INSTITUTE FOR POLICY RESEARCH

Do U.S. Elections Have a Primary Problem?

IPR’s Laurel Harbridge-Yong explains why primary voters wield so much power and how they might skew representation

New York City’s mayoral election is set to take place on Nov. 4. Yet some are already hailing the winner of June’s Democratic primary run-off, Zohran Mamdani, as the city’s next mayor.

That’s not surprising to IPR political scientist Laurel Harbridge-Yong. Her research finds that primary elections can play an outsize role in determining which candidates are ultimately elected to office. 

Why are primaries so important for understanding who wins and whose interests are heard in government? One political party often dominates districts and states, so their primary races are more competitive than general elections. Harbridge-Yong’s research also points to a growing concern for U.S. democracy—that the much smaller groups of voters who turn out for primaries are likely skewing political representation in their favor. This is true in both safely partisan and competitive states.

“To the extent that this is a kind of fringe group of individuals who don’t represent the broader population, we might wonder about who they’re choosing for office, and whether those people ultimately represent the values and preferences and political style of the broader population,” Harbridge-Yong said.

As questions rise about the power of primary voters, Harbridge-Yong is studying how primary elections impact our electoral process and whether they are creating conflict in U.S. politics.

Are Primary Elections a Problem for Democracy?

When the primary electorate disagrees with the general electorate, legislators have incentives to appeal to primary voters. Harbridge-Yong, who has studied these incentives, says that this is problematic because legislators may be tailoring their platforms and messages to the views of a small number of voters over those of the larger constituency.
 
“If primary elections are making a really big difference to who is elected in office, either by determining what the choice is in the general or determining the winner in many of these safe states, it matters whose voices are being heard in this primary election,” Harbridge-Yong said.

This dynamic not only raises concerns about fair representation, but it also can eventually increase polarization and gridlock in Congress. Harbridge-Yong and her co-authors argue that legislators align with primary voters because they are often more unified in their policy preferences and care more about how their representative votes on those issues. 

“In a primary electorate, in most states, legislators are campaigning against people from their own party, so party ID is not a signal or a cue to voters, so voters instead need to rely on other things, such as what the legislators’ roll call voting was or what their issue positions are,” she said. “These positions are going to be more impactful in terms of how people vote.”

Harbridge-Yong’s surveys in California, Michigan, and Nevada in 2024 also show some important differences between likely primary voters and the broader population of the state. Compared to the general population, primary voters are more ideologically extreme and more likely to see the other party as a threat to the U.S. They also have higher affective polarization, which refers to the gap between how much they like their own party and how much they dislike the other party.

When she compared likely primary voters to likely general election voters, she found the same results, except that primary voters are no more prone to see the other side as a threat than general election voters. One reassuring finding, she says, is that primary voters were not more supportive of undemocratic practices than the average American or general election voters.

 Can Election Reforms Make Primaries Fairer?

To bring in a more representative group of voters earlier in races, some states have turned to nonpartisan primary elections, or elections where all candidates are listed on the ballot and voters can pick regardless of the candidate's party affiliation.

Harbridge-Yong studied these election reforms in Alaska and California. While the California results are still preliminary, her research looking at Alaska’s 2022 top-four primary shows that the election reforms likely benefited moderate candidates and encouraged more nonpartisan voters to participate in 2022 versus in 2020. 

Across statewide and state legislative races, they estimate that at least six moderate candidates who would have lost in the old system emerged from the top-four primary and won the general election under the new system. In the race for U.S. senator, crossover voting—voting in another party’s primary—allowed voters from the other party to support the moderate candidate. This helped Republican Lisa Murkowski perform better than Democratic candidates and her co-partisan rival Kelly Tshibaka, who was endorsed by President Trump.

“It actually may be that these lopsided districts or states where one party is pretty dominant and the other party knows they’re the district-wide or statewide minority—that’s where the nonpartisan primary might have the most bang for your buck,” Harbridge-Yong explained.

The votes of Democrats, who are in the minority in Alaska, along with those of nonpartisan voters, tipped the election in favor of Murkowski and other moderate candidates, who went on to win in the general election.

Harbridge-Yong says nonpartisan elections have the potential to reduce polarization because voters in the minority party who know that their preferred candidate won’t win might support a moderate candidate from the other party to keep someone more extreme from getting elected.

While she argues that moderate candidates are not necessarily better than extreme candidates, the U.S. population as a whole tends to hold more moderate positions than many candidates who are elected to office. Harbridge-Yong suggests that any election reforms focus on improving political representation for the general population.

How Do Voters Think About Electability During Primaries?
 
A distinct feature of primary elections is that voters know that they will have the chance to vote again in the second stage of the election, which can cause some people to vote strategically during primaries to try to pick the most electable candidate. Harbridge-Yong’s research shows that primary voters use information to make strategic choices about who is most likely to win later on.

One piece of information that voters use as a key signal of electability is a candidate’s ability to fundraise, according to Harbridge-Yong’s research. Her work suggests that this is especially true of Republicans.

Another forthcoming study shows that primary voters notice the difference between financial contributions candidates receive and self-funding by wealthy candidates. These voters believe that the ability to raise money makes candidates more electable.

Harbridge-Yong and her co-authors also see a connection between affective polarization and how people vote during primaries, including how they think about being strategic with their vote. When they surveyed people across California, Michigan, and Nevada, they discovered that people who scored higher on affective polarization were more likely to choose the candidate from their own party whom they perceived as more electable over the candidate who is closer ideologically to themselves. 

“I think that makes some theoretical and intuitive sense,” she said. “If you really dislike the other party, you’re willing to take your second-best person from your own party rather than risk the other side winning in a general election.”

But they found the opposite pattern between affective polarization and strategic voting in crossover voting. Participants in California’s top-two primary who scored higher on affective polarization were less likely to engage in crossover voting in a nonpartisan primary.

“If you are a Republican who really dislikes the Democratic Party, even a moderate Democrat is someone who you’re not going to want in office,” Harbridge-Yong said.  “You’d rather take your chances with someone from your own party winning.”

Laurel Harbridge-Yong is professor of political science and IPR associate director and a fellow. 

Photo credit: iStock

Published: August 28, 2025.