Sanders’ frontrunner status is a symptom of the devolution of the Democratic Party. Collective action problems could produce Sanders 2020 just like they did Trump 2016.

Media Name: bernie_sanders_the_accidental_nominee.jpg

An elderly candidate with extreme policy views but an energized base of supporters won the presidential nomination of a major political party. But rather than being punished at the polls in the general election, they had the good fortune of facing an opponent from the other major political party who is almost as unpopular as they are. They went on to win the election with a divisive message that rallied partisans by stoking existential fears and piquing their distrust of the other political tribe.

Depending on your political priors, you could read that paragraph as either a description of Donald Trump’s presidential campaign in 2016 or, depending on how the next couple of months turn out, a retrospective on the Bernie Sanders campaign in 2020. Trump and Sanders are two very different individuals from opposite sides of the political spectrum, but they share something in common. In both cases, two presidential candidates who would have once been considered far too radical or otherwise unpalatable managed to win (or nearly win) the nomination of their respective parties.

They did so neither because of some native genius nor because of the mass popularity of their particular messages. Rather, their successes are consequences of the devolution of the modern political party system. Trump and Sanders are symptoms of party decline, not its cause.

As part one of this series explained, the modern party was created in the 1830s in response to the failure of the parties to control the candidate selection process. When a party is weak, it cannot function as a gatekeeper that prevents more than a handful of candidates from declaring their intent to seek higher office. As a result, a large field of candidates splits party support in too many directions, potentially allowing a more marginal candidate with a unified base of support to beat out the more broadly palatable options, as Trump did in 2016.

That scenario is known as a collective action problem. But in 1832 future President Martin Van Buren instituted the nominating convention, allowing the Democratic Party to quickly and efficiently select a single candidate to represent the entire party and to sidestep the collective action problems that had bedeviled the party in the 1820s. Van Buren’s system worked well for the next century and a half, but with the rise of the binding primary in the 1970s, brokered conventions could no longer function as a tool of control over the nomination process; most delegates have since been bound to cast their ballots for whichever candidates win a state’s primary vote. However, the major parties developed other tools that could functionally replace the brokered convention as gatekeeping mechanisms: access to the debate platform and to large dollar donor lists.

However, by the 2000s even these substitute controls had begun to break down, starting with effects of the rise of Fox News on the Republican Party. (See part two of this series for a more detailed description.) If Fox News wanted to host an additional primary debate (or two or five or ten), there was little the party could do to stop them given just how large the channel’s conservative viewership was. In addition, the GOP could no longer act as an effective intermediary between the donor class and candidates since television exposure enabled candidates to bypass the old donor list system.

As a result, the field of candidates exploded from an average of seven people per election cycle before 2000 to seventeen by 2016. This created new collective action problems within the Republican nomination process as broadly palatable candidates split the support of the largest internal factions within the party. In 2016 that allowed a marginal candidate like Donald Trump to win the nomination with barely 40% of the primary vote. Trump was not only the least popular major candidate in modern presidential history, he was also one of the least popular nominees within his own party. But political tribalism kicked in during the election season and Trump won on a wave of “at least he’s not her” mood affiliation.

I See Your Seventeen and Raise You Twenty‐​Eight

Something similar is happening during the 2020 Democratic primaries. The previous record of seventeen major candidates set by Republicans in 2016 has been smashed by the twenty‐​eight Democrats who declared their candidacies this cycle. The crowded field has allowed Bernie Sanders to jump out to an early lead with the support of between a quarter to a third of primary voters. His base might be relatively small, but his supporters are particularly vocal and energized, so much so that other candidates have complained about toxic online behavior from so‐​called “Bernie Bros.” Whether or not that’s a fair accusation, it is clear that Sanders’ supporters are significantly less likely than are other candidates’ supporters to say that they will support the eventual Democratic nominee if it is anyone other than Sanders himself.

Even though Sanders is the frontrunner, it is possible that none of the hopefuls will win a majority of party delegates. This leaves pundits slavering over the prospect of the first truly brokered nominating convention in a generation, along with all the drama and cheap headlines that would entail.

A brokered convention was never more than a slight mathematical possibility for the Republican Party in 2016 because too many of the primaries were winner‐​take‐​all contests, in which a plurality of the vote in a state–even as little as 35%–gave the victor the entirety of the state’s delegates. By contrast, the Democratic process favors proportional delegation, meaning that each of the candidates clearing a certain vote threshold (often 15%) receives at least some delegates from that state. This makes it more difficult for the frontrunner to clinch the nomination before the convention unless they win outright majorities in the primaries.

