CA120: The lowdown on ‘ballot harvesting’

A recent dustup with the California Republican Party using unofficial dropboxes as a version of so-called “ballot harvesting” has brought the state’s ballot delivery process under a national spotlight.

Much of this controversy can be attributed to the misleading way in which the law has been interpreted, most commonly by people who are trying to conjure up scandal and supposed misdeeds by campaigns that organize such efforts and win.
With about a week to go in this election, it seems like a good time to dive into some facts. What does the state’s ballot delivery law, created by AB 1921 by Democratic Assemblywoman Lorena Gonzalez, actually entail? Who does it? And how is it effectively, or not-so-effectively, used by campaigns?

To begin with, if you’re calling it “ballot harvesting” you’re doing it wrong. That is a pejorative term used very disingenuously to conflate the very legal process of authorizing someone to return your ballot with very illegal practices, such as what happened in North Carolina in 2018, when a campaign collected ballots with the intent of changing them, or not returning them at all.

Organizing a campaign that includes picking up ballots and delivering them for voters has always seemed like the by-mail version of picking up a voter and driving them to the polls, something I’ve done on several campaigns. And we didn’t call that “people harvesting.”

The hyperbole about the state’s ballot delivery law would make you think that tons of campaigns are building a whole infrastructure around collecting hundreds of thousands of ballots, and that this is what has tipped campaigns. It became national news when Democrats won several competitive congressional races in 2018, and the “blue shift” of late ballots skewing toward Democrats was seen as a sign of something corrupt in the process.

The policy change allowed for others, such as a neighbor or roommate, to collect the ballot.

Then-House Speaker Paul Ryan openly claimed it was fraud and California’s “bizarre” system led to the Democratic wins:

“We had a lot of wins that night, and three weeks later we lost basically every contested California race. This election system they have, I can’t begin to understand what ballot harvesting is.”

And in a report from the Committee on House Administration on the “weaponization” of the ballot delivery process, we find this description:

“The issue of ballot harvesting was thrust onto the national stage following the defeat of multiple Republican California Members of Congress in the 2018 midterm elections. It was several days after the election, and after the counting of hundreds of thousands of ballots harvested under a new California law, when election observers and California voters started to raise red flags on what they witnessed.”

These reports represent three significant misunderstandings about the use of ballot delivery in campaigns.

First off, the law was a pretty simple change. Existing law allowed a family member to return a ballot when the voter signs over their ballot and identifies the relationship. The policy change allowed others, such as a neighbor or roommate, to collect the ballot.

According to the National Conference of State Legislatures, this is the same law used in 26 states, including several so-called Red states.

This change allows a campaign to organize efforts with staff or volunteers to collect and return ballots from voters they were trying to turn out, as long as they aren’t paying a per-ballot bounty for the collection.

Calling the law “bizarre” is a stretch. According to the National Conference of State Legislatures, this is the same law used in 26 states, including several so-called Red states.

Secondly, ballots delivered through this process have to be submitted to the county registrar within 72 hours. Critics really amped up their rhetoric about ballot harvesting when the late votes were counted in 2018, but coordinated efforts that are conducted for the entirety of the 29-day voting period would not be over-represented in these final ballots.

The very late ballots that were counted near the end of the canvass period in 2018 and “flipped” those races were predominantly same-day registrations that had to be processed as newly registered voters before their ballot was tallied, along with provisional voters who showed up to a polling location without their absentee ballot. These provisional voters were required to vote with a ballot that would be counted only after all the other ballots were processed, to ensure the person wasn’t voting twice.

The much more cited use of the ballot delivery law was simply neighbors and friends returning a few ballots, rarely more than 3 or 4 at a time.

Finally, the “hundreds of thousands” figure is outlandish. Organized ballot delivery programs are small, limited operations and there is no evidence that campaigns are turning in thousands of ballots, let alone hundreds of thousands.

We queried county registrars and only one had seen any large, coordinated, ballot delivery program in their county. This registrar, in a county which has some of the top competitive congressional and legislative districts, stated that they had seen one such effort and in that case it was “not thousands just a few hundred.”

One registrar provided us with the following insight:

“As of yesterday we have received 2,857 ballots dropped off by someone other than the voter. Most who drop off others’ ballots are doing so for family members or other persons they live with, usually between two and four total, and generally on the lower end of that range. ‘Neighbor’ and variations on boyfriend/girlfriend/partner/significant other are the other most common identifiers. To see ten or more ballots dropped off by the same person is rare, and we have not had any instance in this election of someone dropping off dozens or hundreds of ballots, which seems to be a widespread misconception on this issue.”

That county had 100,00 ballots at the time, so this represented just 3% of ballots.

The much more cited use of the ballot delivery law was simply neighbors and friends returning a few ballots, rarely more than 3 or 4 at a time.

