Rose Institute Q&A: CA’s Top-Two Election System (part 4)



Assessing the Impact on Politics

Why is it so hard to get simple answers about the effects of the top-two?

It can be frustrating for citizens, journalists, and advocates to have political scientists provide different assessments about the effects of the top-two reform.  It is important to understand why these disagreements exist.

First: it is hard to isolate the effects of the top-two system because it was introduced at the same time as other major changes.

At the dawn of the new century, the annual failure to pass a budget on time amid partisan bickering in Sacramento caused a serious governance crisis.  Californians, disapproving of elected officials and despairing of the condition of state politics, adopted several reforms.

In 2010, voters passed Proposition 14 to establish the top-two election system.  They also adopted a citizens’ redistricting process (2008), a simple majority budget requirement (2010), and new types of term limits (2012).  Californians also passed new taxes on a ballot measure in 2012, supported by Governor Jerry Brown, which helped alleviate the state’s chronic budget shortfalls.[1]  Moreover, the state became more solidly Democratic during this period.  These overlapping events make it quite challenging to evaluate what any particular reform accomplished.

The difficulty in assessing what any single reform has accomplished should not obscure what has happened.  Approval of the state legislature on PPIC surveys went from 9% to 49% among likely voters in 2010 and 2020.[2]  By 2016, the Los Angeles Times noted the on-time budgets and declared “fiscal gridlock” to be “a thing of the past.”[3]  Given a chance to recall the governor in 2021, Californians instead retained Gavin Newsom in office by a large margin.  California is projected to have a budget surplus in 2022 of nearly $100 billion instead of the yearly shortfall.  This is not to say the state does not face real and serious policy challenges: of course, most voters have ongoing concerns.  Nevertheless, the combination of budget shortfalls, policymaking gridlock in Sacramento, and voter frustration with elected officials have been diminished as the central theme of California politics.

Did the top-two make the difference?  Redistricting?  Simple-majority budgets?  Good fortune?  All of them together?  Figuring out the effect of any given reform is difficult.

Second: political science is a collective and ongoing endeavor.  Scholars try a variety of approaches to answer questions and continue to develop new methodologies.  Each election cycle has produced more data generated with varying external circumstances.  Research done in 2012 would not capture the unusually low turnout year of 2014; research using 2012 and 2014 misses the increases in participation in 2016-2020.  The entire 2012-2018 period of the top-two system took place in the very large shadow of the second governorship of Jerry Brown.  Studies using different methodologies and conducted at different times may produce different findings without any of them being “wrong.”  The variety of data, methods, and results helps scholars collectively build knowledge about how politics works.


What do voters think of these kinds of election systems?

Generally speaking, voters seem to like the top-two system and open primaries more broadly.  Preference over systems does depend to some extent on partisanship, ideology, and worldview.

In a national survey of attitudes towards election rules in 2020, voters most closely aligned with President Trump were the least likely to support the nonpartisan top-two procedure.[4]  This corresponds with earlier polling data in California from 2017 in the PPIC statewide survey: with a few years of experience using the system, 60% of Californians described Proposition 14 as “mostly a good thing” while only 26% thought it “mostly a bad thing” (the rest did not know or gave a mixed answer).  There is clearly a partisan trend in those responses as well, with 71% of Democrats in California considering the top-two rules “a good thing” as compared to 45% of Republicans.[5]  Nevertheless, even though support for the top-two is weaker among Republicans and supporters of former President Trump, overall levels of support are substantial among all types of voters.


Did the top-two produce more moderate or collaborative legislators? 

The original proponents of Proposition 14 promised the system would produce representatives who were “LESS PARTISAN” and “MORE PRACTICAL” (emphasis in the original description).[6]

Scholars have attempted to evaluate this claim, although they have reached diverse conclusions.  In short: some scholars have found little or limited evidence of moderation[7] while others have found some evidence of moderation.[8]  (For additional information, see: “What does moderation mean?”)  Recent work by political scientist Christian Grose finds that not only top-two nonpartisan primary systems are associated with greater moderation, but that open partisan primary systems are, as well.[9]

Skepticism of the moderating effect of the top-two system is rooted in scholarship focused on the ability of parties to bolster their preferred candidates before elections even begin,[10] the weakness of evidence for effects of different degrees of openness among partisan primaries,[11] polarization in the nonpartisan legislatures in Nebraska (and, for the time it was used, Minnesota),[12] and the potential to use endorsements to coordinate party voters.[13]

Aside from the findings directly about moderation in the top-two, reasons for optimism for a moderating effect are grounded in the scholarly literature as well.  The “blanket primary” (ruled unconstitutional in California Democratic Party v. Jones) in some sense provided weaker incentives for moderation than the top-two, but was associated with generating compromise in the legislature.[14]  The mechanism of same-party elections seems like a plausible route for electing more moderate or collaborative legislators;[15] voters seek information about candidates in these same-party general election contests,[16] candidates try to adapt their campaigns to that electorate,[17] and the efficiency of their campaign spending suggests that they can also persuade voters to support them.[18]

It is possible that the power of parties is pushing up against the incentives implied by the structure of the competition, and the outcome depends on the balance of these forces and the presence of outside factors.  A very interesting study (discussed earlier) from political scientist Jesse Crosson suggests that the top-two system works through same-party elections to help advantage moderate candidates, but just not often enough to moderate the legislature.[19]

It is safest to say, at this point, that scholars have not reached a consensus on the moderating impact of the top-two election system, despite valuable research in this area.  It is not particularly surprising that the scholarship has presented a variety of conclusions: researchers are studying different facets of moderation, using different types of data and methods, using data from different elections, and even sometimes including different states.[20]


What other impacts might there be beyond moderation?   