This delegate system was created by design, a reaction to the turmoil of the 1968 Democratic National Convention when unbound Democratic delegates chose Vice President Hubert Humphrey–who had entered no primary contests–over Eugene McCarthy, who had won the dozen or so non‐​binding primary contests. The party responded by shifting towards a more purely democratic process in hopes of injecting a sense of legitimacy back into the process, thus reassuring members that the nominee wasn’t purely a creature of backroom deals by party elites.

“Democratic” is a loaded term that carries many positive connotations, but modern parties eschew a purely democratic nomination process for good reason. Parties are designed to balance the democratic will of their rank‐​and‐​file voters against the party’s interest in winning the general election. The latter requires winning over voters who are not party members, or, if they are members, those who are not as hardcore as primary voters. A purely ‘democratic’ Democratic Party will nominate candidates with views that are too distant from those of independent and swing voters in the general election, decreasing the odds of electoral success. But if the party is too heavy‐​handed with its control of the nomination process, it could generate a legitimacy problem for the eventual nominee. It’s a tightrope walk and when parties lean too far one way, it is all too easy to overcorrect in the opposite direction.

After enacting reforms that democratized the primary and delegate system in 1972, the Democratic Party ran a series of nominees who were too radical for the general electorate, like George McGovern, Michael Dukakis, and Walter Mondale. In response, the Democratic National Committee added in a “superdelegate” system in which several hundred party elites are automatic delegates who are not bound to primary results. In theory, superdelegates shifted the balance back from pure democracy towards party control, but the superdelegate system has not yet been meaningfully stress‐​tested, though that might change if there is a brokered convention in 2020. But the Democratic Party had more meaningful tools for control of the nomination process than the mostly untested superdelegate system: access to big dollar donors and the debate stage.

Filthy Lucre

As to the first, parties have traditionally cultivated relationships with major donors, the kind of people who can drop hundreds of thousands or even millions of dollars on Congressional re‐​election funds or, more recently, to support dark money Political Action Committees. The party could then act as an intermediary between hopeful candidates and this donor class, bringing them together in a series of private forums, salons, and dinners known as the “invisible primary” (or as Ronald Reagan called it, the “mashed potato circuit”) which happen months before the election season formally starts.

Money is a necessary condition for a successful campaign. (Though Michael Bloomberg’s fizzling campaign is a reminder that money, while necessary, is insufficient.) Most campaigns end when fundraising dries up, as happened this cycle to Kamala Harris, Cory Booker, and Beto O’Rourke. Campaigns that are stingy with cash flow can sometimes survive longer and hope for a boost in the polls (and fundraising) from a surprise primary result or debate performance, as has been true of Amy Klobuchar, though that is rarely a successful path to the nomination.

But in general, campaigns have to spend money to make money. They must hope to enter into a virtuous spiral: they buy ads and hire staff in order to gain exposure for their candidate, so that they will poll well, so that the media will cover them, so that voters will support them, so that donors feel comfortable backing them, so that they can hire more staff, and so on.

As long as the candidates all depended on a few thousand big‐​ticket donors, the party as the intermediary had real power over the nomination process. A candidate adjudged to be too radical or controversial could find themselves struggling to get traction with donors whose ears were being filled with whispered sweet moderation by party leadership. No big money meant no shot at a serious, extended campaign.

A Barbaric Yawp

All of that changed in the early 2000s with the rise of crowdfunded political campaigns. A radical Senator from the small state of Vermont named Bernie Sa…err…Howard Dean knew that he would struggle to compete in the “invisible primary” contest for big‐​ticket donors, who would have preferred a known commodity with more centrist views hailing from a larger state, someone like Al Gore, John Kerry, or John Edwards. So Dean launched his exploratory committee very early in 2002, more than two years before the election. That allowed him to start fundraising, but, rather than relying solely on big‐​ticket donors, Dean turned to online donations, the first candidate to do so in earnest.

