[Dear DemocracySOS readers, this political newsletter only survives due to the support of people like you. Please consider becoming a paying subscriber to support this work. Or, if you are already a paying subscriber, consider giving a $5 gift subscription to a friend, colleague or family member. Thank you!]
The hot political gossip in California recently has been watching a potential train wreck in our Top Two primary for governor. Too many Democratic candidates could split the majority Democratic vote and result in two Republican candidates and no Democrats facing off in November. The chances of this happening have receded in recent weeks, but elections can be surprising, and there are a lot of undecided voters.
This episode has given Californians a chance to reflect on whether the Top Two primary is really the best way to elect the Golden State’s political leaders. A recent study by CalRCV shows a better way – Top 5 with an “instant runoff” in November using Ranked Choice Voting.
California has used a Top Two primary since 2012, after Governor Arnold Schwarzenegger led the effort to put it on the ballot in 2009. At that time, California needed a 2/3 vote in the legislature to pass a budget and the Democrats were one vote short in the middle of the housing and financial crises. In order to win a single Republican vote needed to pass the budget, Democrats were coerced into agreeing to put the top-two primary system on the ballot in 2010. So the Top Two primary was born amidst partisan controversy.
In the Top Two, all candidates from all parties – Democrats, Republicans, Libertarians, Greens, independents and more – all compete in a single nonpartisan primary. It’s also called a “jungle primary” because dozens of candidates sometimes appear on the ballot and it can result in a chaotic wild ride. The top two candidates, regardless of party, then advance to a runoff during the general election in November.
The rationale was that the formerly used “closed primary,” in which each party’s primary nominated its own standard bearer who then ran in the November election, was contributing to a polarized legislature. Since most districts were lopsided partisan strongholds due to partisan residential patterns, whoever won the low turnout primary — with usually about half the turnout of the November election, and the most hyper-partisan voters — would automatically win in November.
Despite most elections being decided in the primaries, independent voters had no voice in the closed partisan primaries. In California, 30% of voters are registered either “no party preference” (22.65%) or with a ballot-qualified third party (7.25%), combined more than registered Republicans (25.14%). Those independents were only allowed to vote in the November election, after the party primary had already decided who will represent that district.
Nationwide, closed primaries have resulted in representatives more extreme left and right than the American people. With the mafia, it’s a threat that “you’ll sleep with the fishes.” In politics, it’s the threat of “being primaried” – political murder. For example, many progressives have called for primary challenges to Democrats in Congress who have the audacity to vote for bipartisan solutions rather than spout leftist talking points in protest. Recently Trump has wielded this tool to unprecedented levels to cower the Republican party. In recent weeks, Trump-supported candidates won GOP primaries against incumbents who had opposed Trump in Indiana, Kentucky and Louisiana, including an incumbent US senator.
California has now used the Top Two primary for seven election cycles over 14 years. Eliminating the closed partisan primaries has been effective in stopping candidates from being “primaried,” but the larger promise of benefits has been at best mixed. Indeed, by limiting this “open” primary to advancing only two candidates, the Top Two has produced some anti-democratic tendencies of its own.
A study by my organization, CalRCV, examined all the races under the Top-Two system since the first elections in 2012, and identified a number of defects described below. The study also looked beyond the top two finishers to the next 3 highest candidates eliminated in each race. This allowed a comparison to what would have happened if the top five candidates advanced to the general election. Here’s a summary of the findings:
In 16 percent of races, two candidates from the same party finished in the top two. That meant in the November election, all other party’s voters in those races had to hold their noses and decide which of that single party’s candidates is either acceptable or less despised.
In many races, political operatives regularly try to manipulate which candidates make it into the top two. In a number of races, the leading Democrat has promoted a Republican opponent in campaign ads, thinking the Republican would be a weaker opponent than another Democrat. This tactic was allegedly used by Governor Gavin Newsom, Representative Ro Khanna and most recently Senator Adam Schiff (whose ads mentioned GOP senate candidate Steve Garvey).
This actually defeats the electorate having a choice between a moderate and a liberal Democrat, which was how the Top Two was supposed to offer more “choice” to voters in a lopsided Democratic district, or in a statewide race in heavily blue California. This tactic remains prevalent in many Top Two races today.
Another anti-democratic pattern is that, when too many candidates from the same political party run, that has resulted in voters splitting their party’s votes, as multiple candidates spoil each other. The result is that the other major political party’s candidates finish in the top two slots – just like has been threatening to happen in California’s gubernatorial race. When this happens, the most popular candidate might not end up as one of the top two finishers.
