Skip Navigation

Rationalizing the Presidential Nomination Process

Our democracy will benefit if components from debates to primaries to conventions are given much-needed updates.

October 28, 2021

Long before the first early vote was cast in the November 2020 general election, the lengthy campaign season offered its own challenges to fairness and democracy. None of them were as dramatic or injurious to democratic norms as the post-election assault on the legitimacy of Joe Biden’s election. But this does not mean that the rest of the 2020 campaign should fade away without serious analysis and discussion of potential avenues for reform.

Little in American political life is as ad hoc as the way that we nominate presidential candidates. Over the last six decades, the transformation in how we choose presidents has been based on happenstance as much as rational design. There is no good reason, for example, why the campaign calendar should begin with the unrepresentative Iowa caucuses, other than that Jimmy Carter put them on the political map in 1976 with his surprise win after intense campaigning.

As a result, and to the detriment of our democracy, many aspects of the system reflect the law of unintended consequences. Southern Democrats created the first Super Tuesday in 1984 by grouping together their states’ primaries in an effort to encourage the nomination of a moderate like Ohio Sen. John Glenn. Instead, the Democrats nominated liberal Walter Mondale. But the front-loading of the primary calendar has become a quadrennial tradition as states — in a desperate effort for relevance — cluster their primaries on the same dizzying Tuesday in early March.

I have been covering national politics for four decades and 11 presidential races. Over the years, I have become concerned that the system under which we nominate presidential candidates does not enhance democracy or voter deliberation. What follows are nonpartisan thoughts about plausible reforms for 2024 and 2028, regardless of who is running for president in either party. These ideas are presented with modesty since many of the problems are knotty. And even though the topic is beyond the purview of this essay, my cherished profession of political journalism is culpable as well amid the current crisis of democracy.

This essay is an update of a report that I wrote for the Brennan Center in 2017, “The Chosen One: Thoughts on a Better, Fairer, and Smarter Way to Pick Presidential Nominees.” The 2020 campaign and further reflections on the turbulence of the Trump era have prompted me to modify some of the ideas that I advanced in that prior report. But my goal — then and now — is to bring as much democracy and deliberation as possible to the way that we nominate and elect presidents.

How we got here

Theodore H. White's book, "The Making of the President 1960"

This year marks the 60th anniversary of the publication of what remains the most influential book in history about how America nominates and elects its presidents. The Making of the President 1960, with its riveting backstage account of the election of JFK, won the Pulitzer Prize and sat atop the best-seller lists for a year. Theodore White’s enduring classic can serve as a marker for how dramatically presidential campaigns and elections have changed since the mid-20th century.

In 1960, despite all the attention lavished on Kennedy’s victory in the nonbinding West Virginia primary, almost all convention delegates in both parties were selected in caucuses dominated by political insiders. There were almost no intraparty debates, although Kennedy and Lyndon Johnson made a low-key joint appearance before the Massachusetts and Texas delegations at the 1960 convention. Most convention delegates had only limited free will, deferring for the most part to the wishes of their state’s governor or dominant political boss.

The modern presidential primary system emerged from the tear gas–laden chaos of the 1968 Democratic Convention in Chicago. As Elaine Kamarck, a senior fellow at the Brookings Institution, points out in her invaluable book Primary Politics: Everything You Need to Know About How America Nominates Its Presidential Candidates, one-quarter of the delegates to the 1968 convention had been selected in 1967, long before Eugene McCarthy and Robert Kennedy launched their antiwar candidacies. The obvious unfairness of this system, which helped assure the nomination of Hubert Humphrey, who never entered a single 1968 primary, created irresistible pressures for reform. By 1980, 33 states were selecting their convention delegates in binding primaries as power dramatically shifted from political bosses to partisan voters.

The year 1980 was also a time of transition in terms of debates playing a central role in the fight for the nomination. Despite the aura surrounding the four Kennedy–Richard Nixon faceoffs in 1960, primary debates had been a haphazard and occasional affair. Jimmy Carter debated only twice before he was nominated in 1976, and Ronald Reagan never got to share a debate stage with Gerald Ford that year. All this changed in 1980 when Reagan and George H. W. Bush debated six times on the road to the GOP convention.

Image of TV studio where John F. Kennedy and Richard Nixon engage in a 1960 presidential debate Bettmann/Getty
One of the four 1960 presidential debates between Democratic nominee John F. Kennedy and Republican nominee Richard Nixon.

The structure and sponsorship of these primary debates have evolved over the years from candidate-funded events (Reagan meant it literally in 1980 when he shouted, “I paid for this microphone”) to today’s overhyped cable news ratings bonanzas with their countdown clocks, sometimes raucous audiences, and game show–like production.

