June 5 primary: The chase is on

The chatter online and in the media is all about the June 5 Primary Election. But, for those of us working in these races, the election has been ongoing for weeks. In fact, as of Memorial Day weekend, 1.25 million California voters have cast ballots, approximately 20% of the expected total turnout of by-mail and poll voters.

With this large chunk of ballots already in, we see the beginnings of informed analysis about what is going on in this election cycle. Will there be a big blue wave?  Will Latinos respond to the most diverse ballot in state history? Will turnout continue to favor the San Francisco Bay Area over Los Angeles?

More than two-thirds of the voters 18-to-34 years of age who received ballots this year have never mailed in a ballot.

As in past years, these absentee ballot returns are being tracked daily, from county-level data, by Political Data Inc. The ballot returns, with breakdowns by ethnicity, party, age and geography can be followed here and there is a slimmed-down version for mobile devices here.  Both versions allow the user to see statewide returns or select for specific congressional or legislative districts, even looking down to maps of precinct-level returns.

But, before diving into analysis of this data, it is important to outline two major caveats. First, the wide adoption of vote by mail has changed the kinds of voters who are receiving ballots. Second, this primary election is unlike recent primaries and it is complicating any direct comparisons to past election cycles.

The rate of by-mail voting in California has skyrocketed.

Ten years ago, less than 5 million voters obtained ballots in the mail, by 2014 this reached 8.3 million, and since the last gubernatorial primary, mail-in ballots surged again to 11.5 million.

In five counties, including Sacramento and San Mateo, all voters are receiving ballots by mail.

This growth in vote by mail has been uneven, with little change among older likely voters who were the early adopters of by-mail voting. But there have been huge changes among some of the lowest-turnout voters, particularly younger voters, most of whom have never voted by mail.

More than two-thirds of the voters 18-to-34 years of age who received ballots this year have never mailed in a ballot.

The changes in the by-mail voting universe means that any analysis of voting trends should be done with caution.

Because of these changes in who is getting a mail-in ballot, less than half of them are considered by Political Data to be likely to vote in the coming election — an historic low.

In addition to the changing environment of by-mail voting, primary turnout is even more challenging to compare because not all primaries are the same.

In 2014 there was no Democratic challenger to the incumbent governor, but there was a competitive Republican race. And in 2016 this was flipped – the Republican nomination was sewn up when Donald Trump won in Indiana, but the Democratic race was still raging.

Additionally, independent voter turnout was down significantly in 2016 because Republican-leaning independents were blocked from voting in that primary, and even Democratic-leaning vote-by-mail independents had trouble getting the correct ballots.

These changes in the by-mail voting universe, and the challenge to find an historic election to serve as the baseline, means that any analysis of voting trends — and drawing conclusions about the likely final turnout — should be done with caution.

An instructive lesson can be learned from earlier this year in Texas.

Early-vote numbers suggested that Democrats were in a big blue wave because voters were casting more Democratic than Republican early-vote ballots – a flip from the Republican dominance since 2008.

This is 85,000 more ballots than in 2014, an increase by 7.5%, but 400,000 ballots short of two years ago, a 25% drop.

However, as was seen by Election Day, Republicans ultimately ended the election with a 20-point advantage. Democrats weren’t actually turning out in higher numbers, they were just voting earlier than they had in the past.

With all of these warnings, here is what we see in the early absentee vote returns in California, with some comparison to past elections as reference points.

Total Turnout
Looking at snapshots for Memorial Day weekend in 2018, compared to 2014 and 2016, we can see some growth compared to the last gubernatorial primary, but a very understandable decrease when compared to the 2016 presidential primary.

There have been 1.2 million ballots returned this election cycle, compared to 1.1 million in 2014 and 1.6 million in 2016. This is 85,000 more ballots than in 2014, an increase by 7.5%, but 400,000 ballots short of two years ago, a 25% drop.

Compared to 2016, the Democratic share of the electorate is down 5.3%, Republicans are holding steady, and independents are up 5.3%

Of course, there are a lot more ballots out than in prior elections. In terms of turnout, we can see that this weekend represented a 16% rate of ballot returns in 2016, 13% in 2014, and a current rate of 10.3% turnout of all vote by mail voters for 2018.

Thus far, there is a high number of raw number of ballots being returned for a gubernatorial cycle, but it is not keeping pace with 2016.  And as a share of all absentees, the numbers are lagging and neither measure suggests any kind of historic high turnout for June.

Partisanship
The breakdown by partisanship shows some raw growth over 2014, but significant decreases among Democrats and Republicans compared to two years ago.

These increases over 2014 numbers are rather small, and when adjusted for the raw increase in vote by mail, they fall into the red, with turnout among Democrats who have received ballots down by 3.2% and among Republicans down by 1.6%.

Those aged 18-to-34 are nearly one-quarter of the voters who received ballots in California, but they represent just 10% of the returns.

However, the most important thing for campaigns is how well partisan groups are performing against each other, which helps us consider whether the 2018 primary will be a big blue-wave election, or merely a traditional gubernatorial primary with higher Republican turnout.

As of this time, the analysis is mixed.

Compared to 2016, the Democratic share of the electorate is down 5.3%, Republicans are holding steady, and independents are up 5.3%. But compared to 2014, Republicans are under-performing by 1.5%, a small number representing under-performance by about 7,000 voters at this time.

