It’s Time To Talk About Electoral Reform

Over the past several decades, advocates for improving American democracy have focused their efforts on a few discrete issues. One of these is voting—trying to make it easier to vote, secure the voting process, and ensure equal access to the ballot box. Another is money in politics—working to make political spending transparent and prevent it from corruptly influencing elected officials. A somewhat distant third, until recently, is redistricting—preventing politicians from gerrymandering their districts to keep themselves and their allies in power.

All of these issues are critically important. Democracy means little without the right to vote, and it is at least compromised if money or district lines drive political outcomes.

However, there is another equally fundamental issue that has, until recently, received only niche attention. That issue is electoral reform—examining the most basic rules of the political system and finding ways to ensure that they are promoting effective, representative government. The electoral system is the set of rules that determines how representatives are selected: which candidates and political parties can compete on the ballot, who those candidates and parties represent, and how voters are allowed to choose between them.¹ These rules shape the makeup of Congress and local and state legislatures, and they create strong incentives that guide the behavior of elected officials.

This report describes the important elements of the U.S. electoral system and identifies the reasons why they are not functioning well. It also describes a range of possible solutions that deserve further consideration, including fusion voting, ranked-choice voting, primary election reform, multimember districts, and methods of proportional representation.

Two fundamental problems with the U.S. electoral system are worth emphasizing upfront:

1. Electoral rules discourage problem-solving and reward conflict: America has a two-party political system that encourages candidates to appeal, first and foremost, to members of their own party—while locking out independents, third parties, and other sources of competition. For the most part, instead of working together to solve the nation’s problems, the two major parties engage in an endless tug of war. Disagreement is to be expected in a democracy, but at the end of the day, representatives should be motivated to find areas of agreement and to pass legislation that the public supports.
2. Electoral rules impede representation: Both major parties depend on the support of their own partisan voters, or at least on the prevailing majority of those voters. This means that there are few genuine moderates in Congress and that both parties, at times, are out of step with the median American voter. It also means that diverse viewpoints are underrepresented, whether from people who do not fit neatly on the left-right spectrum, communities of color, or a host of other cultural, geographic, and political groups. There are two points on the political spectrum, the core of the Republican Party and the core of the Democratic Party, that exert a gravitational pull—and in recent years, the core of each party has sometimes veered to ideological extremes.²

According to polling data, Americans are overwhelmingly unhappy with politics, with Congress, and with both major political parties.³ And while voting restrictions, the influence of money in politics, and the corrosive effect of partisan gerrymandering all play some role, the electoral system itself is a major contributor to the current era of political dysfunction.

Recent research shows that many American voters are moderates,⁴ and some have views that do not cleanly fit on the left-right spectrum.⁵ Moreover, whether Democrat or Republican, many voters—if not most—would prefer a government that is professional and responsive, in which politicians work together to solve the nation’s problems.⁶ Unfortunately, however, that is not the government that America’s electoral rules incentivize politicians to deliver.

In the United States, most elections have three main structural components: 1) partisan primaries; 2) first-past-the-post winners; and 3) single-member districts. Each of these pieces of the electoral system contribute to the current era of political dysfunction.

1. Partisan primaries

Most candidates in the United States are chosen in party primary elections.⁷ Typically, the participants in these primaries are strongly partisan voters.⁸ This means that candidates only need to win a bare majority of committed partisans to participate in the general election. And whether voters find themselves in the middle or at the ends of the political spectrum—or not really fitting on the spectrum at all—they will typically have only two candidates to choose from once the primaries are over: a Republican candidate somewhere on the right and a Democratic candidate somewhere on the left.

2. First-past-the-post winners (also known as “winner take all”)

The term “first past the post” is a reference to horse racing: The first horse to cross the finish line is the sole winner, whether it leads by an inch or a mile. U.S. elections are referred to as first past the post because, typically, the candidate who receives the most votes wins—even if they have less than 50 percent of the votes, which often happens when more than two candidates are competing. This strongly disincentivizes independent and third-party candidates from participating. If a third-party or independent candidate enters a race, they are most likely to serve as a “spoiler,” siphoning votes away from the candidate with whom they are more closely aligned and therefore helping their furthest rival. As a consequence, most elections feature only two major-party candidates. And because there are only two candidates, neither candidate necessarily needs broad appeal to win; they simply need to be a hair more appealing than their opponent. Often, the best strategy in these circumstances is to run a negative campaign.