Bear in mind, this was still relatively early in the digitization of commerce; most people didn’t have Amazon Prime accounts and Apple Pay did not exist. Entering credit card information online had only just become an ordinary activity and doing so for a political campaign was still novel. But Dean had staked out a radical anti‐​war position against the Afghanistan occupation and Iraq War, which endeared him to a younger generation of left‐​wing activists. (Dean’s success was in many ways an echo of Eugene McCarthy’s 1968 “Childrens’ Crusade” in New Hampshire.)

Young voters were more comfortable with the internet in all its forms–Dean’s was the first campaign to regularly feature bloggers and use internet “townhalls”–and more willing to donate online, albeit in small amounts. But when it comes to purchasing power, it doesn’t matter whether one person donated $100,000 to your campaign or a thousand people donated $100. By the time he ended his campaign, Dean had raised $50 million, 40% of which came from online, small‐​dollar donations.

Dean didn’t win, thanks in part to his overly enthusiastic yell after the Iowa caucus and the viral memes that followed (a reminder that the Very Online giveth, the Very Online taketh, blessed be the name of the Very Online). But Dean’s approach to fundraising spawned many imitators. Barack Obama built up a network of 1.5 million small‐​dollar donors during his successful campaign in 2008, helping him shock the Democratic Party establishment by overtaking the establishment‐​favored Hillary Clinton, who had relied more heavily on big dollar donors.

However, the most significant foreshadowing of the future of political fundraising in the 2008 elections came from a rather surprising corner. Congressman Ron Paul, a long‐​time libertarian irritant to GOP leadership, broke records by raising $6 million in a single day from small‐​dollar, online contributors. Before online fundraising, someone as marginal as Paul would have been excluded from serious campaigning; GOP leadership would have just flexed its control over access to big dollar donors. Someone with Paul’s outsider views would have had to be independently wealthy to run a major campaign effort, someone like billionaire Ross Perot, who self‐​financed a quixotic third‐​party campaign in 1992.

Online fundraising had the effect of bringing radical outsiders interested in running for President back into the major party fold. Party leadership could no longer keep them off the debate stage or underfunded. The rise of online payment technology had battered down the fundraising gates, leaving the former gatekeepers looking on helplessly. Ron Paul, who ran as the Libertarian Party nominee in 1988, could instead ride a wave of small‐​dollar donations into the Republican primaries in 2008 and 2012. Likewise, in 2016 and 2020 Bernie Sanders could run for the Democratic Party nomination despite a long career representing various minor socialist parties. There are few historical precedents for this new era of partisan permeability.

Feeling Blue?

By 2016 the online fundraising model had morphed once again, moving away from each candidate having with their own donation platform and donor lists that had to be replicated afresh each cycle. Instead, a wave of tech startups were created to bridge from one election cycle to the next and from one candidate to another. (They are essentially Paypal‐​like payment systems with a special focus on regulatory compliance with the Federal Election Commission.)

The largest of these platforms was ActBlue, created in 2004 with the express purpose of supporting progressive Democratic candidates. ActBlue grew steadily over the next several election cycles, but it exploded in 2016 when it surpassed $1 billion in cumulative funds raised. But most of ActBlue’s fundraising haul did not go to the eventual Democratic nominee that year, Hillary Clinton. Twelve years after Howard Dean, another Vermont Senator with radical beliefs, one Bernie Sanders, ran the most successful online fundraising strategy to that point. Of the $228 million he raised during the 2016 cycle, as much as 80% of it came from ActBlue, a truly staggering amount for a candidate who would not have had a chance in an earlier era of party control.

From a radical perspective, Sanders’ ability to delay the anointing of Hillary Clinton is, of course, a good thing. They do not much care about the case of the vapors they are giving party leadership. Democratic socialists like Bernie Sanders are more likely to be conditional Democrats who prioritize ideological purity and the long‐​term, upside potential for systemic revolution over short‐​term, partisan victory in any given election. (Libertarian readers should be able to sympathize, given that a significant number of right‐​leaning libertarians are equally conditional in their support for Republican candidates.)

But from the perspective of the Democratic Party qua the party, Bernie’s small‐​dollar fueled primary campaign was a disaster. Admittedly, you could point to any one of a dozen reasons why Clinton lost in the general election in 2016, from her decision to not campaign in Wisconsin, to FBI Director James Comey’s October surprise, to her odd choice to not bother with evangelical voter outreach; when you lose by a margin of just 80,000 votes in three states, almost any bad decision qualifies as fatal. But those were generally her own mistakes or those of her campaign, not failures of the Democratic Party structure itself.

