When Donald Trump walked on to the presidential debate stage on August 6, he walked into a trap.
Bret Baier began with a question designed to make sure every one of its viewers knew that Trump, and Trump alone, was refusing to rule out a campaign as an independent. He asked broadly if "anyone on stage" with that view would raise his hand, and once Trump did so, Baier hammered his heresy home: "Mr. Trump to be clear, you’re standing on a Republican primary debate stage. The place where the RNC will give the nominee the nod. And that experts say an independent run would almost certainly hand the race over to Democrats and likely another Clinton."
Minutes later, Megyn Kelly informed the audience that Trump was a sexist. "You’ve called women you don’t like 'fat pigs,' 'dogs,' 'slobs,' and 'disgusting animals,'" she said. "Does that sound to you like the temperament of a man we should elect as president?"
Then, Chris Wallace suggested that Trump was full of it on his marquee issue — pointing out that Trump had "repeatedly said" he had evidence that the Mexican government is intentionally sending criminals across the border, but "refused or declined" to share it. "Why not use this first Republican presidential debate to share your proof with the American people?" Wallace asked.
The punishment continued. The moderators hammered Trump for his past support for single-payer health care, his past donations to Democrats, and the bankruptcies of his companies. After the debate was over, commentators on the network kept slamming him, with GOP pollster Frank Luntz trotting out a focus group who said that Trump's performance had been disastrous.
All this looked like a live demonstration of one of the most celebrated recent theories in political science — "the party decides" theory of presidential nominations.
The idea, put forth in a 2008 book of that name by political scientists Marty Cohen, David Karol, Hans Noel, and John Zaller, is that modern presidential primaries aren't driven by the whims of voters. Instead, the most important players are party insiders, who use a variety of tools to try and deliver the nomination to the candidate they prefer — and usually get their way. Indeed, since Trump's rise shocked the political world, many commentators have repeated that phrase "the party decides" like a mantra, assuring themselves that it points to his inevitable downfall.
Now, Fox News, an extremely important institutional actor in the modern Republican Party, was trying to take the bombastic billionaire down, using the debate to discredit him in front of a massive viewing audience. It mirrored a similar pile-on from party officials after Trump said John McCain wasn't a war hero. And afterward, when Trump complained that moderator Megyn Kelly had "blood coming out of her eyes" and "her wherever," there was yet another round of condemnations.
But the voters couldn't care less.
A month and a half later, Trump remains the leader in primary polls. Among Republicans nationally, in Iowa and in New Hampshire, his support didn't decline in the weeks after the debate, but actually improved — many results put him at or above 30 percent in the 16-candidate field. Indeed, the main beneficiaries of the first two debates so far have been two other party outsiders: retired neurosurgeon Ben Carson and businesswoman Carly Fiorina. Meanwhile, Sen. Ted Cruz, despised by his party's establishment, is outperforming most of the career politicians in the field, and independent socialist Bernie Sanders is showing surprising strength on the Democratic side.
If there's one thing wonkish political pundits think they know about presidential primaries, it's that party insiders will doom these candidates' chances. It is "the party elites who traditionally decide nominations," the Upshot's Nate Cohn writes. "Everything we know about presidential nominations screams that Trump has no chance of winning," political scientist Jonathan Bernstein argues at Bloomberg View. I myself have made the case that Trump's weakness among party elites will lead to his defeat. We've all believed that the party will somehow, in some way, end up stopping these outsiders.
But the idea that the GOP's elites, rather than its voters, control the process seems absurd to many of the party's consultants. "Today’s primary dynamic is driven by a dissatisfaction and palpable anger with the current state of the party," says Republican consultant Mary Matalin, who's worked for the Bush family and Vice President Cheney. Steve Schmidt, who ran John McCain's presidential campaign in 2008, goes further. "The party establishment has no ability to stop Trump," he says. "The voters are like the wild animals that escaped the Georgia zoo."
