Every four years a subset of pundits and political junkies speculate about a brokered convention where no candidate comes into the convention with enough delegates to secure the nomination on the first ballot, thus leading to discussions, multiple ballots and deal-making before a nominee is finally chosen. For much of American history this was common, but in the last half century or so that has changed. The last genuinely brokered Democratic convention was 1952 when the eventual nominee was Adlai Stevenson. Early opposition to Donald Trump among many Republicans led to speculation that the 2016 Republican convention would be brokered, but it didn’t work out that way.

Talk of a brokered convention has returned this year because according to longstanding party rules, a candidate must have 50 percent plus one delegates, in this case a total of 1,991, to become the nominee. Although it is still early, with a days to go before Super Tuesday it looks likely that one of two things will happen in the remaining Democratic primaries. Either Bernie Sanders will accumulate those 1,991 delegates before the convention, or nobody will. If it is the former, then Sanders will be the nominee, but if it is the latter there will be some uncertainty.

During the Democratic debate on February 19th, the candidates were asked whether they thought the candidate with a plurality of delegates should be the nominee. To the surprise of nobody who has been following the primary, all the candidates said no except for Sanders. Politicians answering questions to fit their own self-interest, as they all did, is nothing new, but that question was also a little misleading because not all pluralities are the same. For example, if Sanders comes to the convention with 45, or even 43, percent of the delegates and no other individual candidate has even 35 percent, Sanders should and likely will, be the nominee.

To a great extent the challenge for the Democratic Party if the convention is brokered would be to arrive at a nominee that is not Sanders but is acceptable to the Sanders delegates-or at least most of the Sanders delegates

It is also important to remember that no official Democratic Party entity that can make this decision. The only way Sanders could get from having, for example, won 1700 delegates coming into the convention, to winning the nomination on the first ballot is if another candidate releases their delegates and successfully urges them to vote for Sanders. Neither DNC Chair Tom Perez, nor anybody else, cannot simply decree it.

On the other hand, if Sanders has a plurality but fewer than 43 percent, the convention will look very different. If, for example, Sanders has 39 percent of delegates and the next candidate has 32 percent, it is very unlikely he would get to 1,991 delegates on the first ballot. If Sanders were to fail on the first ballot two major things would change going into subsequent ballots. First, delegates, including Sanders’ own delegates, would not have to vote for any specific candidate. On the first ballot, all delegates must vote for the candidate who they are representing unless they are released by that candidate, but on the second ballot that changes. Second, Superdelegates, a group that is made up of party leaders, elected officials and the like, are allowed to vote on the second and subsequent ballots. There are 771 Superdelegates, so beginning with the second ballot, candidates would need fifty percent plus one of the total delegates, which comes to 2,375.5, to win the nomination.

It is possible that if Sanders came very close on the first ballot, the Superdelegates would put him over the top, but it is also likely that failure to secure the nomination on the first ballot would reflect strong opposition to the Vermonter from a large proportion of the party and all but eliminate him from further consideration. The natural person to whom the delegates would then turn would be whoever came in second. While it is likely that Sanders will have the most delegates, and reasonably likely he will win the nomination on the first ballot, we can be much less certain about who the second place finisher will be. If it is Joseph Biden, Elizabeth Warren, Pete Buttigieg or Amy Klobuchar, it is possible that they would be able to appeal to a broad enough swath of delegates, and be acceptable enough to most of the party, that they could be the nominee.

The same cannot be said of the man who is most likely to come in second. While eschewing the plurality winner, albeit in this scenario one without a commanding lead, for a progressive like Warren or a mainstream liberal like Biden, would be something which Sanders supporters would possibly accept, asking them to accept the nomination of billionaire who has embraced conservative economic policies, not least by things like calling Sanders a Communist for the not exactly radical proposal that workers serve on corporate boards, would be a bridge to far, not just for Sanders loyalists, but for labor leaders and others worried about wealth inequality. It is possible that a second place Bloomberg can throw enough money around to get the needed delegates, but that would tear the party apart as not only Sanders supporters, but many others, including those concerned about his previous support for stop and frisk or his treatment of women while running his company, would be outraged.

A brokered convention would not only be politically dramatic, but it would be an enormous procedural and logistical challenge for the Democratic Party

This is where the brokered convention scenario gets most fascinating. There is a real possibility that the two candidates with the most delegates will be unable to get the additional delegates they need on the first or second ballot because they are uniquely unappealing to large parts of the Democratic Party and indeed the electorate. This will mean Democratic leaders and delegates at the convention will have to do two things; first determine who the nominee will be, and second hold the party together with some degree of unity. That will be difficult, but if they get it right the reward could be a big win in November.

If the convention gets to the point where neither Sanders and Bloomberg can get the delegates they need on the first ballot or two, the process will then be wide open as people can be nominated who never ran into a single primary. If that occurs, it is not hard to imagine somebody like Kamala Harris or Ohio Senator Sherrod Brown trying to patch a coalition of delegates together. Similarly, another present or former elected official, a retired general or other public figure could be nominated.

To a great extent the challenge for the Democratic Party if the convention is brokered would be to arrive at a nominee that is not Sanders but is acceptable to the Sanders delegates-or at least most of the Sanders delegates. There are few people who fit that description, but it is clear that if Sanders could not get the nomination after the first two ballots, Elizabeth Warren, regardless of how few delegates she might have of her own, would emerge as a plausible option. This might explain why Warren has been reluctant to strongly criticize Sanders.

Lastly, a brokered convention would not only be politically dramatic, but it would be an enormous procedural and logistical challenge for the Democratic Party. There is almost nobody alive who has participated in a brokered Democratic convention, so the party would have to figure it out and craft the rules as it went along. For a party that has bungled questions of who should be on the debate stage or how to aggregate preferences in the Iowa caucus, a brokered convention would be an extremely daunting challenge. Nonetheless, more than any other time in modern political history it looms as a very real possibility.


Follow me on Twitter at http://twitter.com/LincolnMitchell