CA120: Curing California’s ‘extra innings’ elections

In California, elections go into extra innings.

While in other states, they’re packing up the campaign office in mid-November, in California races are still being won and lost through organized field programs doing ballot signature “curing” that was unheard of just a few election cycles ago.

Nearly 200,000 voters who cast ballots in this election had a signature missing or unmatched at some point in the certification process. That’s over 1% of votes cast, and nearly 5% of all voters 18-24. To better understand the experience of voters, causes and eventual resolution, we did analysis of these voters, the age, ethnic, partisan and other breakdowns, and conducted a survey of nearly a thousand Californians who found themselves with their vote at risk.

Why is this happening?

The change of California law goes back a decade ago when Mindy Romero, then at UC Davis, did a research paper showing that issues with invalid or missing signatures were the second and third most common reasons for ballots to be rejected – only behind late ballots, that at that point needed to be received by Election Day. When that research was done, ballots could only have a signature fix if the change was made prior to the election, and while most counties tried mailing or calling voters, there was a lack of consistency, and little evidence that many ballots were getting cured.

For the 2014 General Election the state implemented a law allowing ballots to count if postmarked by Election Day and received within three days. But the state did not make substantive changes to the rules for signature verification until after a lawsuit by the ACLU (La Follette v. Padilla) which built upon the Romero research, and included additional findings of the unequal processes used by each county, and how certain minority groups, including voters who receive ballots in a foreign language, were more likely to have their signatures rejected.

In 2018 the court found in favor of the plaintiff and ordered that the state allow voters to fix their signatures through the full process of certification – a one-month window after Election Day. This opened the door to new statutes regarding ballot signatures and the extended curing process we have today.

But the “extra innings” seen in this election cycle didn’t come immediately.

In the 2020 General Election, counties began implementing the new law in a way that became a focus of several close races. This was also the COVID election when the state, for the first time, mailed every voter a ballot, and allowed voters to return them by mail, drop box, or at new voting centers or traditional precincts depending on the county.

Campaigns that knew of the law regarding ballot cures started digging into the process, hoping that a fix to even a few ballots in their districts could make the difference in their races. But counties who hadn’t previously been asked for lists of voters who needed a ballot cure were hesitant to provide them – preferring to contact voters themselves, and unsure if it would be allowed for campaigns to assist, or even if it would be helpful.

While counties were hesitant, PDI did obtain lists of voters that needed ballot cures in a few counties and began working with a few congressional campaigns on matching these to the voter file, allowing them to assist in the collection of signature verification forms.

In one of the funnier episodes from that election cycle, John Myers, then at the Los Angeles Times, called a county asking about this process. The county said they had no such list… until he showed them that he had access to these lists in his PDI account. It wasn’t until Secretary of State Alex Padilla sent out a memo to counties encouraging them to release the data that campaigns started getting full access.

Fast forward to 2024 and we see a massive statewide ballot curing process, with campaigns, political parties and nonprofit organizations actively involved. As can be seen in the following table, there were 197,133 total voters that were notified that there was a problem identified with the signature on their ballot. Of these, 113,962 were fixed, leaving 83,171 ballots statewide that were not counted.

Young voters, those aged 18-24 were 28% of the total, even though they are only 10% of voters. They also had a lower rate of fixing their signatures, leaving voters in that small age group to be a third of the total votes that were invalidated.

More than two-thirds of seniors resolved their signature errors compared to just over half for voters under 35. We also saw a slightly higher cure rate for partisan voters than nonpartisans. This could be a function of them not being pushed as much by political campaigns, or it could just be caused by the fact that more young voters are registered independent.

We saw higher rates of ballots being fixed in the competitive legislative and congressional districts and where there were other competitive races, like the Sacramento mayoral race. And social media provided glimpses into this, such as Christine Pelosi on the Democratic side, seemingly traversing the state, curing ballots in Riverside, Orange County and the Central Valley. And, on the Republican side, we saw Chairwoman Jessica Millan Patterson encouraging campaigns to be patient through the cures process as they too were working to make sure all valid votes could be counted.

The final result in Congressional District 13, with Democrat Adam Gray winning by just 187 votes over incumbent Republican John Duarte, could have been determined by their ballot curing operation.

Other recent examples of curing of ballots helping to decide a race include the 16^th Senate District race where Melissa Hurtado won in 2022 by just nine votes after an extensive curing operation and the 16^th Congressional District primary where Evan Low ultimately earned the second spot in the runoff election when cures ended up leading to a tie and a recount resulted in a five-vote win.