At the same time, the culmination of the nomination race — political conventions — are becoming as much of an artifact of the past as cigar smoke–filled back rooms. Over the past 60 years, conventions have gone from decision-making bodies to four-day pep rallies. Not since the 1980s has there been a serious battle on the convention floor over anything, including party platforms. In 2008, at the end of the Democratic primaries, Hillary Clinton was about 200 delegates behind Barack Obama. In prior decades, Clinton would have taken her fight to the convention floor as Ted Kennedy did in 1980, even though he had a much larger delegate deficit. Instead, reflecting the new realities of primary voter sovereignty, ahead of the convention Clinton permitted her delegates to support Obama. By 2016 Donald Trump was anointed as the Republican nominee in early May, when he had won only about 40 percent of the delegates.

The problem with the current system for choosing presidents is that the individual components fail to fit together. The primary campaign period has become virtually endless, with candidate announcements often coming nearly two years before the presidential election. But then, because of the clustering of the primaries, a de facto nominee can be chosen in a month, as was the case with Biden in 2020. The initial hopes of reformers that debates would elevate the tenor of presidential politics have long ago been drowned in a sea of practiced sound bites and evanescent gotcha moments. And almost no one could have imagined that pre-nomination debates would become a profit center for cable TV networks.

In this era of the permanent campaign, Republicans with White House ambitions have already moved on to 2024 regardless of whether Donald Trump runs again, with would-be GOP nominees making exploratory treks to Iowa and New Hampshire. Democrats, in contrast, are frozen by the assumption that Joe Biden or Kamala Harris will be their 2024 nomineebut early indications suggest that the party may jettison the Iowa caucuses.

Despite this initial skirmishing on the Republican side, nothing about the 2024 race is set in stone. It will be months before the political parties and the states establish the order of the primaries. And it will be years before any arrangements for the 2024 debates are made or the convention sites are chosen. In short, everything about the underlying structure of Campaign 2024 is still in flux, and there is time for significant improvements across the board.

A spate of debates

Presidential politics has been afflicted with an inflation problem — a glut of not-quite-serious White House contenders. In 2016 the Republicans boasted a near-record field of 17 candidates, including a reality TV star named Donald Trump. The Democrats easily topped that in 2020 with two dozen. While the definition of “serious” is to some extent in the hands of the voters, it is hard to argue that spiritual adviser Marianne Williamson was a plausible 2020 presidential nominee, yet she was on stage for two Democratic debates. Entrepreneur Andrew Yang — who later admitted that he had only run in 2020 to promote the idea of a guaranteed annual income — was a regular participant in Democrat debates until the New Hampshire primary, in which he received 2.8 percent of the vote.

Before 2016 there were no practical obstacles to each party’s presidential candidates appearing on the same stage, although strategic considerations often played a major role in the scheduling of debates. The only exception was the Democrats’ elephantine 17-candidate scrum in 1976, but there were only three presidential debates that year, all late in the primaries after the field had been narrowed by the voters.

All that changed in 2015. Confronted with an unprecedentedly large field for the first GOP presidential debate that August, Republican Party Chairman Reince Priebus made a misguided decision. The Republicans allocated 10 slots on the main debate stage to the candidates who polled the best in national surveys. Six months before the 2016 Iowa caucuses, with many voters barely paying attention, national polls ranking the candidates precisely were about as accurate as a blunderbuss. Pollster Lee Miringoff, who oversees the respected Marist Poll, actually delayed a national survey to prevent it from being used in the Republican rankings. As he pointed out in a passionate online posting, “Name recognition unduly influences results of early primary horserace polls. Lesser known candidates will now frontload their efforts to try to make the cutoff. Public polls altering campaign strategies? BAD!”

The Republicans in 2015 provided a consolation prize to presidential candidates who didn’t make the polling threshold — a so-called “kids’ table” debate with other lagging contenders. These undercard debates consigned an influential senator (Lindsey Graham) and the runners-up in the 2008 and 2012 GOP nomination battles (Mike Huckabee and Rick Santorum) to perpetual limbo.

2016 Republican presidential candidates, from left: George Pataki, Mike Huckabee, Rick Santorum, and Lindsey Graham engage in a primary debate Justin Sullivan/Getty
2016 Republican presidential candidates George Pataki, Mike Huckabee, Rick Santorum, and Lindsey Graham participate in the last primary debate of 2015.