When Trump, Carson, Fiorina, and Cruz supporters in polls are combined, Schmidt points out, they now regularly make up well over 50 percent of the Republican electorate. This is true both nationally and in key early states like Iowa and New Hampshire. More than half of GOP voters, Schmidt says, are "signaling complete and utter and absolute contempt for the party establishment."
The political consultants have a point. The Party Decides — which, of course, was written as a work of scholarship, not as ammunition for pundits — has many caveats, and the book's claims aren't as strong as its titular catchphrase might suggest. Additionally, the theory fits several recent cycles only messily, and this GOP contest already looks quite different than most of the cases described in the book. Trump certainly has many weaknesses, and it's inarguably true that outsider presidential candidates usually don't end up their party's nominees. But there's no need to write them off entirely — or to assume that it's the power of the party that will bring them down.
So far, the GOP establishment appears paralyzed and uncertain, with Republican elected officials issuing endorsements at their slowest pace in decades. As of mid-September, the party hadn't decided to throw its weight behind anyone. But is this because elites just haven't made up their minds yet — or because voters are refusing to let them?
The Party Decides theory, explained
The Party Decides was the perfect theory to make sense of the bizarre-looking Republican nomination contest of 2012. Mitt Romney began the cycle as the frontrunner, but at various points before and during the primaries, Sarah Palin, Mike Huckabee, Rudy Giuliani, Newt Gingrich, Chris Christie, Michele Bachmann, Rick Perry, Herman Cain, Rick Santorum, and even Donald Trump — yeah, him — led one poll or another.
And yet, when it came down to it, Republicans did what everyone expected them to do. They nominated Romney, the guy who had come in second in 2008.
This is just what Cohen, Karol, Noel, and Zaller expected. Romney was the only one picking up significant endorsements from party figures. He was the one the party wanted, so he was highly likely to win. And he won. Score one for the political scientists.
The authors had started their research years earlier, back in the early 2000s. At that time, the modern presidential nomination system, which replaced backroom convention dealings with a series of primaries and caucuses across the country, was nearly three decades old. Yet serious scholarship about how it functioned was surprisingly rare. Major works written in the early days of the new system, in which nebulous concepts like "momentum" were proposed to explain who ended up winning, didn't seem to hold up. Many agreed, though, that the new system had moved power away from insiders to actual voters.
Yet the 2000 election, in which Al Gore and George W. Bush locked down the vast bulk of party support months in advance and won their nominations, suggested that elites had actually learned how to dominate the new system behind the scenes. Perhaps, the authors thought, the "invisible primary" taking place among these party elites was what truly mattered most.
To test their theory, Cohen, Karol, Noel, and Zaller tallied up endorsements from a broad set of party figures across two and a half decades of primaries. They included everyone from famous elected officials to local pols to activists to celebrities, and weighted each endorser's importance in the party. If one candidate was the clear winner in pre-Iowa endorsements, and won the nomination, then it could be said that the party had decided.1
The authors define the "party coalition" broadly. In their theory, it doesn't just include "the establishment" or even official party members, but also ideological activists and outsider factions. In practice, however, there naturally tend to be more insiders than outsiders in their data.
And that's just what they found. In eight of ten competitive presidential primary contests between 1980 and 2004, endorsements showed that party insiders clearly backed one candidate before Iowa, and that candidate won the nomination. Endorsements were better at predicting the outcome than polls, fundraising numbers, or media coverage.
The authors don't argue that endorsements alone specifically cause a candidate to win. Rather, endorsements are a signifier of how the invisible primary is going — and therefore of which candidate the party network is choosing to favor. "In our theory, party insiders rally to the candidate of their choice, endowing him or her with endorsements, access to fund-raising networks, and pools of talent and volunteer labor," they write.
The finding was impressive, and the theory was counterintuitive — or, as counterintuitive as a theory granting great power to the establishment could be. Forget those roller coaster polls, eye-popping fundraising numbers, and dramatic horse race coverage, it suggested. Forget charisma, authenticity, and "connecting with voters." Forget Trump! The smart people know that the real winner will be determined by the party in the invisible primary. That's how nominations really work.
