You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

Our Broken Presidential Nominating System

The modern primary produces odd candidates and has become less democratic with time. There's a better way.

David Calvert/Getty Images

Portraying himself as a no-nonsense businessman and an anti-politician unafraid to speak his mind (even if his mind changes), and practicing a calculated politics of insult, Donald Trump has racked up early, important wins in the GOP nominating contest. He now has establishment Republicans worrying how his reckless, bellicose style will go down in the fall if, as seems likely, he becomes the party’s nominee. Some Republicans have been quoted recently calling for a brokered convention, reminiscent of the days when a political party could set aside the popular vote and select someone less divisive or more electable than the frontrunner. For that is how the two major political parties used to work. About a dozen times in the twentieth century, parties nominated candidates who were not the top primary vote winners—including William Howard Taft, Woodrow Wilson, Dwight Eisenhower, Adlai Stevenson, and Hubert Humphrey—because they were deemed the best suited to lead the party’s ticket to victory.

But beginning in 1972, the two parties began reforming their delegate systems to remove party and elected officials from their role in vetting presidential nominees at national conventions. Politicians were sidelined, and ordinary voters were empowered in a series of sequential primaries and caucuses. As a result, today he or she who wins the most delegates chosen (directly and indirectly) in sequential contests becomes the nominee, crowned at the national conventions.

This system, of course, has the merits of transparency and apparent democracy. It has led to some 50 million voters participating in the nominating process. But the reforms have also unfolded in unforeseen ways with calamitous results. Party adherents have become bystanders to a process that incongruously results in the selection of their own leader. The so-called nominating convention, which used to be an occasion of decision—when people bargained, pressured, deliberated, and decided—now is no more than a televised spectacle.

RNC Chairman Reince Preibus has even found himself chasing Trump to get him to pledge that he would support the Republican nominee. It used to be candidates sought the endorsement of the party; now it’s the other way around.

Today the main drivers of the nominating process are the candidates, who select themselves to run. This self-selection is a problem. The modern process of running for president—24 months of nonstop fundraising, travel, and relinquished privacy—is so unpleasant and degrading that it requires something like a personality disorder to submit to it. When Ted Cruz’s former college roommate said he would rather see as president someone plucked from the phone book than his former roommate, he was putting his finger on a deeper problem than just one man’s personality. Rather than a selection process, we have an adverse selection: The individuals volunteering for president are often precisely the persons you would never want in that office.

That is the first unforeseen consequence of our post-1972 reforms: a candidate-centric system that produces quirky, sometimes frankly odd, candidates. The second is that, for a system that purports to be democratic, designed to prevent party bosses from overturning the will of voters, it is profoundly undemocratic.

In 2015, more than half a dozen candidates dropped out before a single ballot was cast. So much for democracy. Now that campaigns have to address so many voters, money is vastly more important than it was in the mid-twentieth century. The 1972 reformers did not anticipate that in trying to empower average voters, they would ultimately empower hedge fund managers and casino owners.

When voting does finally start, the problems get worse. After the first few contests, most candidates drop out, as we have seen this past week, depriving later voters of having their say. In 2000, both Al Gore and George W. Bush became unstoppable before 33 states had voted. In 2004, John Kerry locked up the nomination on Super Tuesday, by which time less than 3 percent of the nation’s electorate had voted for him. In 2012, by the time Mitt Romney became the de facto nominee, he had won a majority in just three states, and yet to vote were New York, California, Texas, New Jersey, and Pennsylvania. If democracy requires consulting more than a fraction of voters, our nominating system is hardly democratic.

What makes this early culling of the field worse yet is that when you have more than two candidates in a contest, plurality makes a winner—but plurality victories are notoriously misleading. If you asked voters to rank their second, third, fourth, and fifth choices, only about 50 percent of the time would the plurality winner be the most preferred (as political scientists have known for a long time). (This is one reason Britain’s Labour Party, for example, uses a complex voting system that eliminates the candidate with the least first-choice votes, redistributes his or her supporters’ second-choice votes, and continues this elimination and redistribution until one candidate has a majority.) Thus it’s entirely possible that candidates who drop out early might actually be more preferred than those who remain, and that the person catapulted toward the nomination by early plurality victories is, in fact, among the least popular of the candidates. Trump today, having some of the highest unfavorability ratings in the GOP race, fits the latter category. Sequential voting combined with plurality wins makes for random, unreliable results.

Alexander Hamilton remarked about the Constitution that at last people would impose on themselves a government by “reflection and choice” instead of by “accident and force.” We’re back to accident and force, or more accurately, accident and money.

What to do? In the 1980s, Democrats created super-delegates of elected officials to act as a counterweight to aberrant primary results, but these days super-delegates are expected to vote in accordance with state primary results, and in any case have never determined the outcome of a nominations fight. Once you’ve decided to consult voters, it is awkward to overturn their verdict. That means that if the parties are to awaken from their stupor and seek a role for themselves in their own affairs, they must insert themselves before voting begins, not after.

Nominating conventions could, for example, be rescheduled to occur before the primaries, composed of delegates—party members, elected officials, local notables—who run not committed to any candidate. In the convention, delegates would do as they used to do—listen to candidates, vet them, and engage in the horse-trading and deliberation that are the desiderata of political parties everywhere. Why have a party at all if it doesn’t help sort out national conflicts, resolve differences, and make decisions? After listening and vetting, delegates could nominate not one but two candidates to put before voters with the party’s imprimatur. Why just two? To eliminate the false-winner syndrome of multiple-candidate races. Then primaries and caucuses would occur, much as they do now, but the sequence of contests would not pose the problems it does today, for there would be no plurality wins, no premature culling, and later voters would have a say. After the primaries, whoever won the most primary and caucus votes would be the party’s nominee.

This is not a proposal to bring back smoke-filled rooms where party bosses overturn the will of the people. In this hyper-democratic age, the return to bossism is unacceptable. It is instead a request that we rethink our dysfunctional nominating system. There are other ways to fix the problem. Reinstating the two-thirds threshold for Democratic nominations (and instating such a threshold for Republicans) is also worth considering, as under such a rule no candidate would drop out prematurely and conventions would again become venues of bargaining and compromise and would likely produce nominees with broad party support. There are still other fixes to consider.

If either Ted Cruz or Trump ends up as the Republican nominee—candidates intensely but narrowly supported—and brings his party to a landslide defeat in the autumn, at last the moment may arrive when political parties might awake from their passive stupor, change the nominating rules, and take charge of their own affairs again. If not, we will continue to be saddled with the kind of mediocre candidates we have today, in some cases worse than mediocre.