Failing to anticipate that Donald Trump would prove a major profit center for cable TV news, the Republicans ceded almost all control of the actual primary debates to the networks that broadcast them. That meant that everything revolved around TV ratings rather than informing the voters. Candidates were grouped on stage based on their poll positions (which meant that Trump was invariably center stage), the audiences were permitted to be raucous, and debate answers were so short (usually one-minute responses and 30-second follow-ups) that they resembled haiku. The result was debates that would have undermined serious discussion of the issues even without the disruptive presence of Trump on stage. 

In theory, the Democrats in 2020 should have learned from the 2016 Republican demolition-derby debates. Instead, the Democrats, led by Chairman Tom Perez, found new and different ways to compound the GOP’s mistakes. The Democrats retained polling thresholds and added another wrinkle by requiring candidates to have a certain number of individual donors. The initial requirements for the first debates in Miami in late June 2019 were modest — either 1 percent support in the polls or 65,000 individual donors. To their credit, the Democrats eliminated the kids’ table and began the campaign season with two 10-person debates on back-to-back nights, with the candidates grouped by lot rather than by poll numbers.

But the Democrats soon abandoned the notion of fairness for a format designed to create more dramatic television. The September 12, 2019, debate in Houston required participants to both score 2 percent in national polls and have a minimum of 130,000 donors. Desperate candidates, worried about making the threshold, actually spent as much as $60 in online advertising to attract each $1 donor. Rather than encouraging grassroots fundraising, this arbitrary requirement perverted campaign strategy and spending decisions. The toughened criteria for the September debate played a major role in prompting New York Sen. Kirsten Gillibrand, Washington Gov. Jay Inslee, and former Colorado governor John Hickenlooper to drop out of the race in August. Among the active candidates permanently exiled from the debate stage beginning in Houston were Colorado Sen. Michael Bennet and Montana Gov. Steve Bullock. And before 2019 was over, New Jersey Sen. Cory Booker and Julián Castro, a former Obama administration cabinet secretary, were also barred from debates by failing to meet ever-rising polling and donor thresholds.

Until recently, the voters in early states such as Iowa and New Hampshire winnowed the field of presidential candidates. But the arbitrary debate rules in both parties of late have meant that senators and governors whose campaigns fail to catch fire immediately are either forced to drop out prematurely or rendered semi-invisible by their absence from the televised extravaganzas.

Looking ahead to the 2024 and 2028 nomination fights, there must be a better way to organize primary debates. Voters need help in sorting out large fields of candidates who, because they are in the same party, often tend to agree on most policy issues. Primary debates, for all their inherent flaws, remain the easiest way for voters to develop nuanced views of the candidates.

Solutions: A new way to debate

As a starting point for the future, the political parties should realize that they do not have to cede control to the networks in order to induce them to broadcast the debates. With intense viewer interest in politics, TV networks presumably would continue to vie to host the debates even if the political parties demanded that they tone down the atmosphere and permit longer answers from the candidates.

The political parties could experiment with hosting debates themselves and letting all networks freely cover them as news events. That would allow some debates to be limited to one or two topics rather than bounce from issue to issue in pogo-stick fashion. Away from the control of the networks, occasional debates might also feature questioning from policy experts or academics like political scientists and historians. The problem with journalistic panels is that too often the questions are designed to create short-term controversy rather than anything substantive. And while reporters are sometimes adept at forcing candidates out of their comfort zones, town meeting-style debates with voters quizzing the contenders usually produce prefabricated answers lifted from stump speeches.

There also needs to be a better system to separate out candidates who are seriously running for the White House from those who are trying to boost their speaking fees, promote a single cause, or are off on inexplicable ego trips. As both parties discovered in 2016 and 2020, this can be a daunting task. One approach would be to guarantee a spot on the debate stage to anyone who has won a statewide election in the prior decade. Candidates who do not fit into this category (and this would have included both Trump in 2016 and former mayor Pete Buttigieg last year, among others) could be required to submit a certain number of online petition signatures or meet a similar condition to demonstrate support.

A better notion, though, would be for each party to designate a blue-ribbon group (maybe a diverse set of former elected officials and retired state chairs) to vet the candidates at the beginning of the race for seriousness and plausibility. Yes, this sounds heavy-handed and exclusionary. But before the 1970s, political parties did this in every campaign year, quietly shunning candidates who were, say, secret alcoholics or otherwise known to be unreliable. These days, although the media tries, no one is vetting the candidates, as the nomination of Trump in 2016 demonstrated. Such a blue-ribbon group — especially if it were committed to fairness and a reasonable amount of inclusion — would restore a small role to the political parties in choosing the presidential candidate who would run under their banners.

Who’s on first?

Nothing about the early 2020 Democratic caucuses and primaries followed a predictable pattern — aside from Iowa again being the state that couldn’t count straight.