In our statewide survey we asked voters who had been on a county cures list if they knew about the status of their ballot, who had contacted them, if they received any communication from political campaigns trying to help them, and if they supported the law that allowed them to fix their ballots after election day.

Overall, 67% said they did know there was a problem with a missing or unmatched signature, while 33% said they did not.

In this election the county flagged your ballot as either not having a signature or having a signature that they couldn’t match to your signature that they had on file. Were you aware of this?

This survey was conducted before the deadline for curing, so in some cases voters stated that the first they had heard about their ballot having a problem was from our poll. Among those who said they didn’t know at the time of the survey, most did eventually get their ballot cured and counted – either by doing the paperwork to cure their own ballot, or because a higher level of elections administrator validated the signature.

For those who were aware of their missing or mismatched signature, the most common cause for voters to be placed on the cures list was a sloppy signature, or a name change.

What was the cause of your signature needing to be fixed on your ballot?

The 3% of voters who had someone else sign their ballot (which is not legal) can be better understood by reading three of the open-ended statements left by voters who selected this option. In one case, a woman had been signing for her husband who was paralyzed, in another two spouses had accidentally signed each other’s ballots, and in a third, the voter said they had broken their arm. These anecdotes help illustrate all the variety of unique situations that can arise when you have 16 million ballots being signed and validated.

More than half of the respondents cited the county registrar as their first point of contact and among the “other” responses most reference some official communication. So, in total, it is likely that three-quarters or more learned from their county. Only 6% said their first notification came from a political campaign.

How did you learn that your ballot needed to have the signature fixed?

The most common method of initial contact according to respondents was physical mail, at 38%, followed by texts at 24%, by phone at 14%, email at 14% and just 4% citing someone coming to their door.

In addition to this first contact, one third of voters cited another group reaching out to them to get their ballot cured. The most prevalent was individuals identified with a political party, followed closely by campaign staff who didn’t identify a candidate they were working for, then local races, followed by legislative and congressional campaigns.

It is possible that several people contacted you to let you know that your ballot needed to be fixed, also called “cured” in order for it to count. Which types of people or groups contacted you? Select all that apply.

One interesting datapoint is the higher rate of ballot curing by partisan voters. We can see evidence of campaign efforts that were more aggressively turning out voters registered with their party. Looking specifically at the competitive congressional seats, Districts 13, 22, 27, 41, 45 and 47, we see a rate of 85% cure rate for both Democrats and Republicans and 62% for nonpartisans – a 23-point difference, and significantly higher than the statewide average.

For those contacted by a political campaign, we wanted to know if they were asked who they had voted for before being helped to fix their ballot. Only 9% had that experience. Even in the competitive congressional seats this only grew to 12.5%. This doubled for independent voters, with 24% of those in the competitive districts responding that they were asked who they voted for before getting assistance from the campaign.

At the same time, a total of 8% of voters were encouraged to return ballots by campaigns they didn’t support – 4% only being pushed by a campaign they didn’t support, and 4% getting calls from both sides of a campaign.

Was the campaign that contacted you to fix your ballot or encouraged you to do it, one that you actually voted for?

Being contacted by a campaign did measurably increase the likelihood of the voter fixing their ballot. For voters who only remember getting contacted by the county, 72% of the survey respondents were able to get their ballot cured. However, for those reached by a political organization the share increased to 86%, for those reached by a legislative or congressional campaign it grew to a whopping 91%.

While 68% of poll respondents did get their ballots cured based on data from the County Registrars, only 31% knew for sure their ballot had been counted, 22% believed it had not been counted, and 47% were unsure. This is an issue that was repeatedly raised in the open-ended questions where voters stated concerns that their ballot hadn’t been counted, even though we can see from the county data that they were.

Finally, voters were asked if they supported the law that allowed them to fix their ballot, even after Election Day, and as would be predicted, they overwhelmingly supported it.

The State of California allows voters to fix their ballots after Election Day, as long as those ballots were cast or postmarked by Election Day. Do you support this law?

Going forward we can expect more attention on the ballot curing process – especially in areas where campaigns expect close contests. This means campaign staff aren’t being jettisoned on Election Night – they are being either converted into phone calling, emailing, texting and door knocking ballot cure operations, or being shipped off to somewhere in the state where a race is likely to be won by a slim margin, and where curing can make the difference.

And state and county officials should be looking into more education around ballot signatures – particularly among young voters. In 2020 the state funded millions of dollars in outreach because the state was introducing the mailing of ballots to all voters for the first time. But this kind of funding hasn’t been continued, and as we can see, over 80,000 ballots, disproportionately young and Latino, were disenfranchised by simple issues with ballot signatures.