Blanket Primaries or Ranked-Choice? Why Not Both?
A substantial number of Americans continue to voice dissatisfaction with current American electoral practices. This has put Justice Brandeis’s laboratories of democracy to work by prompting some states to exercise their powers to design election systems to experiment with various electoral reforms. Those powers derive from the state constitutions for elections of state officers; Article I, Section 4 of the U.S. Constitution to “[prescribe] the Manner of holding Elections for Senators and Representatives”, and the U.S. Constitution Article II, Section 1 to determine how electors for President may be chosen.
Many states have allowed their municipalities to experiment, with a few states adopting reforms on a state-wide level. In the latter category, some states, like California and Washington, have adopted what is sometimes described as the top-two, blanket, or Louisiana primary, while Maine has implemented ranked-choice voting. While these reforms have been innovative, study of their effects reveals limited success in achieving advocates’ promises. This post concludes that by combining both reforms, states will be better positioned to access the positive outcomes hoped for.
Former Republican congressman from Oklahoma, Mickey Edwards, has argued that the problem with American politics is our broken election system that rewards partisanship at the expense of good policy. Edwards argues that the culprit is the primary. Primary election turnout is notoriously low, with a recent Pew Research Center study celebrating a surge in participation in the 2018 House primary elections of 56% – a surge that still left turnout in that primary under 20%. The belief is that since political participation in primaries is so low, they are dominated by the most active, and most partisan, members of the parties, limiting the success of more centrist candidates, who are presumably more representative of their districts. In particularly politically active years, incumbents are at risk of being “primaried” by extremists in their party. Thus, the candidates in the general election tend to be more extreme, on both ends, than the district, and the eventual winner is then likely to represent only one extreme, rather than the district as a whole.
To address this concern, Edwards advocates for a reform known as “blanket” primaries, sometimes referred to as “Louisiana,” “jungle,” or, most accurately, “top two” primaries. This reform provides that in the primary election, there is a single ballot, with all candidates on the ballot regardless of party. The top-two candidates who receive the most votes then move on to the general election. This means that the candidates in the general election may both be from the same party, particularly in districts whose residents heavily identify with one party over the other. This may make for more competitive general elections, in that a candidate who is almost certain not to win isn’t on the ballot, in favor of a candidate who actually has a chance of convincing voters to vote for them. Also, along with Edwards’s hope that this will reduce partisanship by ensuring the candidates in the general election better represent the center of the district, advocates have also claimed that it will increase turnout.
Unfortunately, the data provides only marginal support for the proposition that blanket primaries reduce partisanship and increase turnout. An added challenge to blanket primaries is that because only the top-two move on, it is subject to vote splitting – if too many candidates from a party enter the race, they may end up without any candidates in the general simply by virtue of the number of candidates in the race, rather than voter preferences. Thus, blanket primaries are not a panacea, at least on their own, for electing more representative representatives.
Maine has moved in a different direction, adopting ranked-choice voting (“RCV”), a darling of electoral reform advocates for decades. RCV has a long pedigree, first having been promoted publicly in the United States by William Robert Ware in the 1870s. Some states and municipalities flirted with a version of RCV, known as single transferable vote, in the last great shift in party power during the early- to mid-twentieth century. RCV and similar systems have also been successfully used internationally – including in Australia, and Ireland.
While there are many different iterations of ranked-choice voting, Maine has adopted the most typical approach. There, voters may rank candidates first, second, third, and so on. If a candidate gets a majority of first-rank votes, they are declared the winner. However, if no candidate receives a majority of votes, then the candidate with the least number of votes is eliminated, and the voters who ranked that candidate first have their votes redistributed to their second-rank candidates. If no candidate has a majority, the new candidate with the lowest number of votes is eliminated, and their voters’ ballots are also redistributed. The system continues until a candidate has secured a majority of preferences.
Advocates of ranked-choice have also claimed that the system reduces partisanship, since candidates are encouraged to appeal to voters to rank them second, even if they can’t secure their first preference. Advocates also argue that it increases turnout by
making the ballot more reflective of voters’ wishes. As with blanket primaries, however, there is only modest statistical data showing that the turnout hopes are borne out.
States should consider merging the two systems, blanket primaries and RCV, in order to best access the benefits of each. The real problem that neither system effectively can address is the issue of turnout. Primary elections typically draw the most politically aware sector of the electorate that is most invested in who the candidates in the general election are, but turnout remains exceedingly low regardless of the system. Blanket primaries attempt to appeal to an electorate disenchanted by the current partisan model and looking to elect “the best candidate,” but since they continue to rely on a two-stage electoral model, they don’t resolve the fundamental problem. The most partisan members of the electorate participate in the blanket primary, and turnout surges, as it always does, in the general, after the partisans have already selected who will be on the ticket. Combining ranked-choice with the blanket primary would allow there to be a single election, ensuring the highest number of voters considering all the available candidates, not just those the partisans have already selected. Moving to a single election, with candidates on a single ballot regardless of party, and using RCV would ensure that the larger electorate would be able to weigh-in, allow them to rank their preferences, and ensure that the candidate who emerged was the preferred candidate of a majority of voters. This will increase elected officials’ mandate, and provide more information about what direction the district would like to go in.