For the second time in eight years, the overhyped Iowa caucuses failed to deliver a timely verdict. In fact, the Iowa Democratic Party was unable to release any returns on caucus night, February 3, because of the meltdown of its cell phone app. Pete Buttigieg was belatedly declared the winner three days later in a seemingly hair’s-breadth victory over Vermont Sen. Bernie Sanders. But those results were in dispute all through February — and the Associated Press, to the end, refused to crown a winner because of “concerns about whether the results as reported by the party are fully accurate.”

The supposed logic for beginning the nomination fight in four small states (Iowa, New Hampshire, Nevada, and South Carolina) is that this jerry-built system allows voters to winnow the field. The traditional refrain, repeated by generations of political reporters, has been “There are only three tickets out of Iowa.”

But in 2020, all of the five leading Democratic contenders — Biden, Sanders, Buttigieg, Sen. Elizabeth Warren of Massachusetts, and Minnesota Sen. Amy Klobuchar — stayed in the race until after the South Carolina primary. So did the two self-funding, late-entry candidates, former New York City mayor Mike Bloomberg and hedge fund billionaire Tom Steyer.

No candidate in modern history has ever followed anything like Biden’s trajectory to the nomination. When the 2020 Iowa caucus choices were finally tabulated, Biden ran an embarrassing fourth, with initial support from only 15 percent of Iowa Democrats. In 2004, after an analogous fourth-place Iowa finish, Rep. Dick Gephardt of Missouri dropped out of the race. Four years later, after limping home close to the bottom of the pack in Iowa, a veteran senator named Biden abandoned his White House dreams as he told supporters, “I feel no regret.”

But 2020 was a different story. Biden moved on to New Hampshire, where he finished a woeful fifth. The former vice president then stumbled into Nevada, where Sanders beat him by roughly a two-to-one margin in the caucuses. But then Biden’s landslide victory in the South Carolina primary with nearly 50 percent of the vote propelled him to the nomination and the presidency.

Joe Biden delivers remarks at South Carolina primary night election event JIM WATSON/AFP via Getty
Joe Biden delivers remarks at a primary night election event in 2020 in Columbia, South Carolina.

Solutions: Kill the caucus

So, after that roller-coaster ride, what are the lessons for the future from the early 2020 delegate contests?

The ineptitude of the Iowa Democrats destroyed the last shreds of a justification for selecting delegates in a caucus rather than a primary. Political parties — which run caucuses independent of state election officials — simply do not have the skills to reliably count ballots in a contested race. Iowa Republicans demonstrated this failing in 2012 when they initially declared former Massachusetts governor Mitt Romney (and ultimate nominee) the winner before reversing field two weeks later and awarding the crown to former Pennsylvania senator Rick Santorum. And back in the 1988 caucuses, Iowa Democrats deferred to network projections and stopped counting with roughly 750 precincts untallied.

More important, caucuses are inequitable because they are invariably low-turnout events. Traditionally, Iowans in both parties had to meet in person on a Monday evening in the depth of winter to participate in the caucuses. While the rules were always looser in Nevada, the caucuses did require showing up in person. In 2020 the Democrats, to their credit, tried to make it easier for voters in Iowa and Nevada to caucus. Iowa held virtual online caucuses in some group settings for Democrats who worked nights, temporarily resided out of state, or lived in group homes for the elderly. Nevada made provisions for early voting. But neither of these attempted reforms did much to increase turnout.

In 2020 about 70 percent of the New Hampshire voters who would ultimately back Biden in November cast ballots in the Democratic primary. The South Carolina primary attracted roughly half of the eventual general-election Biden voters. In contrast, despite the national attention lavished on Iowa, only about 175,000 Democrats participated in the opening-gun caucuses, which was less than one-fourth of the votes the Democratic ticket would receive from Iowa voters nine months later. The numbers from the Nevada caucuses were worse: just a paltry one-seventh of the state’s voters who opted for Biden in November bothered to caucus.

Caucuses were already an endangered species for the Democrats in 2020, with only two other states (North Dakota and Wyoming) holding them. But even if the holdout states switched to primaries, there would still be the question of which states get to go first. Over the years, both parties have protected a small group of early states — most recently Iowa, New Hampshire, Nevada, and South Carolina — from encroachment by other states jumping the gun. In 2008 Michigan and Florida violated Democratic Party rules by moving their primaries into the protected zone for early states. As a result, no major candidate campaigned in either outlaw primary — and the Democratic Party’s rules committee took away half their delegates.