But is it?
The party decides... sometimes, usually when their choice is obvious, and they don't always get their way
When you look more closely at the endorsement data, and at what's happened over the past few decades, the story looks more complex, and less certain. First, when the party "decides," it's usually a decision to fall behind someone obvious. Second, the candidate that wins endorsements can end up losing. And third, in wide-open races, the party usually fails to decide before the voting begins.
Here's a look at all 16 of the competitive contests in the new system — including those just before and after the period the book focuses on. In just ten of those, the party favorite — defined as the clear endorsement leader before the Iowa caucuses, if one exists — won.
It turns out that in most of the cases in which the party does decide, there was one candidate who clearly had a higher standing than the others from the start. That, The Party Decides co-author and University of Maryland professor David Karol told me, is his main "skepticism" about his own theory. "In the cases where the party has coordinated on a candidate," he says, "it's almost always been somebody fairly obvious."
Indeed, of the ten correctly predicted contests in the table above, eight featured either an incumbent president, a well-respected sitting or recent vice president, or a strong runner-up from the party's most recent contest. In those cases, the party's "decision" actually looks pretty easy. But The Party Agrees When the Answer is Obvious isn't a particularly interesting book title.
The two other correctly predicted cases are the victories of Texas governor George W. Bush in 2000 — who, of course, had many clear advantages and was the front-runner all along by every metric — and Arkansas governor Bill Clinton in 1992, who was little-known among the national electorate. "Bill Clinton is the best case for us," says Karol, because "the party got behind him late in 1991" — before the voters did.
But overall, the book's findings could be interpreted as making the more prosaic point that when there's a very clear heir apparent, that person both gets endorsed by the party and tends to end up winning. That would have some implications for this year's Democratic race, in which Hillary Clinton has long been the clear leader in endorsements. (Bernie Sanders may have recently taken the lead in some Iowa and New Hampshire polls, but he hasn't gotten even one Democratic congressman, senator, or governor to back him.)
Yet even when the party does decide on a candidate, that person doesn't always triumph. Clinton and John Edwards split endorsements before the 2008 Iowa caucuses, according to the authors' metric, but a candidate far behind them — Barack Obama — won. And Edmund Muskie, the overwhelming establishment favorite in the 1972 Democratic race, was toppled by antiwar liberal George McGovern, who was considered a complete outsider by party elites. The Party Decides authors view this outcome as a fluke of a poorly-understood new system, and argue that party officials have since become much better at preventing such a turn of events. However, they also write that elite favorite Walter Mondale's "bare" victory over endorsement-poor Gary Hart in the 1984 Democratic race shows that it's certainly possible that "the insider favorite" can "be beaten."2
The authors separate out the 1972 and 1976 nomination contests — which resulted in the shocking nominations of George McGovern and Jimmy Carter for Democrats, and the more-expected but close nomination of incumbent President Gerald Ford for Republicans — in their analysis. They argue that, back then, the parties had not yet learned to operate in the new system, and made various basic strategic errors that are now unlikely to be repeated. That's reasonable, but it's hard to ascribe the complete and shocking collapse of consensus party frontrunner Ed Muskie to simple bad strategy.
This year's Republican race, meanwhile, presents a different problem — the party isn't uniting around anyone at all yet. Jeb Bush hoped to replicate his brother's 1999 success at locking down party support early, but it hasn't happened. Among higher-level officials, just three senators, a handful of congressmen (mostly from Florida), and zero governors have backed him. That's enough to give him a narrow lead in FiveThirtyEight's endorsement tracker, but the numbers are so low that that hardly matters. The real story of GOP endorsements so far is that there have been remarkably few of them.