For example, in one California race for a heavily GOP state senate seat, six Republicans split 59% of the vote, sending two Democrats to the general election. A FairVote study of the first Top Two primary in 2012 found that far more than half of the Top Two races had outcomes that split the vote and may have resulted in the wrong candidates advancing to the November runoff.
The Top Two also has the undemocratic tendency of using a low turnout June or March primary to decide which candidates are eliminated and which two are advanced to the November election. The June election usually has a voter turnout of about half of the November election, and it’s disproportionately lower among voters of color, younger voters and poor people. Given that the low-turnout primary election often decides who will win in November, that means the election is being decided when the participating electorate is overwhelmingly whiter, older and wealthier than California overall.
Another undemocratic result of the Top Two is that it has wiped out what was once a vibrant third party culture in California. Third party candidates used to regularly appear in the November general elections for governor, attorney general and the state legislature, but not any more. Per the CalRCV study, third party candidates finished in the top two and appeared the November election in only 19 races (1.7%). Thirteen of those were because there was only one major party candidate (e.g., a Democrat frontrunner with no Republican candidate). One of the only two ways a third party can retain ballot status as a registered party is to receive 2% of the vote for a statewide office in the November election. But Top-Two has de facto taken that possibility away from California’s smaller parties. Since the Top Two, third parties hardly ever appear on the November ballot, when most voters are at the polls.
Top-Two has been effective in stopping candidates from being “primaried.” But the larger promise of benefits has at best been mixed, with the flaw being the bottleneck of only two candidates advancing. We have to find a better way.
The defects of Top Two can be fixed by two changes: 1) expanding the number of candidates that advance from the primary to the November general election from two to five; and 2) using an “instant runoff” (ranked choice voting) in the general election to ensure the winner has a majority of the vote, and to prevent spoiler candidates and split votes. Below are some other advantages of moving to Top Five, as noted in the CalRCV study:
More women candidates in the November election. In 152 of 1111 total races (14%), there was at least one woman candidate in the top 3rd-5th primary vote getters, but only two men advanced to the November election in those races. Top Five thus would have given the voters in the higher turnout November election more women candidate options in those 152 races.
More minority candidates in November elections. In 141 races out of 1111 total races (13%), there was at least one minority candidate in the top five in races in which no minority candidate advanced in the top two. The biggest negative impact was on Latinos. One of the reasons for minorities being locked out of the Top Two is that minorities have disproportionately lower turnout in the June or March primary elections, as reported by FairVote in studies of Sacramento, San Bernardino, Oakland and San Jose.
More third party candidates. In 219 races (20%) a third party or independent (NPP = No Party Preference) candidate was left out of the top two but would have finished in the top five (slightly over half were NPP candidates). That number is surely depressed because third parties often have not fielded a candidate out of frustration over the overwhelming hurdle of making it to the November election in a Top Two primary.
Alaska sends the four highest finishers in its open primary on to the November general election. Oftentimes the general election includes both Democratic and Republican candidates and third-party or independent candidates. In the two elections where Alaska has used a Top Four primary plus RCV in the November election for governor, senator, U.S. representative, state senate and state house, over a third of the primary races saw a third party or independent candidate finishing in the top four (35% of the 2022 races and 36.5% of the 2024 races). Of those, about two-thirds were independents and one-third were third parties.
Following the Top Four primary, Alaska has used ranked choice voting in the November general election to elect majority winners. Alaska Senator Lisa Murkowski survived Trump’s attempt to primary her because of Alaska’s Top Four system. And Alaska elected a moderate Democrat to the House over right-wing Republican Sarah Palin. Also, in the new cross-partisan culture fostered by RCV, the majority Republicans in the legislature formed a coalition with moderate Democrats, rather than letting legislation be held hostage by a handful of their extremist MAGA Republican colleagues (as often happens in the US House).
A Top Five primary would be even better than a Top Four. By sending five candidates instead of four (or two) to the general election, and running the general election with the instant runoff feature of ranked choice voting, there can still be vote-splitting in the primary but, with five candidates advancing, any vote-splitting would have less impact. The odds of a general election with no Democrat or no Republican are infinitesimal. In addition, the odds are increased even more than in a Top Four primary that we would see more third party and women candidates and more candidates of color making it to the November election. And in November, the spoiler effect would also be eliminated by the use of Ranked Choice Voting.
It is time for California to upgrade its deficient Top-Two primary to a Top-Five primary plus RCV to ensure majority winners. That would produce a great win for democracy.