From time to time, proposals have been floated to replace the individual state delegate contests with a national primary or a series of regional primaries. The most potent argument against these reformist notions is the example of the 2020 presidential campaign of Mike Bloomberg, who squandered a staggering $1 billion in his quest for the Democratic nomination. Tom Steyer spent more than $300 million on his own presidential ambitions. A national primary, or even a sequence of regional ones, would require something close to nine-digit spending on TV advertising for a little-known candidate to compete. Under this kind of system, there would be almost no way for a long-shot candidate (aside from self-funders) to break through based on personal campaigning and affordable TV ads in one or two early states.

True, no contender has replicated Jimmy Carter’s 1976 feat in going from “Jimmy who?” to the nomination and the White House. But centering early campaigning on a handful of small states has allowed underdog candidates like Arizona Sen. John McCain in 2000 and Bernie Sanders in 2016 to emerge as serious alternatives to the high-flying front-runners. And then there’s Rick Santorum. While his 2012 presidential race is mostly forgotten, he came surprisingly close to knocking off Mitt Romney for the GOP nomination.

There is an inherent logic to starting the presidential race with primaries (not caucuses) in four smaller states in different regions of the country. The necessity of personal campaigning undoubtedly helps the candidates understand the sprawling nation they are hoping to govern. And, frankly, the essence of democracy lies in a candidate like Joe Biden spending 90 minutes speaking to and patiently answering questions from 125 Iowa Democrats in a high school cafeteria in Knoxville (population: 7,595) just 14 months before he was inaugurated as president.

The conundrum, of course, is which four smaller states go first? Because by ordering the primaries, all states are equal, but some states are more equal than others.

The first two Democratic delegate contests in 2020 were rightly criticized for their lack of diversity; Iowa and New Hampshire are among the 10 states with a white population of more than 90 percent. In contrast, 56 percent of South Carolina Democratic primary voters were Black, according to exit polls. And despite their comparatively low turnout, the Nevada caucuses were reasonably diverse, with 17 percent of the participating Democrats Hispanic and another 11 percent Black, again according to exit polls. (Diversity is less of a consideration on the Republican side since the party’s base is disproportionately white.)

There is a strong case for retaining New Hampshire and South Carolina at the front of the pack.

In New Hampshire’s favor is a rate of political participation and primary turnout that is off the charts and a highly educated, politically independent population that is emblematic of the Northeast. Making South Carolina the second primary would enhance diversity, especially since candidates traditionally campaign by hopscotching back and forth between the first two states on the calendar.

Prior to the Trump era in the GOP, New Hampshire and South Carolina together captured a broad swath of Republican sentiment. New Hampshire boasted a quirky independent streak that embraced outsider candidates ranging from Pat Buchanan to John McCain. South Carolina, in contrast, was the state where the GOP establishment regrouped, which is why its primary was pivotal in securing the nomination for Bob Dole in 1996 and George W. Bush in 2000.

There are also real-world considerations buttressing the special roles of New Hampshire and South Carolina. An idiosyncratic state law gives the New Hampshire secretary of state the unilateral power to move the primary to a date as early as necessary (even into the odd-numbered year) “to protect the tradition of the . . . first-in-the-nation presidential primary.” With Biden in the White House and former South Carolina Senate candidate Jaime Harrison now chairing the Democratic National Committee, it is almost impossible to envision another southern state replacing South Carolina on the primary calendar.

The uninspiring track records of the Iowa and Nevada caucuses offer a compelling argument to award the third and fourth positions on the calendar to new states. As a substitute for Iowa, Kansas offers similar rural demographics and an agricultural pedigree. Kansas, in fact, aside from its overwhelmingly white population, is surprisingly close to providing a cross section of the nation in terms of median age, income, and education. Another appealing notion might be to replace Iowa with Wisconsin, a good proxy for the industrial Midwest. With no party registration in Wisconsin, an early presidential primary would allow both parties to test the appeal of its candidates to independent voters in one of the great 21st-century battleground states.

Nevada received its prime place on the political calendar largely because of former Senate majority leader Harry Reid’s clout within the Democratic Party. But while Nevada has one of the highest percentages of Hispanic voters in the country, the dominance of Las Vegas and casinos makes the state atypical. As a regional alternative, award the final spot on the early calendar to Arizona (a swing state like Nevada), Colorado, or New Mexico.

But even if the order of the 2024 primaries were New Hampshire, South Carolina, Wisconsin, and Arizona, it would not solve all the problems with the primary calendar.