Remember that the goal of any modern party over the last century and a half has been to select a single candidate as quickly and efficiently as possible. But because the Democratic Party was devolving from a modern party into something more primitive and less structured, it could no longer as effectively perform that core function. It could no longer control the nominating convention or access to fundraising. So Sanders could raise the tens and hundreds of millions of dollars necessary for a deep campaign run and there was little the party could do to stop him.

This forced Clinton to spend time and money on her primary campaign instead of gearing up for the general election contest. After all, while the structural fundamentals of the 2016 political landscape and economy favored the GOP, the Democratic Party should have had at least some compensating retail advantages. Clinton was a member of the outgoing administration; in an earlier era she would have had the nomination wrapped up by February at the latest and then spent the next several months looking presidential and raising money while the chaos of the Republican primaries left their eventual nominee bruised and battered. Instead, Clinton was battling tooth and nail with Sanders until June, months after Trump had secured the GOP nomination.

To Debate or Not to Debate? That is the Question

However, Democratic party leadership did have one remaining tool of control over the nomination process in 2016 and they wielded it energetically. Money is the lifeblood of campaigns, but its purpose is to buy exposure for the candidates through advertisements, staff for door knocking, and other campaign outreach efforts. But money can’t buy access to the debate stage. (Michael Bloomberg’s $400 million golden ticket in 2020 is the exception that proves the rule.) On the primary debate stage, candidates can make their pitch directly to potential voters and donors.

Parties have traditionally decided how many debates to hold. Debates are high risk, high reward events; exceed expectations; the winner can potentially get a significant boost in the polls and fundraising totals. This is especially valuable if you are a lower‐​tier candidate. But the frontrunner, who already has widespread name recognition and donor support, has less incentive to accept the downside risk of a bad debate performance. Thus, having fewer debates conveys an advantage to leading candidates, while having more debates favors challengers, those scrappy upstarts from small states or with radical views that place them outside of the party mainstream. For challengers, each debate is a raffle ticket; the more you have, the better your chance of pulling off a surprise upset.

The Democratic Party leadership in 2016 wasn’t interested in playing the raffle. Everyone not living under a stone since the 1990s had heard of Hillary Clinton, so a lack of exposure was not her problem. And big‐​dollar donors were lining up to give to the assumed nominee. Primary debates were all downside risk with minimal upside for the Clinton campaign. Sanders, on the other hand, could benefit from his willingness to boldly stake out radical positions on the debate stage without worrying about the effects on swing voters in the general election.

The Democratic Party leadership, many of whom owed political favors to the Clintons, went to great lengths to rig the process in Hillary’s favor. Officials discussed planting questions at townhall debates that would highlight Sanders’ extreme views and they leaked questions in advance to the Clinton campaign. In a reminder that the truth is stranger than fiction, the DNC’s manipulations were exposed by Russian hackers with an assist from the whistleblowing anarchists at Wikileaks.

Although the leaked emails were embarrassing for the DNC, the primary mechanism of party control was much more mundane. The DNC could help the Clinton campaign and hinder Sanders by holding fewer debates and scheduling them for time slots that would attract few viewers. Six debates were initially sanctioned, fewer than was normal for a contested election cycle. And the timing was suspect with most of the debates falling on weekends, when they had to compete head‐​to‐​head with football games, or in proximity to holidays. Other campaigns immediately accused the DNC of playing favorites with the debate schedule, but even the best laid plans of mice and party officials go oft awry. After Sanders beat Clinton in a surprise New Hampshire primary upset, the Clinton campaign changed its tune and asked for more debates.

In any case, suspicions of skullduggery by the DNC over the debate schedule fed into a broader narrative about how the party establishment was behaving undemocratically. This led some alienated Sanders supporters to stay home from the polls in the general election or even to vote for Republican nominee Donald Trump instead of Hillary Clinton.

Learning the Wrong Lessons

The belief that the Democratic Party was not behaving democratically in 2016 has had ripple effects for the 2020 election. Modern parties always have to balance their control of the process in order to maximize general electability with the need to convey party‐​unifying legitimacy on the eventual nominee. But when the leaked emails in 2016 let the public see how the party sausage got made, it forced the DNC to correct course for 2020, especially given that the former target of party machinations four years prior was now the frontrunner with a massive grassroots network of ardent supporters.