The bottom line is that we are seeing slightly higher turnout among independents, but turnout to date points more to a Republican over-performance, with stable but not high Democratic turnout, consistent with past gubernatorial and likely voter projections.

Age
It almost gets tiring writing that young people don’t vote. But it always seems to be true, particularly in the lower-turnout mid-term primary elections. And 2018 isn’t an exception.

Those aged 18-to-34 are nearly one-quarter of the voters who received ballots in California, but they represent just 10% of the returns. At the other end of the scale, seniors are also roughly one-quarter of the voter file, but they account for half of the ballots that have been returned.

California saw historic registration and turnout of young voters in the 2016 election cycle, but you have to wonder how many of these voters have moved to college, or came back from college, or just switched living situations in the last two years.

California has two large ethnic populations that have each been referred to as “sleeping giants,” with the potential to awaken and send shock waves through our elections.

The transient nature of younger Californians clearly is a contributing factor to their low turnout, but other factors could be in play here.

Younger voters don’t really use the mail the way their parents do and they often don’t see the primary elections as their responsibility.  Additionally, younger voters are much more likely to be independents, and we know independent voters are lower turnout in partisan primary elections.

While the impact of young voter under-performance might not mean much in the primaries, it could be critical for the General Election, particularly in the competitive congressional districts.

This is set to be an old primary, with the median age around 59. Millennials may be the largest part of the voter file, but with a turnout boost we can expect the Baby Boomers to decide most of these races.

Ethnicity
California has two large ethnic populations that have each been referred to as “sleeping giants,” with the potential to awaken and send shock waves through our elections.

The Latino community is 38% of the state’s population and 25% of the state’s registered voters. However, Latinos represent only 14% of ballots returned.

The Asian community has consistently performed at a rate roughly equivalent with their share of the voter file, with some examples of extreme over-performance, such as Vietnamese voters in Orange County supporting Vietnamese candidates, Koreans backing candidates for Congress and City Council in Los Angeles, and others.

In this election there are prominent Asian candidates for governor, controller and state treasurer.  However, Asian turnout appears to be consistent with their projected turnout.  It is possible that Asian vote will spike in certain areas with prominent Asian candidates, but that hasn’t made itself shown yet.

The Latino community is 38% of the state’s population and 25% of the state’s registered voters. However, Latinos represent only 14% of ballots returned.

The returns are slightly higher than in the 2014 Primary, with 38,000 more Latinos having returned ballots by Memorial Day, but they reflect a  lower turnout by about the same amount compared to 2016.

Looking just at the Democratic side, Latinos are 30% of those with ballots, but just 19% of returns.

Even if Latino turnout doubled in the coming week, from 14% to 28% of all ballots, all the way to and including Election Day, total Latino turnout would only reach 19% of ballots cast.

Many campaigns will be looking at the Latino turnout, with a particular eye on what it will mean in the governor’s race and down ticket.  As we have seen in a recent article on polling in the Governor’s race, Latino turnout has been projected by pollsters to be anywhere from 18-28% of the total vote – a wide variation, much of which is outside of past history.

The math going forward for Latino turnout begins to get tricky.

Even if Latino turnout doubled in the coming week, from 14% to 28% of all ballots, all the way to and including Election Day, total Latino turnout would only reach 19% of ballots cast. And this math will get even harder if Latinos continue to stagger in the by-mail period of voting.

However, strong Election Day turnout could get Latinos close to 2016 numbers. At the end of vote-by-mail in the 2016 primary, Latinos were only 13% of turnout.  But slumping Republican primary turnout and a high Election Day vote among Latinos helped prop those figures up to give Latinos a final 20% share of all votes cast.

Regional Strength
As was discussed in this recent Capitol Weekly article, Los Angeles County has more voters, but the San Francisco Bay has better turnout – and this is continuing in the early votes.

The Bay Area has higher rates of absentee voting, and includes San Mateo with an all-mail system, so it starts with 24% of all ballots mailed to voters, and they are currently accounting for 24% of ballots returned.  They aren’t by definition over-performing, but they are holding their weight.

We should be wary of grand statements about what it all means. We won’t have all the data for real analysis of the 2018 primary for weeks or months after.

Los Angeles County, on the other hand, starts with 19% of ballots mailed, and drops 6-points to just 14% of returns.  LA County alone has more registered voters than any other region, yet the S.F. Bay Area, Central Valley and OC/San Diego regions are all returning more ballots.

This has a significant impact on statewide contests as we have seen in past elections, with Bay Area candidates generally over-performing similarly situated Angelino candidates thanks to their built-in turnout advantage.

These early numbers are a big deal to campaigns, and they should be taking stock of the early vote, comparing the incoming votes to their projections for turnout, and tracking what areas of their district reflect their strengths and weaknesses.

But, for those of us simply observing, the analysis of these early results must be taken with a grain of salt.

We aren’t exactly sure yet which movements are due to mechanical changes in the voting process and which are signals of true voter enthusiasm. And we should be wary of grand statements about what it all means.  We honestly won’t know until Election Day, and we won’t have all the data for real analysis of the 2018 primary for weeks or months after.

Just in time to gear up for November and do it all over again!
—
 Ed’s Note: Paul Mitchell, a regular contributor to Capitol Weekly, is the creator of the CA120 column, vice president of Political Data and owner of Redistricting Partners, a political strategy firm.