The Iowa caucuses are still over four months away, so there's still time for Republicans to fall in line and endorse Bush, Marco Rubio, or someone else. ("The book is not about who's leading in August," Karol says.) But, in the few recent races that have lacked an heir apparent, it's common for the bulk of party insiders to stay undecided for quite a while, and for those who do endorse to split their support among several contenders. This happened to Democrats in 1976, 1988, and 2004, and to Republicans in 2008. In all of these, the party failed to unite before the voters had their say. It's Bill Clinton's late surge in the 1992 contest, which benefited from a late series of party endorsements before Iowa, that's the fluke.
And in recent years, it's looked increasingly normal for the party to be indecisive for a long time. Neither John Kerry in 2004, nor Barack Obama in 2008, nor John McCain in 2008 was a clear endorsement winner before Iowa. In all these contests, party elites followed voters — not the other way around. That means that Mitt Romney's 2012 victory is the only race since 2000 in which the authors' data shows that a party made an early decision and got its way. "Since we started writing the book," Karol says, "we've had messy races."3
Kerry and Obama were both third place in endorsements in their respective nomination battles. For Republicans in 2008, McCain was technically in first, but Giuliani and Romney were so close behind him (35 percent to 32 to 27) that the authors view the result as a failure of the party to decide. The authors never calculated 2012 data, but the alternative endorsement counts out there all put Mitt Romney in first.
In these recent contests, party elites haven't looked at all like puppet masters — instead, they're frequently unhappy with the choices available to them, and frantically scrambling about how to respond to what voters seem to want, as expressed in polls and early state results. The GOP's elites, especially, have looked similarly hapless this year.
Even the 2012 Republican contest, seemingly the perfect example of the theory in action, can be interpreted rather differently. The man who emerged as Romney's main rival, Rick Santorum, was an ex-senator who had lost his most recent election by 20 points, was popular only among the religious right, raised hardly any money, and is now barely making a dent in the polls in his repeat bid this year. Yet this candidate, flawed and weak by practically every traditional metric of political strength, managed to win 11 states over the party favorite.
Viewed that way, these recent contests start to look less like exceptions — and more like a trend. Could the parties have lost their power to decide — if, indeed, they ever had it at all?
In case of insurgent candidate, please break glass
The nominations of John Kerry, John McCain, or Barack Obama don't truly fit the theory, but they don't exactly shatter it, either. Those candidates didn't clearly win pre-Iowa endorsements, and McCain in particular had serious enemies in the GOP. But they were all at least elected officials in their parties, national figures of some renown, and potentially strong general election contenders. The party may not have hand-picked them, but once voters started leaning toward them, it had no need to go all out to stop their nominations from happening.
The triumph of Donald Trump — or Ben Carson, Carly Fiorina, Ted Cruz, or Bernie Sanders — would be an entirely different kettle of fish. These are true outsiders. Trump, Carson, and Fiorina have never held elected office. Cruz is despised by his party's elites — as, of course, is Trump. Sanders, meanwhile, has never been a member of the Democratic Party, and has embraced the "socialist" label that mainstream Democrats try to avoid at all costs. We haven't seen anyone remotely like them be nominated since Carter and McGovern, who totally lacked ties to their party's national elites.
Indeed, in every contest starting with 1980, both Democrats and Republicans have ended up with a nominee that party insiders can live with. No nominee since has been a fringe figure. So some have argued that even if the parties can't always handpick the candidates they want, they have been able to exert their influence to block candidates they view as truly unacceptable. That is, even if the party can't decide, it can veto.
Yet while it's definitely true that parties have many tools they can use to influence the nomination process, it's not clear how strong and reliable those tools are — or how decisive they've been in recent contests.
The key question here is what mechanisms the parties use to decide. They can, for instance, deploy various procedural tricks — such as influencing the primary calendar, changing ballot qualification requirements, and limiting the number of debates. But these changes can have unintended consequences, and heavy-handed interventions aimed directly at blocking a popular outsider candidate would likely lead to a voter backlash. The only formal move against Trump the GOP has felt comfortable making so far was its successful effort to get him to sign a pledge not to run as an independent.