The chaos of clustering

Minutes after the South Carolina primary polls closed at 7 p.m. on February 29, 2020, the TV networks declared Joe Biden the overwhelming winner. The next morning on NBC’s Meet the Press, Chuck Todd began the broadcast by declaring, “Joe Biden gets the landslide victory he needed in South Carolina.” By the end of the day, Pete Buttigieg — the Iowa winner who finished a close second in New Hampshire — had dropped out of the race, telling supporters that it was “the right thing to do when we looked at the math.” The next day, Amy Klobuchar ended her campaign by endorsing Biden at a Dallas rally where Buttigieg also backed the former vice president.

The race for the Democratic nomination had been upended in 48 hours in one of the biggest turnabouts in modern political history. And then on March 3 — aka Super Tuesday — 14 states, including California and Texas, held presidential primaries. A staggering 15.7 million votes were cast on a single day, as Biden won 10 of 14 contests. Even though they had already dropped out, Buttigieg and Klobuchar received almost 800,000 votes combined, with most of them presumably coming from absentee and early balloting.

Deliberation should be an integral part of democracy. Primaries, in particular, should offer voters time to reflect on their choices, since the differences among candidates in the same party tend to be nuanced. But the clustering of 14 primaries on the same day was a stampede rather than an opportunity for considered judgment. (Let me stress that this is a commentary on the process rather than the outcome.)

The glut of primaries on March 3 was inevitable once the Democrats followed tradition and issued a rule reserving February 2020 for the four designated early states. As a result, state legislatures gave way to the irresistible impulse to make sure that their states mattered in nominating a president by holding primaries on the first permissible Tuesday. In 10 of the 14 Super Tuesday states, primaries for other offices were held at a later date.

Joe Biden smiles at a crowd from a podium, while Jill Biden and Valerie Biden applaud behind him David McNew/Getty
Joe Biden attends a Super Tuesday event in Los Angeles as part of the 2020 Democratic primary with his wife Jill Biden and sister Valerie Biden.

The March 3 rush offered a case study in the law of unintended consequences — the clustering became so intense that no state, not even California, mattered that much. The risk of such a massive Super Tuesday is that it becomes made-to-order for a super-rich self-funder like Mike Bloomberg. He was the only 2020 candidate with the resources to advertise heavily in all 14 Super Tuesday states. Had Elizabeth Warren not memorably eviscerated him in a February 19 debate in Las Vegas, it is conceivable that Bloomberg could have bought his way into a delegate lead on March 3.

Solutions: Push back and break up Super Tuesday

This is not a new problem. Both the Democrats and the Republicans have been grappling with the clustering of the primaries since the 1980s. But it has been hard enough for the national parties to exert their limited powers to guarantee a protected period for the four early states. Threats from national party headquarters have limited impact when the power to set the dates for presidential primaries rests entirely with state legislatures.

Instead of a heavy-handed approach, the national parties have offered various blandishments over the years to states in an effort to space out their presidential primaries. The carrots have ranged from awarding bonus delegates to states that don’t jump the gun to the Republicans in 2016 permitting states to hold winner-take-all primaries beginning in mid-March. But nothing has worked. The bonus delegates have been mostly ignored as too paltry a prize, and the 2016 GOP winner-take-all primaries (which are barred under Democratic Party rules) had the boomerang result of hastening a rush to judgment without giving voters enough time to assess the candidates.

Oddly enough, the best example of states deriving tangible benefits from delaying their primaries occurred by accident. In 2008 North Carolina and Indiana were the largest of the seven states that waited until after May 1 to select their delegates. As it happened, Barack Obama and Hillary Clinton were still jousting for the nomination that spring, and the two popular Democrats lavished two weeks of intense campaigning on North Carolina and Indiana before their May 3 primaries. This burst of attention played a major role in boosting Democratic organizing in both states. And, partly as a result, Obama carried both states in November 2008 — the only time in this century that either North Carolina or Indiana has gone Democratic.

There is, of course, no way to institutionalize a protracted nomination battle. But it might help to space out the primary calendar if political parties were not so overtly fearful of having a delegate race stretch into June. In years past, it seemed rational for party leaders to be obsessed with anointing a de facto nominee in March or early April. The logic was that an early choice would allow the candidate to get a head start on fundraising and lessen the chances of lasting internecine feuds. But in this era of intense partisanship, raising money is no longer a daunting challenge (about $4 billion was spent on the 2020 presidential race, including funds from super PACs and hidden “dark money.”) And a strong argument can be made that the made-for-television drama of a hard-fought battle for the nomination can aid a political party in November, especially since in these partisan days it is much easier to forge party unity for the fall campaign.