The focus on moderation reflects the campaign promises of Proposition 14, and is certainly relevant for many of the most important debates in American politics today; nevertheless, it is not the only potential consequence of changing the election rules, nor did every voter supporting Proposition 14 have the same objectives or expectations for the reform.

For example, voters could reasonably hope for an increase in the competence of elected officials.  Even if the ideology of legislators and executive officeholders remains largely constant, state government can produce better-designed programs, greater efficiency, decreased corruption, improved administrative oversight, and more effective representation.  One of Hirano and Snyder’s main findings in Primary Elections in the United States is that primaries (broadly) help select more qualified candidates and encourage voters to hold them accountable for performance while in office.[21]  While these authors did not extend that claim to the top-two system in particular, these are types of outcomes that stand apart from ideological moderation.[22]

It is worth observing, moreover, that some democratic goods voters seek, such as competence in their elected officials, descriptive representation, or performance accountability, might work at cross-purposes with achieving ideological moderation.  Consider a Republican voter facing a choice, from his or her perspective, of Democrat A (more moderate, and possibly less competent) or Democrat B (more liberal, and possibly more competent).  If the Republican voter preferred the more competent candidate despite that candidate’s liberalism, is that necessarily a failure of the top-two system?

We also know that legislators do a great deal beyond simply occupying a location on the ideological spectrum.  Some could be more effective at constituent services, more responsive, or better able to provide local political benefits.[23]  In addition, research suggests that voters seek descriptive representation in same-party general elections (on the basis of race, ethnicity, and gender) when party cues are not available.[24]  Even two candidates of roughly similar ideological orientations may choose to focus on different issues, affecting everyone working in those policy areas.[25]  Politics is a very messy business with a host of inputs and a host of outputs; as additional research is conducted, we should gain a clearer insight into the main question (moderation or not) as well as more exploration of these other potential consequences.  Even if these effects are not what the campaign for Proposition 14 promised, they may be outcomes that California voters like and are willing to sustain.


Did the top-two system make the Republican Party uncompetitive in California?

In brief, no.  Other forces besides the top-two system are responsible for the Republican Party’s weakness in California.

Political scientist Kenneth P. Miller has identified a range of factors that have contributed to the Republican Party’s decline in California, such as the state’s underlying progressive political culture, its demographic and economic changes, and the national realignment of the two parties. In combination, these and related factors caused the state’s partisan balance to tip decisively in favor of Democrats near the turn of the twenty-first century—before the introduction of the top-two system.[26]

By 2007, Republican Governor Arnold Schwarzenegger (the last Republican to be elected statewide) told the GOP state convention that the party was “dying at the box office.”[27]  In 2008, Barack Obama, the Democratic candidate for president, won more than 60% of the statewide vote—a pattern that would be cemented in subsequent presidential elections.

In 2010, the last year with the prior partisan primary, and nationally a very good year for Republicans, the California Republican Party lost every statewide office.  Democrats won the majority of the U.S. House seats, California Assembly seats, and had a comfortable majority in the State Senate as well.

As of yet, the top-two election system has not helped Republicans reverse this trend, but it is doubtful that it has accelerated the party’s decline, either.  Whether by party registration, presidential voting patterns, publicly available survey data, or the outcomes of elections covered by Proposition 14, it is clear that most Californians are disinclined to favor Republican candidates.

In 2014, in the first statewide election using the top-two system, Republicans were equally as unsuccessful in the contests for statewide office as they had been in 2010 with the old partisan system.  In 2010, despite vast expenditures and the “Tea Party” wave, Republican gubernatorial candidate Meg Whitman won only 40.9% of the general election vote against Jerry Brown.  In the lieutenant governor’s race, Abel Maldonado–who, along with Governor Schwarzenegger, was one of the architects of the top-two system—won only 39% of the vote against Gavin Newsom.  Fast forward to 2014: Republican gubernatorial candidate Neel Kashkari won 40% of the vote against Jerry Brown and the Republican candidate for lieutenant governor, Ron Nehring, won just under 43% of the vote against Gavin Newsom.  Whitman, Maldonado, Kashkari, and Nehring are fairly different candidates – but they all carried the “R” label into the election in a state continuing to trend toward Democratic dominance.

The top-two election system does, though, provide a way to have competitive elections even as one party continues to decline.  Competitive states, and districts within them, are rare.  For California, and other states interested in such rules, nonpartisan election procedures are one potential way of continuing to have some kind of meaningful democracy even when the vast majority of a state’s citizens prefer one party.