Then there's money. Traditionally, many presidential candidates who don't get party elites' approval have been either starved of financial resources and forced to quit the race, or hammered by negative ads. But nowadays, there are more and more alternative sources of cash available for insurgents — such as Super PACs that can be funded by just a few idiosyncratic billionaires, and small-donor internet fundraising aimed at enthusiastic true believers. And, of course, a billionaire like Trump can self-finance and may not be responsive to these pressures.
Expertise is another factor. It's difficult to mobilize voters, gather ballot qualification signatures, and master a complex series of rules governing delegate selection in 50 states and several territories. The people who know the system and how to work it best are often closely tied to the party. But motivated outsiders, like George McGovern's 1972 campaign team, can also master the rules and see possibilities the insiders miss. Plus, money can buy expertise — the chief operative behind Rick Santorum's shocking 2012 Iowa caucus win, Chuck Laudner, is now working for Trump.
Though all these are important, in the end, the authors of The Party Decides write that "political persuasion — a cue from partisan leaders to partisan voters — may be the most important mechanism of insider control." In other words, the party gets its way by telling voters, both through implicit signaling and explicit endorsements, which candidate is best.
"Most primary voters are partisan voters who care about their party and they like the party leadership," Noel, a professor of government at Georgetown (and, as of earlier this month, a Vox.com contributor at Mischiefs of Faction), says. "So if the party coordinates and says, 'that’s our guy,' then those voters are going to respond to that."
But this year's Republican voters haven't seemed in the mood to follow their elites' lead. Indeed, polls have shown that, in contrast to Democrats, a majority of the GOP electorate thinks its party leadership isn't doing a good job representing their views on major issues. Voters are frustrated and angry, Matalin says, because of the party's "seeming inability to deliver conservative policies despite monumental electoral gains at all levels."
Meanwhile, the judgments from party elites that have been expressed so far almost seem to act as an anti-endorsement. Condemnations from those elites, like those aimed at Trump, have been taken as a badge of honor. When the billionaire calls GOP leaders "stupid," as he so frequently does, voters tend to agree with him. If that's the case, why would they care what the party's eventual decision may be?
So far, it looks like the voters — not the parties — are calling the shots
When The Party Decides was written, it offered valuable pushback against the conventional wisdom that parties had lost all their influence on the nomination process. And its focus on endorsements is a helpful alternative to early polls that have frequently been wrong. But in the time since, "the party decides" has become the new conventional wisdom among some wonky pundits — despite the small number of modern contests and the many messy exceptions, especially in recent years.
Party elites' influence matters and should be taken seriously. But it's time to start treating the theory with a bit more skepticism. It doesn't explain, or purport to explain, how every contest will turn out. "The key contribution is not so much that we can tell you who's going to win, but rather that we should think of this process in a way differently than most people do — that we should look at it from the point of view of the party," says Noel.
Furthermore, the idea that we should use findings mainly based on what tends to happen in contests with clear heirs apparent to try to explain this wide-open GOP race seems strange. We could end up with an outcome like the others, of course. But so far, it looks like the years when the party fails to make a decision early — the type of contest the book's endorsement statistics do not explain particularly well.
When I asked Karol whether anything could have changed to reduce the party's grip in recent years, he cited two main possibilities. First, party elites' control could be weakened if there's a particularly controversial hot-button issue splitting them from their voters. "When we first presented The Party Decides, [UCSD professor] Sam Popkin said, 'I basically buy your story — except when the there's a really divisive issue and the party elite can't really get behind somebody," Karol says. For instance, the Vietnam War empowered the rise of George McGovern in 1972, and the Iraq war was a major issue in the Democratic primaries of 2004 and 2008.4
The first of those contests is frequently cited as a good case for The Party Decides, because Kerry, a candidate acceptable to the establishment, ended up winning over antiwar outsider Howard Dean. But the party didn't anoint Kerry — he was actually third place in endorsements, behind both Dean and Dick Gephardt. The voters surged to Kerry in Iowa before the party did. "The party didn't stop Dean, they couldn't have," says Joe Trippi, who worked on Dean's campaign.
