To blame our current party system for the dysfunction of our democracy is not to argue that we would be better off without parties. As Drutman says, reinvigorating our party system is “the only path forward.”
Drutman also rightly argues that we should reform the laws governing parties if we want to increase political dynamism. But he is wrong to think that fusion voting will bring back that dynamism or create multipartyism.
The heady, mid-nineteenth-century period of party politics that Drutman admires was the product of an era when parties were unregulated, private entities. The problem with our party politics today isn’t that fusion ended. The problem is that the party apparatus was co-opted by the state around the same time.
As Drutman points out, states began to ban fusion voting in the late nineteenth century after they took on responsibility for printing ballots. No longer would parties send voters into the booth with tickets printed by their parties; instead, candidates’ names would now be printed on one public ballot, which voters could use to keep their vote secret. But just as states took responsibility for running the general election with the use of unified ballots, they also began to take responsibility for running party primaries. Starting in the early twentieth century, taxpayer dollars were committed to funding primaries that selected the candidates from each party who would appear on the unified ballot in the general election.
But public money always comes with strings. Along with taxpayer funding came new state requirements about what it takes to form a party. In the early nineteenth century, parties formed simply by organizing; groups of people came together to develop and execute a shared agenda. With early twentieth-century reforms, states began to require that parties meet certain criteria—through signature gathering and party conventions, charters, and platforms—to gain the right to place a candidate on the ballot. States also introduced a minimum number of votes required to maintain status as a party.
In other words, running a party became a state-regulated activity—and diverse rules proliferated around the country. The rules significantly raised the barrier to entry for new parties, and the diversity of rules across states made matters even worse. In order to function nationally, parties today have to master fifty different sets of procedures. Even though the Libertarian Party was founded in 1971, it did not achieve ballot access in all fifty states until 1980, and it has not maintained that access consistently, achieving it again only in 1992, 1996, 2016, and 2020. The Green Party, founded in 1984, has thus far maxed out at forty-four states.
Fusion voting simply cannot solve this problem: the difficulty of getting a range of candidates on the ballot in the first place. Indeed, fusion voting seems to make very little difference even where it does exist. Consider the Working Families Party in New York. In 2000 the party claimed roughly 0.07 percent of registered voters. This year, after nearly a quarter-century of effort, it represents about 0.42 percent. Over the same period, the share of Democrats grew from 46 to about 49 percent. This is not favorable evidence that fusion will promote party dynamism.
A better reform would use taxpayer dollars only for public functions—like keeping our votes confidential—while returning to parties the functions that should be the responsibility of civil society associations, like selecting their standard bearer. Alaska’s new election system, passed in 2020 and first implemented in 2022, does just that.
In particular, the state now runs just one, unified primary. Instead of requiring candidates to go through a party process to get their names on the ballot, the ballot includes everyone who meets certain requirements established by state law and that vary by office (typically, the gathering of a sufficient number of signatures or the paying of a fee). All candidates can choose to be identified by the label of the party with which they affiliate as a way of providing voters with information about their positions. Meanwhile, parties can still hold conventions and endorse a candidate—and the endorsement will show up on the ballot—and they are fully responsible for funding that process. Parties also cannot block candidates who affiliate with them but aren’t endorsed, provided those candidates have met the state’s requirement for ballot access.
The result is an all-candidates primary on a unified ballot. If four or fewer candidates qualify for ballot access, the primary is canceled, and all candidates move on to the general election. If more than four qualify, voters pick the one candidate they prefer, and the four candidates who win the greatest number of votes move forward to an instant runoff in the general election, where ranked choice voting is used to ensure a majority winner. In the general, voters are invited, but not required, to vote not only for a first choice but also for a second and third. This structure is important; since it enables voters to select backups, third-party candidates can maximize their vote share. This method should quickly make third-party strength visible as it starts to grow, which in turn will draw media attention and create a positive feedback loop that will enhance public learning. That should make it possible for third parties to establish their footing better.
This kind of reform is necessary and urgent, and other states should follow Alaska’s lead. But by itself, it won’t restore the system-wide dynamism that Drutman advocates. We also need to change the rules that determine when an organization has earned the status of a political party and can put a candidate on the public ballot.
That will be much harder, but I think Drutman is right when he says that we should focus on the sweet spot where impact and feasibility meet. Right now, that is democracy, Alaska-style.