[1] Ben Adler, “The Make-Or-Break Moment Of Jerry Brown’s Second Governorship,” January 3, 2019,

[2] “PPIC Statewide Survey: Californians and Their Government – March 2022: Likely Voters – Time Trends,” March 2022,  This is not just an artifact of the pandemic, either; legislative approval was relatively steady in 2019-2022.

[3] John Myers, “Remember When California’s Budget Was Always Late? Here’s Why Fiscal Gridlock Is a Thing of the Past,” Los Angeles Times, June 18, 2016,

[4] Sinclair and Sinclair, “Primaries and Populism.”

[5] “PPIC Statewide Survey: Californians and Their Government, December 2017” (Public Policy Institute of California, December 2017),

[6] Bowen, “Official Voter Information Guide: California Statewide Direct Primary Election, Tuesday, June 8, 2010”; see also: McGhee, “Open Primaries.”

[7] Eric McGhee, “California’s Top Two Primary and the Business Agenda,” California Journal of Politics and Policy 7, no. 1 (February 5, 2015),; Eric McGhee and Boris Shor, “Has the Top Two Primary Elected More Moderates?,” Perspectives on Politics 15, no. 4 (December 2017): 1053–66,; Thad Kousser, Justin Phillips, and Boris Shor, “Reform and Representation: A New Method Applied to Recent Electoral Changes,” Political Science Research and Methods 6, no. 4 (October 2018): 809–27,

[8] Christian R. Grose, “Political Reforms in California Are Associated with Less Ideologically Extreme State Legislators” (University of Southern California Schwarzenegger Institute, March 2016); Steven Sparks, “Polarization and the Top-Two Primary: Moderating Candidate Rhetoric in One-Party Contests,” Political Communication 36, no. 4 (October 2, 2019): 565–85,; Christian R. Grose, “Reducing Legislative Polarization: Top-Two and Open Primaries Are Associated with More Moderate Legislators,” Journal of Political Institutions and Political Economy 1, no. 2 (June 10, 2020): 267–87,

[9] Grose, “Reducing Legislative Polarization.”

[10] Marty Cohen et al., The Party Decides: Presidential Nominations Before and After Reform, Chicago Studies in American Politics (Chicago: University of Chicago Press, 2008); Hassell, The Party’s Primary.

[11] McGhee et al., “A Primary Cause of Partisanship?”

[12] Masket and Shor, “Polarization without Parties”; Masket, The Inevitable Party.

[13] For an evaluation of the value of endorsements, see Kousser et al., “Kingmakers or Cheerleaders?”

[14] R. Michael Alvarez and Betsy Sinclair, “Electoral Institutions and Legislative Behavior: The Effects of Primary Processes,” Political Research Quarterly 65, no. 3 (2012): 544–57.

[15] J. Andrew Sinclair, “Winning from the Center: Frank Bigelow and California’s Nonpartisan Primary,” California Journal of Politics and Policy 7, no. 1 (2015),

[16] Sinclair and Wray, “Googling the Top Two.”

[17] Sparks, “Polarization and the Top-Two Primary.”

[18] Steven Sparks, “Campaign Spending and the Top-Two Primary: How Challengers Earn More Votes per Dollar in One-Party Contests,” Electoral Studies 54 (August 1, 2018): 56–65,

[19] Crosson, “Extreme Districts, Moderate Winners.”

[20] E.g., are we studying California alone, the closely related California and Washington rules, or all (structurally) nonpartisan rules?  There are good reasons to make all of these different choices as a researcher; these choices, though, influence the results and can help explain some of the divergence.

[21] Hirano and Snyder, Primary Elections in the United States.

[22] “There is some evidence from California and Washington that the top-two system has the potential to elect more moderate candidates.  However, since it is such a recent reform in these two states and has only been used in one other state [for Congress], we cannot yet draw firm conclusions.”  Hirano and Snyder, 306.

[23] Not all the electoral outputs in the classics of legislative behavior are about ideology.  See David R. Mayhew, “Congressional Elections: The Case of the Vanishing Marginals,” Polity 6, no. 3 (1974): 295–317,

[24] Sadhwani and Mendez, “Candidate Ethnicity and Latino Voting in Co-Partisan Elections”; Sara Sadhwani, “Asian American Mobilization: The Effect of Candidates and Districts on Asian American Voting Behavior,” Political Behavior, May 6, 2020,; Stauffer and Fisk, “Are You My Candidate?”; Sara Sadhwani, “The Influence of Candidate Race and Ethnicity: The Case of Asian Americans,” Politics, Groups, and Identities, February 1, 2021, 1–37,

[25] An argument made in Sinclair et al., “Crashing the Party.”

[26] Kenneth P. Miller, Texas vs. California: A History of Their Struggle for the Future of America (New York, NY: Oxford University Press, 2020), 121–22.

[27] Evan Halper and Scott Martelle, “State GOP Is ‘dying at the Box Office,’ Gov. Says,” Los Angeles Times, September 8, 2007,