Mattias Polborn is an economics professor at the University of Illinois who co-wrote a paper that studied the trade-offs between voter coordination and candidate quality under different primary election systems. As the race to determine the 2016 Democratic and Republican presidential nominees lurches toward Super Tuesday on March 1, Polborn, also a professor of political science at Illinois, spoke with News Bureau business and law editor Phil Ciciora about his research on primary elections.
Your research points to the current primary system as being unwieldy but ultimately the best way to winnow the candidates down to the “Condorcet winner” – that is, the candidate who would prevail in a head-to-head election against any one of the other candidates. Why is that?
Whenever there are three or more candidates who appeal to different types of voters, who to vote for is a much more difficult decision than when there are only two candidates.
Candidates with similar positions compete for the same pool of voters. For example, in New Hampshire, the Republican "establishment candidates" – John Kasich, Jeb Bush, Marco Rubio and Chris Christie – received a combined vote share of 45 percent. That’s more than Donald Trump's 35 percent, but it was split among the four of them, which means that they get fewer convention delegates than if there was only one consensus establishment candidate.
Similarly, there is a reason why Ted Cruz’s campaign, during the Iowa caucuses, passed on the rumor that Ben Carson was dropping out of the race. Most religious conservatives who were thinking about supporting Carson would probably move to Cruz if Carson indeed dropped out. But as long as Carson is in, Cruz and Carson split their votes.
Why not just have a national primary election day, in which all 50 states and the District of Columbia vote at once?
The problem with a national primary day is the coordination problem I just explained. If we had a national primary day, it would become a lot more likely that the "wrong" candidate is nominated.
I’ll use the 2010 Illinois Republican primary for governor as an example. There were seven candidates, but only state Sen. Bill Brady came from downstate, while the remaining serious candidates all came from the Chicago area. Brady received only 21 percent of the statewide vote and most likely was not the strongest general election candidate that Republicans could have nominated. But Brady won the primary nevertheless because the Chicago-based candidates split the vote there very evenly. Brady went on to lose what should have been a very winnable general election for the Republicans.
What’s your prediction for the current Democratic and Republican primaries?
On the Democratic side, there are two candidates left in the race, and I don’t see that either of them will drop out very soon. It’s going to be a long race, more or less like in 2008. Hillary Clinton probably still has a slight edge, but with any new negative information about her, Bernie Sanders may very well win the Democratic nomination.
Interestingly, the betting markets currently view Sanders as the stronger candidate for the Democrats in the general election. According to the betting markets, if Sanders wins the nomination, his chance of winning the White House is about 68 percent, while it’s 59 percent for Clinton.
On the Republican side, we will soon see the race narrow down in a few weeks to effectively three contenders: Trump, Cruz and one more out of Rubio, Kasich or Bush. And they will fight fiercely for it among themselves.
After that, the next one to drop out is either Trump or, more likely, Cruz. I think this will happen by the end of March – the later it happens, the more beneficial it is to the establishment candidate.
Even if I’m wrong about the sequence in which candidates will drop, it will be very unlikely that the Republicans will go to a brokered convention. Dropouts will occur early enough for one candidate to accumulate a majority.
Critics argue that the current system affords disproportionate influence to the early primary states – Iowa, New Hampshire and South Carolina, for example – that don’t necessarily reflect the demographics of the rest of the U.S. Would it be better off to have different states in the early stages of the primaries? Say, Illinois instead of Iowa, perhaps?
I am not a big fan of this “representativeness argument.” In a primary, candidates are not chosen by the population at-large of a state – Democratic primary voters choose the Democratic candidate, and Republican primary voters choose the Republican candidate.
It may well be that, this year, Democrats and Republicans choose two very extreme or flawed candidates, but the fundamental problem is that Democratic and Republican primary voters are quite extreme. I don’t think that this problem could be fixed by changing the sequence in which states vote.