Many observers falsely assume that California’s electorate gets uniformly bluer between primary and general elections, giving Democratic candidates an edge in competitive races at the congressional and legislative levels. Reality is more complicated than any statewide average admits. Shifts don’t just vary based on political environments; they’re also affected by different demographic and geographic characteristics of individual districts.
To define when, where, and to what extent California’s “blueshift” actually occurs, we broke down the average shift by election cycle into eight distinct buckets based on race and region. Averaging these district classifications allows us to see how specific groups shifted relative to the state as a whole in a given year.
Our findings disprove the common assumption that the California electorate undergoes monolithic shifts. The 2020 cycle, for instance, produced a R +4.1 shift, 13 points right of the 2012 shift, which saw the electorate get 8.8 points more Democratic.
In 2012, the GOP presidential primary was held concurrently with the congressional nominating contests. Although all of Mitt Romney’s challengers apart from Ron Paul had dropped out by June that year, Democrats won just 52% of the combined primary vote, which artificially inflated Republican strength in the statewide primary, and therefore, the congressional primary as well.
Obama’s November reelection win margin of 23% also contributed to the D +8.8 shift, bringing Democrats out to the polls in competitive districts. While Republicans in crossover seats like Jeff Denham (CA-10) and David Valadao (CA-21) outperformed Romney, Obama’s strong showing likely made their reelections closer than what the primary partisan shares would have suggested at face value.
2014 saw California’s statewide offices contested under the nonpartisan blanket primary system for the first time. As voter turnout was historically low, indicating a Republican-favoring primary environment, the primary-general shift benefited Democrats under the pre-Trump partisan coalitions. This was enough for Democrat Pete Aguilar to win the Inland Empire-based seat that he had been locked out of in 2012. The combined congressional vote was still D +19, four points redder than Obama’s win two years earlier.
In 2016 a competitive Democratic primary helped bring more Democrats to the polls than there were in 2012 or 2014. This helps explain a manifestation of a small, but perceptible rightward shift between the primary and general, as the Republican (and American Independent) nominee Donald Trump had already been crowned by June.
Democrats also enjoyed a strong leftward shift in 2018, albeit for a different reason. They struggled with untapped turnout potential and a nascent level of organization in the primary. In November, more independents turned out to vote for Democrats and reject President Donald Trump’s agenda and vulnerable Republican representatives ㅡ causing a D+5 shift.
Similar to 2016, 2020 saw a competitive Democratic primary running alongside an uncompetitive Republican primary. With the outcome of the Democratic Presidential primary still up for grabs, Democrats throughout the state turned out in overwhelming numbers on Super Tuesday for their preferred candidates. By swamping out the comparatively tepid Republican turnout, the primary environment painted a more Democratic environment than the final general electorate, leading to a strong Republican environment gain between the two elections.
In 2022 we found that the Golden State shifted 1.4 points to the right between last year’s primary and general elections. This is likely due to the implementation of statewide vote-by-mail, which makes voting in the primary more accessible, decreasing the motivational barrier voters must climb to have their voices heard during an election.
Groups which Democrats typically rely on turning out more relative to the whole electorate at large, such as Hispanic or young voters had greater accessibility to voting, meaning the primary-to-general motivation increase no longer causes large, disproportionate turnout increases along demographic lines.
Regional & Demographic Analysis
By breaking California’s congressional districts into eight demographic buckets, we discovered several distinct demographic patterns. The most consistent pro-Democratic shifts were observed in the Central Valley’s white/Hispanic seats, a result of low-propensity Hispanic voters being more likely than their white counterparts to turn out only for general elections.
Interestingly, the four “mixed” districts (under the old lines: 6, 13, 37, 43) swung sharply toward the Republicans in 2020 and 2022. Containing roughly-even concentrations of at least three of the four main racial groups, these seats are sometimes home to latent minority populations sympathetic to conservative causes under the right circumstances.
Mixed seats generally are almost uniformly Democratic, meaning that Republican candidates in such districts are generally heavily reliant on independent voters who typically only come out in general elections. Conservative, non-college minority voters are also more likely to abstain from the primary and opt to vote only in general, if at all.
An overall look at the data since 2012 highlights another important conclusion. At the demographic level, midterms featured more Democratic primary-general shifts than the last two presidential elections because 2016 and 2020 primaries were more competitive for Democrats than Republicans.
The white/Hispanic districts in the Central Valley continue to have the strongest blueshift thanks to differential turnout and racial polarization, but education polarization has also played a palpable role, possibly driving redshifts in Asian/Hispanic, Hispanic, and mixed seats.
Given that California’s minority voters generally do not have college educations, it’s possible that 2020 (and to a lesser extent 2016) ushered in a miniature-realignment along educational lines in the Golden State’s electorate by reversing the partisanship of latent low-propensity voters. This has some interesting implications for 2024.
We recently debuted our preliminary 2024 House ratings characterizing the chamber as a Tossup. In California, we placed five Republican-held districts in the Tossup category: CA-13 (Duarte), CA-22 (Valadao), CA-27 (Garcia), CA-41 (Calvert), and CA-45 (Steel). Our findings suggest that educational polarization had a lesser impact on the Central Valley than it did elsewhere in California, simply due to the high incidence of racial polarization in its place.
This dynamic could make or break Mike Garcia’s reelection chances in 2024 should the miniature realignment hold up. Garcia won narrowly in 2020 thanks to his margins in Simi Valley before notching a comfortable victory in a rematch against Christy Smith in 2022.
Despite an unfavorable redistricting cycle, Garcia’s recent reelection proves that Democrats need huge Hispanic margins and turnout in a general election to beat him, as white voters in northern Los Angeles County still lean Republican. Democrats’ task could be eased by presidential turnout in 2024, but it should be noted that they haven’t won the seat since 2018 ㅡ the last cycle in which Southern California white/Hispanic seats had an average blueshift.
Calvert’s seat also fits into the white/Hispanic category, but it’s unlikely the same dynamics driving white voters rightward in the 27th will hold in the 41st. There are, after all, obvious differences in cultural identity and voting patterns between whites in both districts, exemplified by Palm Springs, which is whiter than surrounding cities in the Coachella Valley but significantly more liberal.
Finally, in Steel’s 45th, the Asian/Hispanic coalition in Orange County is different from those in SoCal’s two other Asian/Hispanic seats. Unlike the 28th and 31st, the 45th got more Democratic between primary and general in 2022 independent of educational polarization.
Steel’s strength comes in her margins with Vietnamese and Korean voters ㅡ key voting blocs necessary for any Democratic victory. Additionally, white voters in northern Orange County are still Republican leaning like their counterparts in northern Los Angeles County.
After calculating the average shift between primary and general elections in each of California’s House districts for every cycle between 2012 and 2022, we broke the results down into eight demographic-regional categories: white, white/Hispanic NorCal, mixed, Asian/Hispanic NorCal, white/Hispanic Central Valley, white/Hispanic SoCal, Asian/Hispanic SoCal, and Hispanic SoCal.
We also excluded seats that were not contested by both parties in the general election and those that only featured write-in major party opposition in the blanket primary to remove outliers from the dataset. Additionally, groups not categorized in the region and demographic breakdown could play a role in the margin shift, limiting the extent to which these finding are conclusive.