UCSD study: Recall election could be unrepresentative of California

As Gov. Gavin Newsom faces a recall election on Sept. 14 which could have political ramifications both in the state and across the nation, the off-cycle nature of the election could lead to low voter turnout, which may favor recall, according to UC San Diego research announced Friday.

Low and uneven turnout is a serious problem for local democracy, said Zoltan Hajnal, professor of political science at the UCSD School of Global Policy and Strategy. However, simply moving off-cycle, local elections to be held on the same day as statewide and national contests doubles voter turnout and leads to an electorate that is considerably more representative in terms of race, age, class and partisanship, according to his research.

Hajnal’s study looked at all local elections held in California over an eight-year period. The findings could have major implications for the state’s recall election — the mail-in ballots for which are already arriving in voter’s mailboxes.

“What our results tell us is that the more unusual we make the timing of an election, the fewer people are going to turn out to vote and the more unrepresentative those votes are going to be,” Hajnal said. “With a recall election that is on an unexpected date, we are essentially making it more difficult for voters to participate. As a result, we might see lower turnout that is less representative of the California populace.”

He added that when local elections are not held on the first Tuesday of November with other statewide and national contests, it can create barriers because local voters need to learn the date of their local election, find their polling place and make a specific trip to the polls just to vote on local contests.

The study, published in American Political Science Review, finds that voters in California during off-cycle elections are 67% white on average, but that number drops to 57% during statewide or national elections. By contrast, the share of voters who are Hispanic increases almost 50% during on-cycle, local elections — growing from 18% of the vote to 25%, while the Asian American share of the vote increases by 30% — shifting roughly from 8% to 10.3%.

Effects are even more pronounced in terms of voter age. Older Americans are represented twice as much in off-cycle elections than their numbers are reflected in the overall population. But that overrepresentation is reduced by roughly half in on-cycle contests.

Finally, Democrats are generally slightly underrepresented in off- cycle elections, but moving to concurrent local elections produces a near proportional turnout of both parties. The paper also found significant effects for class, which is measured by income and wealth.

The findings reveal that in some places, off-cycle elections lead to politically active conservative, wealthy, older, white voters having a disproportionate sway over local government.

“The result is that an extraordinarily unrepresentative set of residents determines how local governments distribute services and spend the almost $2 trillion that local governments control,” Hajnal said.

A poll from the UC Berkeley Institute of Governmental Studies co-sponsored by the Los Angeles Times found that 50% of likely recall voters wanted to keep Newsom and 47% wanted to recall him, and the poll difference was within the margin of error. An energized group of Republican voters skewing white, older and wealthier voting in a low-turnout, off-cycle election could tip the balance toward recall.

The study analyzed 2,000 elections of roughly 500 California cities from 2008 to 2016. Municipalities, such as San Diego, that moved election timing during this period drove the results of the study.

In San Diego, mayoral elections held during off-cycle dates have turnouts that are much lower. Turnout in the last mayoral election held off-cycle on Feb. 11, 2014 was 290,192. By contrast, turnout in the most recent mayoral election on Nov. 3, 2020 was 619,549.

The results are even more dramatic for Los Angeles. It experienced a 400% increase in voters casting ballots in local elections for the city’s first on-cycle election in March 2020, compared to the last off-cycle election in March 2015. In the 2020 primary elections, 604,000 more people cast ballots in city races compared to the stand-alone, city primaries in 2015.

While the paper focuses on California, Hajnal and co-authors did the same analysis in Florida and found similar effects. Although the magnitude is slightly smaller in Florida, when cities moved to on-cycle elections, it increased their representativeness in terms of race, age and income.

“Across the nation, about 70% of all city elections are off-cycle, so the vast majority are under a system that essentially minimizes and skews turnout,” Hajnal said. “On-cycle elections likely will not lead us all the way to proportional participation, but they get us much closer.”

More from Politics