3. Single-member districts

Unlike legislators in many other democratic systems, legislators in the United States are generally elected from single-member districts—districts where there is only one winner. Single-member districts have at least three major disadvantages. First, since there is only one winner, the voters who do not support the winning candidate are simply out of luck; no matter how diverse the district is, it only gets one representative. Second, single-member districts make the emergence of third parties difficult, even without first-past-the-post elections, because third-party candidates must achieve a top finish in an entire district to win any representation at all—a difficult feat when every non-major party in America lacks the infrastructure of the major parties. Finally, single-member districts enable gerrymandering, which worsens all the problems already discussed; it gives parties more power to rely on their most partisan supporters and to lock out competitors. Without first-past-the-post elections and single-member districts, gerrymandering may be nearly impossible.⁹

In summary, these three features of the U.S. electoral system usually give voters no more than two choices: one Democrat and one Republican. Those candidates must cater to the strongly partisan voters who elect them in the party primaries, but they still win in general because first-past-the-post elections in single-member districts do not allow for meaningful alternatives.

To be sure, factors outside the electoral system make this dynamic even worse. Perhaps the most important is that governing, like elections, is winner take all. In Congress and most state legislatures, the party that wins the majority of seats also wins control of the legislative agenda. With only two major parties, this creates a strong incentive for each side not to work with—let alone make compromises with—the other side.¹⁰ It is a two-sided, zero-sum competition. While it might make sense, in the short term, for the minority party to work together with the majority, any perceived “victory” that the majority earns is a loss for the party that wants to win control of the legislature. The minority party therefore often defaults to opposing everything that the majority party suggests, even in areas where policymakers may otherwise have found common ground. In other words, the current U.S. electoral system, in this governing context, is a recipe for partisan animosity, vitriol, and gridlock.

Yet it is possible to take this critique too far. The fact that American government is currently stuck in a state of high conflict and low productivity does not mean, necessarily, that nothing is achieved, or that there is no democratic accountability, or that a democratic crisis is inevitable.

From time to time, even an embittered, closely divided Congress is able to get things done. As a number of commentators have pointed out, 2022 was an unusually productive legislative year; but it was also in some ways an exception that proves the rule. Much of that productivity was only possible because the same party controlled both houses of Congress and the presidency, ¹¹ a rare circumstance in recent history and one that ended in 2023. A number of legislative conflicts were resolved through bizarre procedural maneuvers: Although the minority party tacitly allowed majority proposals to pass, it did so in ways that allowed its members to publicly disavow and oppose them.¹² And while congressional members reached across the aisle to pass some serious bipartisan legislation, a number of the key members who bucked their party were retiring or otherwise not up for reelection.¹³

Members of Congress should certainly try to keep this streak going. The likelihood, however, is that electoral rules will incentivize a return to the default mode of conflict and dysfunction. Unless the rules change, then at least in the medium term, American politics can be expected to continue along its current path: two parties, a substantial distance apart, locked in an indefinite conflict and only sporadically able to get things done.

Americans, even the relative few who work in and around politics, tend to take the electoral system for granted. Partisan primaries, single-member districts, and winner-take-all elections have been the standard for a long time. It is doubtful that many Americans have much awareness of the alternatives.

The fact is, though, that many alternative systems exist for choosing representatives, and they can be found not only in dozens of other democracies worldwide but also in a small but growing number of U.S. states and municipalities.¹⁴

Moreover, there is a lot of potential for change. Few of the elements of the U.S. electoral system are enshrined in the Constitution. The most important exception is the requirement that each state must elect two senators. While the Constitution originally tasked legislatures with this choice, the passage of the 17th Amendment required that senators be “elected by the people”¹⁵—and members of the U.S. House of Representatives must also be chosen “by the People.”¹⁶ But the method through which the people must elect members of Congress is almost entirely unspecified.

The final sections of this report describe a range of different ways that states and localities, as well as Congress, could reform their electoral systems, with the hope of creating a more positive political dynamic. Some of these changes would be relatively minor; others would be substantial. This report does not advocate for any one particular change; rather, it advocates for more consideration of and experimentation with a variety of electoral reforms that could improve the political status quo.

Ultimately, any reform is worthy of consideration if it advances at least one of two criteria:

1. Improves incentives for problem-solving and collaboration: Any electoral reform should at least maintain, if not improve, the motivation for legislators to govern collectively and responsibly. An ideal legislature would contain people with diverse points of view; after all, most people in any given community have diverse points of view. But it would also encourage legislators to work with their colleagues in good faith to get things done.
2. Improves representation: Any electoral reform should at least maintain, if not improve, how well the public is represented in local, state, or federal legislatures. The bar is low: Due in part to gerrymandering, parties that won fewer votes in some U.S. states have been able to hold majorities in the legislature,¹⁷ failing a basic test of democracy. More broadly, voters feel that their system of government needs major reform¹⁸ and that most members of Congress do not deserve to be reelected.¹⁹ Electoral reforms should aspire to provide more meaningful choices to voters and greater alignment between the public and their representatives.

