The United States has a ‘primary problem.’ As described by Unite America, the 2026 general election is months away, but 93 percent of U.S. House and 80 percent of Senate races are already decided. Because many districts and states are safe for one party, low-turnout primaries often determine who represents a community, giving a small share of voters disproportionate influence.

Primary election reform is billed by advocates as a solution to the primary problem and the concerns stemming from the oversized influence of primaries in American politics. While reform broadens participation, its effects are likely to be small and conditional unless it is accompanied by greater mobilization of moderate voters.

The Problem

The primary problem stems from the combination of safe seats and low-turnout primaries. Turnout in primary elections tends to be less than half that in general elections, often around 20 percent of eligible voters in congressional primaries.

Legislators have incentives to respond to the interests of primary voters rather than those of general election voters. Because primary voters tend to be more unified in their issue positions than the general electorate, and because primary voters tend to put more weight on issue positions when voting (since party is not an easy cue), legislators have electoral incentives to align with primary voters on issues where the primary and general electorates disagree.

Primaries might also contribute to polarization, with downstream consequences for partisan conflict and gridlock. While the empirical evidence linking primaries to polarization is mixed, there is some evidence that primary voters hold distinct views relative to the general electorate and use primaries to select aligned candidates.

When primary voters are not representative of the general electorate (or even all co-partisans), primaries skew representation and can reinforce party loyalty. Primary losses such as those in the 2026 Republican primary by Thomas Massie in Kentucky, Bill Cassidy in Louisiana, and John Cornyn in Texas illustrate these dynamics.

Is primary reform the solution?

Advocates of primary reform focus on two types of reforms. The first is a shift from closed to open primaries. In partisan primaries, voters choose among candidates within a party. In closed primaries, voters must be registered party members. In open primaries, voters can pick either party’s ballot in the primary, regardless of their partisanship. Variants like ‘open to unaffiliated’ allow unaffiliated voters to participate in either the Democratic or Republican primary while requiring that partisans vote only in their own party primary.

Advocates of open primaries focus on giving voice to unaffiliated voters (a large and growing bloc of citizens in many states), giving elected officials incentives to listen to their broader electorate, and reducing polarization. However, evidence suggests that turnout increases are modest – only a few percentage points – and primary electorates remain dominated by partisan voters. The result is that polarization is similar across primary types. Primaries are low-information elections, and the partisan base may have stronger incentives to participate. However, as the bloc of unaffiliated voters grows, robust participation from these voters could have a meaningful impact on primaries.

The second type of primary reform is a shift from partisan to non-partisan primaries. Non-partisan primaries narrow the field of candidates before the general election but do not nominate candidates for each party. Currently, California, Washington, Alaska, Louisiana, and Nebraska have some form of non-partisan primaries for state elections. All candidates are on the same ballot (regardless of party) and all voters can participate. As a result, the gap in turnout between the primary and general tends to be smaller in non-partisan primaries and both partisans and independents have a greater likelihood of participating in non-partisan primaries than closed partisan primaries. Advocates suggest that non-partisan primaries allow all voters to participate in meaningful elections and may reduce polarization. Empirical patterns provide some suggestive evidence that the rate of increasing polarization slowed in California after the adoption of the top-two primary system and that newly elected members of Congress from states with non-partisan primaries are less extreme than members elected in partisan primaries.

Ironically, non-partisan primaries might have the biggest impact on reducing polarization in places that are very lopsided in terms of political competition. When voters of a statewide or district minority party know that they have no chance of electing a candidate from their own party, they can vote for a more moderate candidate from the opposing party, ensuring that candidate makes it to the general election. In the general election, the more moderate candidate is likely to win. In top-two or top-four primaries this might mean that the primary yields two candidates from the same party, with the more moderate candidate winning the general election. For instance, in the 2022 Alaska primary, a large fraction of Democrats reported voting for Lisa Murkowski in the top-four primary. Murkowski – a more moderate Republican – would not have been likely to win a partisan primary against Kelly Tshibaka, who received support from a majority of Republican voters.

Non-partisan primaries are not without their flaws. They winnow the number of candidates but are not a means of nominating candidates within the parties. This can lead to some perverse outcomes in terms of representation. Since California adopted the top-two system in 2010, a small number of general election matchups featured two candidates from the same party in districts that favored the other side. Many Democrats worried about this possibility in the 2026 California gubernatorial primary where a splintered Democratic field and greater Republican consolidation could have allowed two Republicans to advance despite statewide voter registration favoring Democrats by a wide margin. Non-partisan primaries (and even fully open partisan primaries) also allow people outside of the party to pick the party’s candidate. Some advocates for stronger parties have argued that democratic politics works best when parties can choose their own candidates – selecting those who align with the values of the party and who are best positioned to help the party win in November. Of course, the counterargument is that the party establishment hasn’t done a great job protecting the party brand in recent years and voters who participate in primaries may not be thinking about electability or the importance of compromise in our governing institutions.

Ultimately, primary reform addresses the symptoms rather than the causes of the primary problem. Safe seats make primaries decisive and give primary voters leeway to select ideologically extreme and loyal partisans without much consequence in the general election. More competitive districts would enhance accountability, in part by reducing the importance of the primary electorate. Competition would also encourage voters to consider their party’s chances of winning in the general election – what we call ‘electability’ – which may lead them to pick more moderate candidates. Party elites are more likely to coordinate around a candidate in the primary in competitive districts. If the party is better off in November with a more moderate candidate that appeals to swing voters, electability considerations and coordination might work to the advantage of relatively moderate candidates.

Primary reform broadens participation, but its effects on representation and polarization depend on whether independents and moderates participate in large numbers. Without greater engagement from those voters, reform is unlikely to substantially change electoral outcomes.