Political parties do possess the power to do one important thing to add a note of deliberation to the primaries: to mandate a pause of at least a week between the last of the four early primaries and the inevitable Super Tuesday. In 2020 that would have given Democratic voters a chance to digest the South Carolina results, the turnabout in Biden’s political prospects, and the endorsement of the former vice president by Buttigieg and Klobuchar. Pushing back Super Tuesday would also lessen the chances that any early or absentee voter would have cast a wasted ballot for a candidate who dropped out of the race.

But in the end, the front-loading of the primary calendar directly flows from another sea change in politics — the end of conventions as decision-making bodies. If candidates are no longer scrapping to the end for delegates to take to the convention, then May and June primaries in most campaign years will merely offer voters the meaningless opportunity to ratify a choice that has already been made.

Conventional wisdom

The pandemic destroyed the last illusions that a political convention is anything more than free television advertising dressed up as a news event. Instead of thousands of delegates (and seemingly as many reporters) flocking to Charlotte for the Republicans and Milwaukee for the Democrats, both parties, out of necessity, went mostly virtual in 2020. About the only memorable live backdrops during the conventions were Joe Biden greeting supporters in their honking cars in a Wilmington parking lot after his convention address and Donald Trump, with dubious legality and worse taste, commandeering the White House for his acceptance speech.

Joe Biden is shown on a screen delivering a speech to a crowd of people and cars Alex Wong/Getty
People gather to listen to Joe Biden accept the 2020 Democratic presidential nomination at a parking lot outside the Chase Center in Wilmington, Delaware.

The broadcast TV networks, bowing to the remnants of civil obligation, devoted an hour of primetime for four straight nights to each convention, from 10 to 11 p.m. Eastern. Cable TV and PBS ran the full array of evening sessions, but even this extended coverage had its built-in limitations. There were no delegates to interview, no controversies, and scant traditional content. The Republican, reflecting Trump’s scorn for substance, even dispensed with a party platform. The Democratic convention was aptly likened by a Syracuse University media scholar, Robert Thompson, to “a 1970s variety show.”

The irony is that just six months earlier, in February 2020, it seemed like everyone in politics was predicting a “contested convention” for the Democrats in Milwaukee. With no candidate dominating the race after Iowa and New Hampshire, pundits and politicos were eagerly imagining scenarios under which the Democratic nominee would be selected on the convention floor for the first time since 1952. Nate Silver’s influential statistics-based website, FiveThirtyEight, put the odds at 41 percent that voters in the primaries would not award a majority of delegates to any Democratic contender.

Of course, there is virtually no one left in politics or journalism with any experience with a convention as an actual decision-making body. The most recent floor battles for the nomination were Ronald Reagan’s hard-fought challenge to incumbent Gerald Ford in 1976 and Ted Kennedy’s last-ditch bid to dethrone incumbent Jimmy Carter in 1980. Back in those days, there was no cable news, let alone smartphones, and no ability for anyone other than a network reporter to show video. In short, it seems almost as long ago as 1924, when it took the Democrats 103 ballots in steaming heat at New York’s old Madison Square Garden to pick a nominee.

Senator Ted Kennedy shakes President Jimmy Carter's hand Bettmann/Getty
President Jimmy Carter and Sen. Edward Kennedy shake hands at the conclusion of the 1980 Democratic National Convention in New York City. Kennedy, who had been the president’s rival in the primary, waited at his hotel during Carter’s acceptance speech before driving to Madison Square Garden to appear beside Carter in a gesture of party unity.

Had Covid-19 not intervened, 4,750 Democratic delegates would have attended the Milwaukee convention. It is almost impossible to imagine the bedlam if such an unwieldy group had been required to make a binding decision on the presidential nominee.

Despite the fantasies of political junkies, voters have displayed mixed reactions to the idea of allowing convention delegates to pick a nominee who had not dominated the primaries. In the spring of 2016, when it appeared that Trump would fall short of winning a majority of GOP delegates before the convention, pollsters asked Republican voters how the nominee should be selected. Typical was an April 2016 NBC News/Wall Street Journal poll finding that 62 percent of Republicans believed that the candidate with the most primary votes (even if it were less than a majority) should prevail at the convention. In late February 2020, before the South Carolina primary, Fox News polled Democratic voters, asking an analogous question about what should happen if no candidate were to corral a majority of delegates. By a margin of 50 to 38 percent, Democrats said they preferred to have the delegates choose the nominee rather than automatically deferring to the candidate with the most support in the primaries.