People who write about and advocate for electoral reform have plenty of disagreements about which reforms have the most potential to improve American politics. They also engage in vigorous debates on matters of strategy, including whether more modest reforms are a gateway to more ambitious reforms or a recipe for disappointment and backsliding. They must also consider the tricky matter of advocating for reform without overpromising. Electoral reform is an important, overlooked piece of the democratic puzzle, but it is not the only thing that matters. Any given reform may not produce immediate results.

All that said, both activists and ordinary citizens almost universally agree that the political status quo is in need of change. Reformers should be willing to consider whatever reforms are politically achievable that move the ball forward.

This section briefly describes a variety of possible electoral reforms that show at least some potential to improve incentives for governance or representation. They are presented roughly in order of how significantly they would change the current electoral rules, from the most minor changes to the most substantial.

Fusion voting

Fusion voting is the practice of allowing more than one political party to nominate the same candidate. That candidate then appears multiple times on the ballot, under each different party label. Although it is a relatively minor change—one that a handful of states, including New York, already use²⁰—it creates more opportunity for third parties to influence elections and develop distinctive brands. The hope is that allowing voters to express more nuanced views on their ballot could help break up the rigidity of two-party conflict and encourage candidates to build broader coalitions. Some advocates have argued that fusion voting could be a step on the path to bigger reforms that encourage or require more than two political parties.²¹ Others argue that the impact of fusion voting is relatively limited.²²

• Status: Although common in the United States in the 19th century, fusion voting is now banned in 42 states, partly a result of the major parties’ efforts to consolidate power.²³ However, there is currently renewed interest in the practice, including an effort to overturn a ban on fusion voting in New Jersey, a possible precursor to broader nationwide efforts.²⁴

Ranked-choice voting

Ranked-choice voting (RCV) is a system where, rather than selecting a single candidate, voters can rank multiple candidates in order of preference—first, second, third, and so forth. After the votes are counted, the candidate who received the fewest first-choice votes is eliminated and their supporters’ votes are reallocated to the supporters’ second choice. This process repeats itself until a single candidate has received a majority of the votes.²⁵

The hope of RCV advocates is that it upends the first-past-the-post system, eliminating spoilers, creating more space for additional candidates, and potentially generating less animosity in campaigning and governing—because candidates would have an incentive to win the second- and third-choice support of the voters who ranked their rivals first. RCV does not inherently eliminate partisan primaries, and if voters continue to make partisan candidates their first choice, then it may not have much impact on outcomes.²⁶ But many advocates of RCV support pairing the reform with either nonpartisan, multicandidate primaries or with multimember districts. (discussed below)

• Status: Currently, in the United States, two states and 60 localities use some form of RCV—in many cases paired with other electoral reforms.²⁷ The state of Maine, for example, has elected to use ranked-choice voting for all federal elections. Rep. Jared Golden (D-ME) has twice won election to the U.S. House from Maine’s 2nd Congressional District only after receiving second-choice votes from voters who initially supported an independent candidate.²⁸ Efforts to promote RCV have gained steam in recent years, with one state and eight localities voting to adopt RCV in the 2022 elections alone.²⁹

Nonpartisan, multicandidate primaries (in combination with RCV)

Another approach to electoral reform is to eliminate partisan primaries. More than a decade ago, California adopted a reform known as the “top-two” primary: Instead of having separate primaries for each party, California now has one primary for candidates of all parties, as well as independents, and the top-two vote-getters advance to the general election. Although top-two has its proponents,³⁰ it is widely perceived to have had a much more limited impact than its advocates had hoped for,³¹ and at times, it has led to dubious outcomes when multiple candidates split the votes.³²

However, a version of primary election reform that has more recently been embraced by advocates combines a top-four or top-five primary with ranked-choice voting. The idea behind this system—which the state of Alaska adopted in 2020—is that, like top-two, it eliminates partisan gatekeepers, but it also gives general election voters a real diversity of choices, since multiple candidates advance to the general election and since RCV allows voters to pick their genuine first choice without fear that their vote will be wasted.

• Status: Alaska just completed its first election under its new system, a top-four primary combined with RCV. It is still early to judge how the system will affect future elections, but it does seem to have ushered in a number of moderate candidates who align well with Alaska voters and who may have lost in a traditional partisan primary.³³ Voters in Nevada also recently approved a top-five system; according to that state’s law, the system will have to be approved again in 2024 in order to go into effect in 2026.³⁴

Multimember districts (in combination with RCV)

One limitation of all the reforms discussed so far is that they maintain single-member districts—which means that some voters will still cast only protest votes, that the barriers to entry for third parties and independents will still be significant, and that partisan gerrymandering may persist. An even further-reaching reform that has growing support is the idea of creating larger districts and allowing each district to elect multiple representatives. This system has been written into a draft federal bill, the Fair Representation Act (FRA),³⁵ which would divide each state with multiple representatives into one or more multimember districts. In turn, each of those districts would elect between three and five members to the U.S. House, chosen through RCV. So, for example, if a third of the voters in one of these large districts supported a Republican, a third supported a Democrat, and a third supported an independent or third-party candidate, all three could be elected—a diverse district with diverse representation.