This discouraged the DNC from using access to the debate platform to control the nomination process. The number of sanctioned debates was doubled to twelve. And while there was a threshold that candidates had to clear to make it into the debates, it was set so low that twenty (!) candidates qualified, forcing the party to split the debate into two panels of ten. The resulting cacophony of voices satisfied nobody; the event turned out to be less a debate than a series of disconnected and overly‐​rehearsed bits.

Maximizing the number of debate participants and holding more total debates allowed the party to avoid the appearance of rigging the system, but it came at a severe cost. The sheer number of candidates on the debate stage made it hard for the top tier candidates to consolidate their respective lanes.

Take, for example, the contest over the moderate lane, which was divided between at least half a dozen top tier candidates (and perhaps another half dozen from the lower tier). The moderate frontrunner in the polls was former Vice President Joe Biden, but he spent the greater part of the first several debates standing behind a podium and listening to a dozen other moderate hopefuls attack him. Indeed, the “winner” of the early debates was often defined by whichever moderate challenger was most effective at blasting Biden’s old policies and gaffes.

Adding to the confusion, billionaire former New York City Mayor Michael Bloomberg bought his way into debate qualification with $400 million in ad spending, immediately siphoning off support from other moderate candidates. Four moderate candidates have either won double‐​digits in one of the early state contests or, in Bloomberg’s case, in the polls. Combined, they arguably represent more than half of Democratic voters, but their lane is hopelessly divided.

At this relatively late stage in the primary process, when the lanes are typically consolidated behind two or three final hopefuls–as was true for Democrats in 2016–there are still eight candidates in the race. This kind of packed field benefits whichever ideological lane has the fewest contestants. In 2016, the beneficiary was Donald Trump, who could watch on in amusement as the other Republican candidates jostled over control of the moderate and Christian Right lanes. In 2020, Sanders has been the beneficiary. He is the only Democratic Socialist in the running and although there is a good amount of policy overlap between him and Elizabeth Warren, she is the only major candidate contesting his control of the left‐​wing of the Democratic Party. (And she has been relatively gentle in her attacks on Sanders during the most recent debates, leading analysts to speculate that she is angling for a Vice Presidential nod.)

Bernie 2020

As appears increasingly likely given that a third of all delegates are up for grabs on Super Tuesday in early March, the inability of the moderates to coalesce behind one candidate could allow Sanders to gain an insuperable lead among delegates in the next week despite winning a mere plurality of Democratic voters. That would pose a lose‐​lose situation for party leadership going into the convention. If they argue that a plurality of delegates is insufficient for the nomination and hold a true brokered convention, which then leads to another candidate taking the nomination on a later ballot, they will face accusations of anti‐​democratic meddling, potentially leading to defections among Sanders supporters in the general election. But if they simply anoint Sanders, they could face similar complaints from other campaigns of denying the revealed preference of a majority of voters for other candidates.

If that happens, remember that either outcome was foreseeable and perhaps even avoidable with the right combination of bold leadership and vision. The Democratic National Committee should have looked at what happened to the Republican Party in 2016 as a cautionary tale about the dangers of party devolution. Instead, by thinking only of their own mishandling of the 2016 Democratic primaries, they created the perfect preconditions for collective action problems.

Given that election prediction models currently favor the GOP in November–although an economic downturn in the intervening months could change those forecasts–Democrats will need every little edge in retail politics they can find. Instead, because of both the transformation of political fundraising and the self‐​inflicted failures of party leadership, their eventual nominee, whoever it may be, will limp into the general election after a prolonged and bruising primary process.

The Democratic Party is devolving for different reasons than the Republican Party, but the end effect has been similar. Candidates who would have once been considered too radical can no longer be prevented from running by party leadership. As gatekeepers have lost influence, the fields of candidates and the number of primary debates have swollen. The subsequent collective action problems favor candidates backed by small but dedicated followings, allowing them to gain victory with mere plurality support. And then, because presidential elections primarily hinge on factors other than retail politics, those previously unthinkable candidates end up sitting in the Oval Office, from where they pursue radical policies that contribute to a growing climate of institutional distrust, party polarization, and the decline of American democracy.

Next up is the fourth and final entry in the “Death of the Modern Party” series; what will the future of American politics look like and what, if anything, can be done to address the devolution of the modern party system?