Three counties cited ballot delivery programs they knew of at local nursing homes, one of which was working with the county and had identified a resident that was collecting the ballots, and would meet a member of the registrar’s staff to turn them in before the election.

Looking at the scale of ballot delivery programs it becomes clear that they are not the big boogeyman of the state’s election system.

Another election official cited the use of the ballot delivery law in hospitals where residents can’t leave, and family members can’t visit, but a nurse or doctor can be the go-between to get ballots delivered legally, in a way that wasn’t legal before the passage of the AB 1921.

Looking at the scale of ballot delivery programs it becomes clear that they are not the big boogeyman of the state’s election system. Campaigns I have spoken with target these efforts at a small portion of the electorate – generally the 20% of voters in their political party least likely to turn them in on their own.

And with competitive races only really covering 15% of the state and even those counties suggesting that coordinated efforts are a fraction of ballots delivered through the state’s ballot-delivery authority, the numbers get small really quick.

Just 20% of the electorate in 15% of the counties show some small single-digit percentage of ballots received by registrars being submitted in this fashion.

This suggests that less than 1% of ballots are returned through these programs. That means more than 99% of ballots returned by someone other than the voter are done by roommates, neighbors, fellow residents in a nursing home, or some other local, individual effort.

The GOP has been consistently railing against the law, casting aspersions about Democrats’ use of ballot delivery programs.

There have probably been more hours spent by reporters writing about “ballot harvesting” in 2020 than actual hours spent by campaigns doing it.

The widespread misconception of the issue — and the outsized media attention given its  de minimis usage by campaigns — is potentially what caused the California GOP to get caught in a controversy regarding their use of unauthorized drop boxes as a part of their ballot delivery program.

The GOP has been consistently railing against the law, casting aspersions about Democrats’ use of ballot delivery programs, and suggesting they were being done on a massive scale to flip key contests. Anyone who was devising a program to compete against this fictional view of ballot delivery in California might think about a grand-scale program like the one they sought to create.

And the question becomes, aside from the legal ones, is this: How effective could that program even be in driving election outcomes? Did the GOP expenditure of time and money to create these dropboxes and implement their ballot delivery program have more effectiveness than if those same resources had been used for other kinds of outreach or voter persuasion?

The fact is that Democrats and labor have a lot more infrequent voters and need more outreach to try and achieve the same turnout levels.

Looking at vote history, 82% of Republicans voted in the 2016 General Election. This figure was even higher, at 85%, in the competitive districts which were being targeted. That suggests that, at best, only one-in-seven of the ballots collected would even be considered a fruitful use of the resources.

When the GOP program initially came under attack, they had reportedly garnered ballots in the range of a hundred or more in one county that made the information available. If, like the rest of the GOP electorate, this population of ballot delivery was already 85% likely to turnout, then we are looking at very small marginal gains, maybe in the range of dozens of ballots in that county, not thousands or tens of thousands.

Talking to several Republican operatives, there has been years of consternation about the GOTV efforts by Democrats, and big concerns internally that they are being outdone by Democrats and organized labor with their large-scale grassroots operations.

But the fact is that Democrats and labor have a lot more infrequent voters and need more outreach to try and achieve the same turnout levels. This is particularly true in low-turnout elections and it was even more important when mail-in ballots in most counties required postage.

We can expect to see ballot delivery used increasingly in future years, particularly if voters like the vote-by-mail options this year…

In a high turnout election, where all ballots are postage-paid and with the proliferation of dropboxes throughout the state, any organized staff-driven ballot delivery program by either party is unlikely to be a big driver of outcomes. Put bluntly, it is probably a waste of time and money.

What could drive outcomes would be encouraging your voters to use one of the more pedestrian examples of the law, such as my mom taking the elderly neighbor’s ballot to the vote center or my stepson taking his and his roommate’s ballot to a dropbox.

If I was a campaign interested in building turnout, that might be a better tactic. Republicans could, for example, text message all pure-Republican households that don’t appear to be entirely families —  a so-called “non-traditional” household — and let them know that they can turn in the ballots of a friend or roommate. That would be much less resource intensive and target how the law is being used in practice, rather than how it has been demonized in press releases.

We can expect to see ballot delivery used increasingly in future years, particularly if voters like the vote-by-mail options this year, and more counties adopt the Voters Choice Act use of by-mail ballots and vote centers. And the impact of coordinated ballot delivery programs will likely be much greater in elections like we could have in 2022, with new district lines and anemic turnout that we saw in 2010 and 2014.

By 2022 we could have campaigns from both political parties engaged in aggressive, smart, and legal ballot delivery programs. And, hopefully, by that time we can stop attacking it as “ballot harvesting” and agree to see the law for what it really is.
—
Editor’s Note: Paul Mitchell, a data expert and a regular contributor to Capitol Weekly, is the creator of the CA120 column, vice president of Political Data, Inc., and owner of Redistricting Partners, a political consulting firm.