The use of ranked-choice voting makes it much more likely that voters, even minority voters in red or blue states, could help elect at least one like-minded candidate. And the candidates themselves would have at least some incentive to campaign and govern cooperatively in order to earn the second- and third-choice support of voters who ranked their opponents first. Among reform-minded academics, “multi-winner ranked choice voting”³⁶ is perhaps the electoral reform that has attracted the most positive attention.³⁷

• Status: Portland, Oregon, recently became the first major U.S. city to adopt a multimember, ranked-choice system for its city council elections,³⁸ a change that advocates will be watching closely and that may inspire other state and local reform efforts. The multimember RCV system, known internationally as “single transferable vote,” is also used in several jurisdictions outside the United States, including Ireland and Australia.³⁹

Open-list proportional representation (OLPR)

There is another way to eliminate single-member districts that, in some ways, is more straightforward but might be perceived as a more significant change to America’s political status quo. OLPR is a common system in other advanced democracies in which all the voters in a given jurisdiction vote for a candidate and also have their vote count for the candidate’s political party. Each political party gets seats in the legislature in proportion to the votes they earn; so, if a hypothetical Orange Party wins 20 percent of the votes, it would get roughly 20 percent of the seats. But voters also have a role in determining which Orange Party candidates get elected. The order of the candidates is determined by how many votes they have received; if the Orange Party gets three seats, those seats would go to the three Orange Party candidates who received the most votes.

In an OLPR system in the United States, there could be no real red or blue states, as every vote cast would help determine how red or blue—or something else—the legislature would be. And it is much easier, in these systems, for additional parties to develop and win seats because they can get a few seats in the legislature with a smaller number of votes cast over a wider area.

One downside to this system is that it is relatively unfamiliar in the United States. Voters would have to become comfortable with the idea that support for political parties plays a bigger role in representation and that some fringe political parties could get a foothold in legislatures.

• Status: Although OLPR is widely used overseas, in 40 different countries according to one count,⁴⁰ it is not currently in use in any jurisdiction in the United States. However, several proponents have argued for it as a possible alternative to multimember districts with RCV.⁴¹

Other systems, including other forms of proportional representation

It is worth at least keeping in mind that there are many possible variations of the ideas above, including numerous other forms of proportional representation used by democracies globally. For example, another system that is well-regarded by scholars but not used in the United States⁴² is mixed-member proportional representation, which combines first-past-the-post single-member districts and proportional multimember districts.⁴³ Another common system is closed-list proportional representation, in which voters only cast a vote for a political party and candidates are then elected from the party’s predetermined lists. Some systems and features, such as the closed party list, probably do not make sense in an American context. But advocates should be aware that there are many different viable ways to design an electoral system,⁴⁴ and there may be good ideas that have not yet entered the public debate.

Though electoral reform has long been a niche issue, it is finally starting to attract serious attention. The New York Times recently reported on an open letter by more than 200 political scientists calling on Congress to adopt multimember districts.⁴⁵ Ranked-choice voting and other reforms have attracted interest from across the political spectrum, including from Cato Institute libertarians,⁴⁶ prominent conservative columnists,⁴⁷ left-of-center think tanks,⁴⁸ and many others in between. Specifically, Alaska’s adoption of top-four primaries and RCV—a major electoral change that affects an entire U.S. state—may have softened doubts about what kind of changes are politically possible.

The organizations and advocates who made this happen deserve an enormous amount of credit. Many of them have spent years working on electoral reform with extremely long odds and very limited support; but their persistence has now put these issues on the verge of becoming mainstream.

Nonetheless, there is a lot more work ahead. Political incumbents—whether elected politicians, political parties, or allied interest groups—tend to resist changes to the system that put them in power. Fortunately, now more than ever, many of these incumbents see the current political status quo as alarming, even untenable. But many more will have to be persuaded that reform can coexist with their personal and political interests.

Moreover, most Americans have never experienced anything outside the long-standing, partisan, winner-take-all political system. They may dislike the way that politics is currently working, but it is going to take significant and sustained education and advocacy to help Americans imagine how things could be different—and to figure out what version of electoral reform would best address their concerns.

To meet these challenges, it is incumbent on those who care about democracy—organizations, advocates, funders, and commentators—to make electoral reform a bigger part of their collective work. It is increasingly clear that electoral incentives are a big part of what is driving the dysfunction in American politics. A serious effort to addressing that dysfunction requires serious attention to electoral reform.

The author would like to thank Lee Drutman, Kristin Eberhard, Ben Olinsky, Walter Olson, Will Roberts, and Chris Tausanovitch for their thoughtful feedback.