Drutman is not alone in his diagnosis of the problem with American democracy: that most roads, if you follow them long enough, lead back to the two-party system. Research has linked the system to a list of maladies: hyperpolarization, inept governance, uncompetitive elections, depressed voter turnout, underrepresentation of minorities, and even a greater risk of political violence. If there is such a thing as a scholarly consensus in political science today, it is that America’s two-party system, arguably the world’s strictest, is not doing our stumbling democracy any favors—and that the path out of this morass runs through more and better parties.
In fact, if there is such a thing as a consensus among Americans generally today, it is the same. Dismay with the two parties has reached new lows, with “unaffiliated” now the country’s largest voting bloc. Registered partisans tend to not even like their own party as much as they simply dislike the other. A full 70 percent of Americans wish there were more. How is it, then, that in a democracy where a sizable majority prefer something else, no viable alternatives are, for the most part, ever available? The answer is both simple and arcane, located in a 1967 statute passed by Congress. Bringing multipartyism to America will require changing it.
Political competition is a function of electoral rules. Just as a duopolistic market does not come about by chance, a party system confined to only two players is a predictable consequence of rule choices. And one rule, more than any other, governs the number of nationally competitive political parties: the number of representatives elected from each legislative district.
In the United States, each congressional district sends one representative to the House. When only a single seat is up for grabs, two parties typically emerge. That is, single-member districts tend to generate predominantly two-party systems. By contrast, in a multiseat district, multiple parties have a chance at winning. More opportunities to win a seat and a lower threshold to secure one incentivizes more parties to compete. Countries like Uruguay with a modest number of seats per district (on average, 2.5) generate modest multiparty activity (there, three large parties), while others like Israel (120 seats in one nationwide district) can generate unwieldy party systems. The more seats per district, the greater the number of political parties that are likely to contest them.
This logic follows 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.
The United States inherited its first-past-the-post electoral system from Great Britain: it was “the only game in town in 1787 and for some generations thereafter,” as Robert Dahl observes. “The Framers simply left the whole matter to the states and Congress, both of which supported the only system they knew.” What was bequeathed as a colonial artifact is now codified in federal law. The 1967 Uniform Congressional District Act (UCDA) is the latest, and least flexible, in a line of mandates dating back to the nineteenth century requiring the use of single-member districts for the U.S. House. Over the course of the twentieth century, as most other democracies experimented with multiparty systems, federal law largely locked the U.S. party system into place.
Any realistic effort to allow for multipartyism in America at the federal level must then contend with the UCDA. While the mandate itself is simple—it is all of a single sentence—the question of what to put in its place is less so. Instead of one representative per district, should it be six, where possible? Eight? How variable might it be from one district to the next? What about states with only a single representative, or two or three? Should Congress establish general parameters, and then otherwise let each state decide? The decisions bear directly on what a new multiparty landscape would look like.
Amending the UCDA could force other policy decisions that would also influence the nature of America’s party system. For example, in multimember districts, parties should win seats in rough proportion to their votes. In a six-seat district, if a party wins 50 percent of the vote, it should win three seats. If that party had won 49 percent of the vote, it should still probably get three seats; but some standard allocation formula would be required to say so—and certain formulas tend to favor larger parties over smaller ones. Or consider the size of the House, which also influences the number of viable parties. It, too, should probably increase, but by how much? Without expansion, Massachusetts’s nine districts could be easily collapsed into three, each sending three representatives to Congress. But nothing would change for neighboring Rhode Island with its lone representative.
Of course, policymaking can only go so far in designing a multiparty system. Various features of the U.S. political system are likely to cut against the number of new national parties. The presidency, for example—an inescapably single-winner race—would almost certainly continue to be a contest between the two major parties (though as in other presidential systems, each could form electoral coalitions with minor parties). And while federal law can create space for more parties, parties must still be built—and a variety of state-level regulations, such as restrictive ballot access laws, inhibit the ability of new parties to form and contest elections. Among the most significant barriers are state laws banning fusion voting, arguably the most successful ballot mechanism in U.S. history that supported the development of minor parties.
Loosening these restrictions is a here-and-now place to start for reformers, as Drutman rightly argues. Several organizations—including ours, Protect Democracy—are litigating to overturn bans on fusion and exploring ballot measures to advance multipartyism in state legislatures. Contending with federal law may be an important long-term ambition, but in the short-term, we need not wait for Congress to start the work of building a multiparty America. Despite different timescales and levels of government, the locus of reform is the same: revisiting the rudimentary rules that quietly structure the American party system. The goal should not be to circumnavigate parties, no matter how dismal their current performance, but to build more of them.
Here again we agree with Drutman: no matter how broadly or deeply disliked, political parties are indispensable to modern democracy. They are the institutional vehicles that connect citizens across common sets of interests, mobilize them, marshal resources, mediate coalitions, cohere policy ideas, and coordinate policymaking. When incumbent parties calcify, other parties should compete to speak to the disaffected—and in most democracies, they do. “In a multi-party democracy, politicians, activists, and voters can leave a party if they are dissatisfied with it and join or even create another,” write political scientists Hans Noel and Seth Masket. But “this is close to impossible in the United States.”
The reason is not for lack of trying. Today, third-party bids are drawing in millions of dollars, but they will inevitably flounder. “Independents” will outnumber both Republicans and Democrats in the 2024 election, though nearly all will vote for major party candidates anyway. Primary voters are signaling displeasure with their choices on both sides, but most will fall in line by November. While a flurry of popular reforms promising to “give voters more choice,” from open primaries to ranked choice voting, are proliferating across the states, none will result in more parties winning seats.
That’s because, despite the consensus and energy for change, the single-member rule, more than any other, binds the United States to a two-party system. Until it changes, we’re bound to its consequences, too.
