In the wake of recent redistricting battles, proportional representation is having a moment. Ezra Klein devoted an entire episode to touting it as a solution to gerrymandering. A flock of Substackers and influencers chimed in. Even Kamala Harris and Pete Buttigieg came out as PR curious.
From the hubbub, you might assume that most students of US elections believe that multi-winner districts with proportional representation are the only full solution to the redistricting wars, and you’d be right. But beyond that point of consensus, agreement breaks down. Proponents split into camps over which type of proportional representation to pursue.
Proportional representation, after all, is not one thing. Every country that uses it—and most do—has its own version: closed list, open list, mixed member, proportional ranked choice voting, and countless hybrids.
Ranked choice or open list?
These myriad alternatives have elicited passionate debate among specialists for decades. Two factions currently dominate the movement in the United States.
1.FairVote and allies have spent more than three decades building momentum for ranked choice voting (RCV), including its proportional version. They contend that RCV is the best fit for America’s traditions and political structure, partly because it combines single-winner executive and proportional legislative forms and can work in both nonpartisan local elections and partisan races.
2.Protect Democracy and allies have built a newer effort to promote party-based proportional representation through methods such as open-list elections. They maintain that party-based voting methods yield stronger multi-party democracies than do ranked choice voting and cite the voluminous literature from political scientists underlining the centrality of strong parties to a healthy democracy.
In principle, these differences seem stark; in practice, they may collapse into insignificance. The US Constitution makes moving wholesale to proportional methods difficult: a single-winner elected president, Senators elected one at a time, and states with small delegations all bake in advantages for the two major parties. The upshot is that RCV might yield as many parties in Congress as would a party-based method. In this light, the arguments between these factions may turn out to be the proverbial two bald men fighting over a comb.
Constitutional constraints on multi-party democracy
A canon of political science is Duverger’s Law. It holds that any system of democracy that relies on single-winner races with pick-one voting and plurality victors favors two-party dominance. Conversely, proportional voting methods eventually nurture more parties.
Presidency. The preeminent role of the president in American government and culture, and the method used to select him or her, puts a classic Duverger situation on the main stage of American politics. The internal logic of a separately elected president, chosen through state-by-state winner-take-all electoral college competitions, discourages voters from marking their ballots for anyone but the major party nominees. The presidency itself does not rule out multiparty democracy but constrains it.
US Senate. Each state elects its Senators one at a time, in different years, by plurality, pick-one elections, which makes Senate elections classic Duverger terrain: as in presidential races, two parties naturally predominate.
Because the Senate is the more powerful chamber of Congress, the highest-stakes US elections, for president and Senate, both entrench two-party dominance. These commanding heights give the major parties the vast authorities of these offices. They also confer other political advantages: their candidates are automatically on almost every ballot they contest and are taken seriously by every voter.
The presidency and Senate let the two major parties build their war chests, institutions, and reputations; control media attention and public narratives; recruit promising candidates and operatives; and win entrée to virtually any audience they want to address.
50 States. The US House adds to this two-party entrenchment by virtue of its composition: half of states lack enough representatives to support a fully robust multi-party composition. That’s a matter of arithmetic.
A panel of experts assembled by the American Academy of Arts and Sciences notes:
. . . something close to a law in political science. Knowing the average number of seats per district, along with the total number of seats in a legislature, will generate a remarkably accurate prediction of the number of nationally competitive political parties in any given country.
More seats per district equal more parties. To produce an American politics with more than two parties, the panel recommends multimember districts with a minimum of three, and preferably as many as eight, House seats in each.
If every House district has three representatives, three major parties become mathematically possible, but two major parties remain the most likely outcome. If the aim is a true multiparty democracy with five or six parties, districts of seven or eight seats would be best.
Unfortunately, 26 states do not have seven or more seats; indeed, six states have only one each. Some 18 percent of House members come from states that are too small to reliably send five or six parties to Congress. That’s another built-in advantage for the major parties.
Voters and other political actors will think hard about supporting parties that have no shot at the White House, the Senate, or the House seats from more than half the states.
International experience offers hope for both paths
For the United States, therefore, a multiparty democracy with five or six main parties seems far-fetched, regardless of the model used. American proportional representation might get to three or four parties, but two would probably still predominate.
That result would still be a gratifying upgrade; indeed, it’d be much like the admirable proportional systems of New Zealand and Australia, which have two main parties and a few minor ones. The former employs a party-based voting method called mixed-member proportional; the latter, a version of proportional RCV.
Indeed, proportional RCV can yield as many parties as party-based voting methods.
Consider the island nations of New Zealand and Ireland, two former British colonies with about 5 million citizens each. Ireland has used proportional RCV since its independence in 1922, and New Zealand has used mixed-member proportional voting since 1996. Since 1996, Ireland has averaged 10 parties in its legislature, along with several independent legislators, while New Zealand has averaged around 7. In this comparison, RCV yielded more parties than did mixed-member proportional voting.
Thus, the argument that open-list or mixed-member proportional elections will give the United States a more robust multi-party democracy than ranked-choice voting is strained. The Constitution and international experience suggest otherwise.
So while proportional representation is having a moment—its US prospects have brightened, and its potential is exciting—many obstacles litter its path. Ultimate success likely depends on unity among proponents, on advancing whichever method is politically attainable in the circumstances of each locality and state, and on no longer fighting futile battles over a comb.
Alan Durning is the founder and executive director of Sightline Institute in Seattle, a nonpartisan, nonprofit think tank for Cascadia.
