Democrats Have a Primary Problem for 2020
Democrats have a problem. And I’m not talking about the often-cited issues -- Republican dominance in Congress and the states, a tough map in the 2018 Senate elections or a potential fight between progressive and moderate factions. Democrats are sitting on a system of presidential primary rules that make the party vulnerable to a contested convention, and it appears they’re not really discussing it.
Before getting into the details, here’s a quick refresher on Democratic nominating process. Delegates have historically been allocated proportionally based on how much of the vote candidates get (provided that they reach a certain threshold of support). In recent cycles, Democrats have also had “superdelegates” -- elected officials or party leaders who can vote for whichever candidate they choose.
Proportional systems like this encourage some level of consensus, but that can be a double-edged sword. The advantage is that it’s difficult for a factional candidate who fails to build a majority of support to run the table on delegates early by winning pluralities (the GOP rules are more conducive to that). The disadvantage is that if the field doesn’t narrow to a few candidates quickly, the rules could lead to a contested convention.
To show this, I ran a simple, illustrative scenario of a 2020 primary between former Massachusetts Gov. Deval Patrick and Sens. Kirsten Gillibrand and Sherrod Brown. I assumed that each state would have the same number of delegates that it did in 2016 (an incorrect assumption but, again, this is illustrative). In every state where blacks made up more than 20 percent of the overall population, I gave Patrick 40 percent of Hillary Clinton’s delegate haul (Clinton’s coalition in those states was often majority black). In every other state, he got 20 percent of Clinton’s share of the delegates and 15 percent of Bernie Sanders’s. In 2020, Patrick might perform significantly better or worse than this (if he were to run), but the point is that he might take some votes from a more center-left candidate like Clinton (especially in heavily black states) and a bit less from a died-in-the-wool progressive like Sanders. I gave Gillibrand any Clinton delegates that didn’t go to Patrick, and Brown, a progressive, got the remainder of the Sanders delegates. Under this scenario, Gillibrand, Brown and Patrick won 42 percent, 39 percent and 19 percent of the pledged delegates (those won in primaries and caucuses), respectively.
In other words, no candidate would have majority in an explicitly majoritarian system.
In past Democratic primaries, superdelegates might have swooped in to solve the problem. For example, in 2016 Clinton won a majority of pledged delegates but failed to get a majority of all delegates, so the superdelegates decided to push her over the top well before the convention. And in this fictional scenario, they might coordinate and give Gillibrand (the plurality winner) the nomination.
But there’s no guarantee that would happen in 2020. Before the 2016 primary ended, many Sanders supporters argued that superdelegates should be bound to vote according to the results in their home state. They’ve already made some progress in states like Maine. Their basic argument -- that the presence of unbound superdelegates dilutes the power of rank-and-file voters and makes the Democratic primary less, well, democratic -- is strong. But if they significantly reduce the number of unbound superdelegates, Democrats will have less power to prevent a multi-candidate fight. Those fights might have to be resolved by the candidates and delegates before or at the convention.
Obviously this is all highly speculative. Maybe by the time 2020 rolls around, one candidate will have emerged to steamroll the opposition and win the nomination outright. Or maybe it’ll be another two-candidate contest like those in 2008 or 2016 -- neither of which ended in delegate-swapping deals or a convention floor fight. Or maybe there is a multi-candidate melee, but the plurality winner resolves it by making a deal with one of his or her competitors.
But the point remains that current Democratic rules -- where the overwhelming majority of delegates are given out proportionally -- is more fragile than most realize. We don’t even need a candidate with as much sway as our hypothetical Deval Patrick to show that. In the 2016 GOP primary, Ohio Gov. John Kasich won 161 delegates out of the 2,472 available -- only 6.5 percent of the total (and he drew just 13.8 percent of the popular vote). That’s a better showing than most presidential candidates, but it didn’t put Kasich anywhere near winning the nomination.
But if a candidate as weak as Kasich had exclusively pulled delegates away from Hillary Clinton (who won 54.4 percent of the non-superdelegates), she wouldn’t have had a majority of pledged delegates heading into the convention. Clinton likely would have still won the nomination in that case (through dealmaking or superdelegates), but the point is that adding even a weak third candidate could keep a strong front-runner from claiming a majority of votes or pledged delegates.
One could also argue that a contested convention almost assuredly won’t happen because there are incentives for relatively weak candidates to drop out. This theory makes sense intuitively. Running for president is difficult, costly and, frankly, not a lot of fun. Moreover, there’s no guaranteed prize for second place, so once a candidate knows he or she won’t win the nomination, there aren’t many good reasons to stay in the race.
But this argument has trouble dealing with the realities of the 2016 election. Bernie Sanders was never in a strong position to beat Hillary Clinton in a one-on-one matchup. The delegate and demographic math didn’t add up for him -- a fact that most observers realized well before the primaries even started. But Sanders persisted anyway, either erroneously believing he would win or in an attempt to spread his message more broadly throughout the Democratic Party.
Similarly, many 2016 Republican presidential candidates stayed in the race for an extended period because they were responding to a different set of incentives. Each of them believed (not without reason) that if they were to end up in a one-on-one matchup with Donald Trump that they would win. So Kasich, former Florida Gov. Jeb Bush, Florida Sen. Marco Rubio and others all stayed in the race even after it was clear they wouldn’t win the nomination.
In other words, the traditional incentives that winnow the field don’t always work. Moreover, if a convention looks like it might be contested, then any candidate with a decent number of delegates has a big incentive to stay in and try to increase their bargaining power (or maybe even win it all).
It’s also important to note that this scenario would be unprecedented. Since the dawn of the modern presidential primary, Democrats have managed to settle on a choice before they headed to their convention. So it would be easy to argue that if Democrats were vulnerable to a contested convention, we would have already seen it happen.
But risks that haven’t materialized yet still exist.
For example, in the RNC’s 2013 Growth and Opportunity Project (more commonly known as “the autopsy”), top Republicans argued that an unnecessarily long and contentious primary season had bruised Mitt Romney and that the process should be accelerated. Theoretically, that would have allowed a consensus candidate to emerge more quickly and get a head start on his or her general election matchup with the eventual Democratic nominee.
But many Republicans at the time failed to realize that their system -- which includes a number of winner-take-all or winner-take-most primaries -- was vulnerable to being swept by a factional candidate who won pluralities in a crowded field. Every previous nominee had been a more or less a mainstream, traditional conservative, so party officials thought the 2016 process would produce another such candidate. Instead, they got Donald Trump. Had he lost the general election (and he only narrowly won), many Republicans would have blamed the primary system for not producing a more unifying candidate.
Democrats shouldn’t make an analogous mistake. They should at least be aware of the so-far-unrealized structural vulnerabilities in their system as they begin to hash out the details of their next primary.
Finally, a contested convention seems objectively unlikely. There are an enormous number of possible scenarios for 2020, and the vast majority of them don’t end in a convention fight. But it’s worth noting that the Democrats’ rules (like every set of rules) create winners, losers and risks. And, in my view, the risks on the Democratic side are underappreciated.