Complicating everything is the hostility of many Democratic activists to the existence of superdelegates, who are the nearly 800 elected and party officials who are automatically selected without having to endorse a presidential candidate in a primary. At first glance, it seems logical that these superdelegates — many of whom will run on a ticket with the presidential nominee — are entitled to a significant role since they have so much more at stake than a typical Democratic voter. But in recent years, Bernie Sanders’s supporters regarded superdelegates as an illegitimate mechanism for the party establishment to dictate the nominee. As a result of the continuing controversy, a compromise was reached before the 2020 primaries, giving superdelegates decision-making votes only on a putative second ballot at the convention.

The endless Democratic wrangling over superdelegates — which also occurred during the Obama–Clinton nomination struggle in 2008 — serves as an indicator of the controversy that would likely surround a nominee emerging from a contested convention. Democratic Party rules since 1980 have allowed delegates to vote their conscience (Rule 13-J in 2020) rather than robotically follow the result of the primaries in which they were selected. But it doesn’t take much to envision the uproar if a pivotal group of delegates switched candidates on the convention floor under the “conscience” rule.

Solutions: All pageant, no power

Since Biden would have been the consensus Democratic nominee even without the pandemic, it may seem odd to get caught up in might-have-beens about a contested convention. But sooner or later, a political convention is going to be transformed from a four-day pep rally into a decision-making body that may well choose the next president of the United States.

Elaine Kamarck of the Brookings Institution offers a plausible scenario: What if John Edwards had won a majority of the delegates in the 2008 Democratic primaries and, on the eve of the convention, the sex scandal that destroyed his political career hit the news? Because the delegates could “vote their conscience,” there was no chance that the Democrats would have gone forward in suicidal fashion with Edwards as the nominee. Another possibility would be a controversial or badly vetted vice president nominee unveiled at the last moment. (As a reporter, I covered both the controversy over the real estate dealings of VP candidate Geraldine Ferraro’s husband in 1984 and the shock at the 1988 GOP convention in New Orleans at the surprise selection of Dan Quayle for the number two spot).

Nothing equips a 21st century convention to make these kinds of in-emergency-break-glass decisions about the presidential ticket. The Democrats now allot a ludicrous number of convention delegates plus alternates. In contrast, the 1960 Los Angeles convention that nominated John Kennedy had fewer than one-third as many delegates as the 4,750 chosen in 2020. Imagine trying to get recognized for a parliamentary point of order from the convention floor when that “floor” extends halfway up the seating in a basketball arena.

Ronald Reagan and Gerald Ford shake hands at a podium Bettmann/Getty
A defeated Gov. Ronald Reagan wishes President Ford well in his campaign for the presidency at the end of the 1976 GOP convention in Kansas City.

There is a tempting argument to officially make conventions what they have become: a four-day televised celebration of the party’s presidential ticket. Decision-making power in the rare cases of a deadlock or a scandal would rest elsewhere — say, with the equivalent of Democratic superdelegates. This is not as outlandish as it may sound since the Democratic and Republican national committees currently have the power to choose a replacement candidate if there is a vacancy on the ticket after the convention has concluded. In 1972 it was the Democratic National Committee that selected Sergeant Shriver to replace Tom Eagleton as George McGovern’s running mate following revelations about Eagleton’s mental health.

If we have learned anything from the tumultuous 2020 political season, it is that any weakness in the system under which we nominate and elect presidents can be exploited. That is why — even though I long to witness the drama of a second ballot for president — I have sadly concluded that 21st-century conventions should be all pageant and no power.

•  •  •


The solutions to so many problems hobbling American democracy require overcoming obstruction in the Senate and rolling back state laws that interfere with voting and vote counting. But the only legislation involved in the nomination of presidential candidates are the state laws setting the primary dates. That means that the only real bulwarks standing in the way of practical reforms to the nomination system are tradition, the inertia of the political parties, and the profit-and-loss ledgers of cable news networks.

It would not take much to structure presidential debates around fairness and voter education rather than TV ratings. Both parties have the power to change their rules in the belated recognition that political conventions as decision-making bodies are political artifacts in the 2020s. Even the order of the early delegate contests can mostly be set by the national committees of both parties, especially since, as 2008 indicated, candidates would be reluctant to campaign in states that violate party rules by jumping the gun. Only the problem of the clustering of primaries on a Super Tuesday would require legislative action by multiple states.

There is, of course, no perfect way to nominate presidential candidates. But just because our current method of nominating presidents evolved almost by accident is no excuse for inaction as the 2024 races begin to be glimpsed on the far horizon. Republican presidential contenders may start declaring their candidacies in as little as 16 months. That is why the time to make changes in the system is now, before any alterations in the primary calendar and the rules for debates risk being viewed as boosting some candidates and hindering others.

At a time of rightful fear over the future of American democracy, it would be bracing to solve over the next year or two fixable problems in how the nation nominates would-be presidents.