble that they could have easily picked up the seat in 2010 with a decent contender.

But that kind of a mismatch is much less likely to occur at the national level. The lengthy presidential nomination process can ruthlessly expose a candidate鈥檚 flaws. Hacking your way through the primary jungle requires considerable political skill.

鈥淕enerally it produces a competent mainstream candidate,鈥� says Mr. Lemieux of The College of Saint Rose.

But 鈥済enerally鈥� isn鈥檛 the same as 鈥渁lways.鈥� Ideology is a crucial dimension of presidential candidates 鈥� if voters think candidates are too far out of the mainstream, candidates can suffer at the polls. In 1964, GOP nominee Barry Goldwater鈥檚 conservatism seemed out of step with much of his party as well as the public, and he suffered a historic defeat at the hands of Lyndon Johnson. In 1972, Democratic nominee George McGovern correspondingly seemed too liberal to many voters. Also, his campaign was a mess. He got crushed by Richard Nixon.

鈥淎 lot of political scientists would agree that ideology makes a difference,鈥� says Ms. Azari of Marquette University.

Beyond that the relative importance of the person on the podium isn鈥檛 a settled issue in academia. Some political scientists 鈥� we鈥檒l call them 鈥渢he fundamentalists鈥� 鈥� think candidate personalities matter little and that the economy and other fundamentals are everything. Others 鈥� perhaps labeled 鈥渂ehaviorists鈥� 鈥� think there鈥檚 evidence that certain candidate qualities can appeal to some voters and even change their minds.

Consider the basic question of likability. Favorability ratings are a staple of political journalism 鈥� who鈥檚 liked, who鈥檚 not liked, who鈥檚 so disliked they鈥檙e Donald Trump. Commentators routinely use favorability as an apparently empirical means of rating the appeal of candidate personalities.

But favorability polls have a chicken-and-egg problem. Do voters develop a favorable view of candidates, and then decide to vote for them? Or is it the other way around? Maybe voters decide to back candidates because of party affinity or economic conditions, and then (惫辞颈濒脿!) suddenly decide that, darn it, that candidate seems like a nice person after all. Maybe we鈥檒l do brunch after this election thing is over.

Romney鈥檚 2012 adventures might illustrate the limits of punditry based on a candidate鈥檚 personality. Commentators routinely noted Romney鈥檚 empathy gap, the fact that more voters thought Obama 鈥渃ared about people like them.鈥� Few noted that Republican presidential candidates almost always trail on this measure, because of the historic association of the party with business and wealthy individuals, and that some of those GOP standard-bearers won the White House anyway.聽

In the campaign鈥檚 final days, some media outlets breathlessly tracked Romney鈥檚 favorability rating, noting that he was catching up to Obama on this measure. Some polls had the pair even on likability in the end. But subsequent analyses showed that Romney鈥檚 gains weren鈥檛 a reflection of (largely nonexistent) swing voters moving to his side. They came from wavering or undecided Republicans 鈥� voters who were highly likely to support him anyway.聽

This leads to a broader point about elections, according to Sides and Vavreck: Winning and losing don鈥檛 necessarily depend on who the candidates are as people, or what the electorate thinks about their personal strengths and weaknesses.

鈥淧erceptions of the candidates seem to be much more the consequence, and not the cause, of how people vote,鈥� they write.

鈥� 聽 聽 鈥� 聽 聽 鈥⒙�

Here鈥檚 a corollary to the assertion that candidates aren鈥檛 all-important: You can鈥檛 fix Washington by electing the right person to a position of national power.

No one leader can bust up Washington鈥檚 partisan stagnation. No savior exists who through sheer force of personality will persuade Congress to stop arguing and start passing important, difficult stuff.

And maybe that鈥檚 not so bad. Maybe it鈥檚 actually OK. Maybe it鈥檚 preferable to the alternative.

The US government, with its clashing institutional powers and multiple centers of leadership, was designed to foster disagreement. It was designed to slow things down and make progress incremental. The Founding Fathers were afraid of a monarchy. They wanted democracy and not dictatorship. They wanted stable policies and not lurching change.

鈥淥ur political system is not one that operates under a cult of personality,鈥� says Victor of George Mason University.

Yes, that鈥檚 frustrating when big problems need to be addressed. Yes, it鈥檚 particularly frustrating now, when divisions seem so deep and political disagreement so ugly.

But conflict is what free politics is about. Disagreement is a natural byproduct of democracy, says Victor, and maybe we should learn to value that more.