So perhaps unauthorized immigration, the issue that first catapulted Trump to the front of the polls, is now having a similarly divisive effect on the GOP. It's long been a troublesome topic for the party's elites, who tend to either explicitly or tacitly support a path to legal status for these immigrants, while their base favors a much harder line. John McCain managed to survive his heresy on the issue back in 2008, but this year, Trump's tough talk about deporting 11 million people seems to excite conservatives most. And now that Scott Walker has quit the race, both of the remaining candidates who seem formidable to elites — Jeb Bush and Marco Rubio — are very closely tied to immigration policies that deeply anger those conservatives.
Second, Karol said, elites' control could be weakened due to changes in the media. The 24-hour news cycle of intensive coverage, the rise of partisan media like Fox News, and the internet have made recent presidential campaigns feel quite different. "Fox, especially, has changed things" on the GOP side, Karol says. "How did Republicans even know who Ben Carson was? Fox News has been building these guys up, and making them celebrities." Karol doesn't think these changes really explain the rise of Trump, but suggests they may have "made it harder for traditional party elites to steer the process."
Steve Schmidt thinks this has happened. "Parties once had the power to restrain the impulses or the sovereignty of the individual voter," he says. "They could communicate in a hierarchical, top-down way, and it was more often than not received as credible." But nowadays, Schmidt continues, "people trust people like them, and consume news through social media, and through websites and blogs and Twitter and all manner of other sites, where their opinions are validated, not challenged." Which means that it's easier for fans of Trump or other insurgents to find messengers that will tell them what they want to hear, and tune out messages from party-approved organs that tell them to be more reasonable.
There are many reasons to believe that, in the end, voters will choose to be more reasonable. It's still quite early, and much of the public hasn't been closely following the race so far (though those two debates did get record ratings). Some evidence suggests that, as the election gets closer, voters start caring more about electability. Early polls haven't proven predictive in recent years, and many recent early state winners have emerged quite late. The field has already winnowed and will winnow further.
There are also many reasons to believe that Trump and the other outsider candidates will fail. Presidential politics is a difficult business, and many novices who start out looking quite good at it soon stumble. Other newcomers to politics might have skeletons in their closets. Trump, of course, has a volatile personality and could self-destruct, and the second debate exposed his ignorance on many issues. But if Trump brings himself down, or GOP voters simply decide they'd rather not support him, it doesn't necessarily follow that it was the party that was responsible.
But the lingering question remains — what if the anti-establishment feelings that have boosted outsiders in the polls aren't just a passing fad, and the wild animals truly have escaped onto the streets? If the voter infatuation with Trump lasts into early state balloting — or is transferred to another outsider candidate with a somewhat different set of skills, like Ted Cruz— it's not at all clear that the party would be able to stop him. The same holds true for Bernie Sanders on the Democratic side. Elites would certainly try to kill these candidacies, and if the voting results turn out very close, there's room for insider shenanigans to tip an outcome.
Yet despite beliefs that the establishment will surely come up with "something" to stop these insurgents if they catch on, I don't really see how that would work. Democratic legitimacy matters nowadays. If someone like Trump, Cruz, Fiorina, or Sanders wins the primaries fair and square (emerging from the process with a clear lead in pledged delegates), it's very difficult to imagine how the party could get away with deposing that person — his or her supporters, numerous by now after voting across the country, would be enraged. Not only would such a divisive move hurt base turnout in the general election, but the supporters of those outsider candidates could later vote in party primaries, like the one that famously deposed House Majority Leader Eric Cantor.
At the end of the day, parties do need their voters. Elites may or may not end up falling behind one particular candidate, but if they do, it's still the voters who will take a look at that person and decide whether he or she is worth backing. So the party better hope that it can still persuade — or it may find it quite difficult to get the animals back in their cages.