California VBM Primary Return Indicate Democratic Advantage

Through Friday, 1.1 million VBM ballots were returned for the June 3 primary. In 2012, 2.2 million VBM ballots were returned prior to election day. Another 1.2 million VBM ballots were returned on election day and 1.9 million voters voted in precincts on election day.

I have the ballot breakdown by political party for the 2.2 million pre-election day ballots. The breakdown for those 3.1 million subsequent ballots exists, but I don’t have it. The first 2.2 million were 43.9%D/38.4%R/17.7%O. It’s widely believed that VBM ballots cast on election day and precinct ballots are more Democratic than the early VBM ballots. There are two reasons for this 1) the vote is more Democratic than the VBM returns would suggest 2) Democrats gain in post-election day counting. That’s exclusively ballots turned in or cast on election day.

Since the ballots turned in already are the same VBM as the 2.2 million 2012 ballots we have data for, we can do an apples to apples comparison. The 1.1 million turned in so far are 44.2%D/34.9%R/20.9%O, a significant Republican drop that’s gone almost exclusively to DTS and minor party voters. That means that the Democratic ballot return advantage in 2014 is 9.3% over Republicans, a jump from 2012’s 5.5%.

Why has the Democratic advantage gone up? Will it hold for the remaining ballots? What does this mean for the primary results? What does this mean for November?

Why has the Democratic advantage gone up and will it hold?
Since I don’t know have day by day data for 2012, I can’t tell you whether Democratic return is up or Republican return is down. The chart below shows that Democrats have returned 51% of their 2012 total, while Republicans have returned 46%, and others have returned 60%. We’ll have a better idea when all the ballots are turned in whether Republicans are down or Democrats are up.

One possibility is that this is just an aberration and Republican ballots will catch up as we get closer to the election. Democrats have been trying to emphasize getting their voters VBM ballots earlier. The reasoning for this is that once a voter gets his ballot in, you don’t have to spend time and money trying to get him to do so. You can focus on those other voters.

I dismiss this because the daily returns have been fairly consistent and not getting more Republican and because there’s been an increase in DTS/third party voting. Democrats won’t be pushing DTS voters to return their ballots because they don’t know who they’ll be voting for and because they aren’t likely to respond to Democratic pitches. I’m inclined to believe this is a real change that will hold through the primary and result in a better Democratic primary results.

Could this be a consequence of the increased Democratic and DTS voters as a result of the October 2012 online registration push? That’s part of it. Since May 2012, there are 235k more Democratic voters and 416k more other voters, but 145k less Republican voters. That makes it sound like people are leaving the Republican party, but really it’s due to a purge of inactive voters. Since the October 2012 online surge, Democrats have shed 288k voters, while Republicans have shed 315k.

If we assume Republicans have the same return rate as they did in 2012, 16.5%, we’d see that Democrats index 4% higher and DTS/other index 16% higher. So the change in number of voters is likely part of it, as Democrats are only up slightly. In fact, the Democratic increase, compared to Republicans, is nearly negligible.

It’s likely that the other part of the DTS surge is the intended consequence of the top two primary. Previously, they couldn’t vote for major party candidates and now they can. Their votes matter more, so more people are becoming involved. Their return rate still doesn’t approach Republicans and Democrats but it is higher.

What does this mean for the primary results?
Better Democratic results than 2012. That’d be a relief to Democrats, who famously got burned in the CA-31 primary because more Republican leaning voters voted. We see that’s reversed. In 2012, Republicans got 51.5% of the vote. If that’s 49.5% in 2014, the Republican chances of getting both top two spots go down.

One other races where the percentage of the Democratic vote could come into play is CA-17. In CA-17 Ro Khanna is hoping to finish top two along with fellow Democrat congressman Mike Honda. He stands a better chance if Republican voters are lower in the primary. They are, but so are Democratic voters. Since Honda is expected to win Democrats handily, the less Democrats there are the better it is for Khanna.

What does this mean for November?
It could mean a better than expected November for Democrats, but I’m not inclined to think so. In prior years we’ve had a big disconnect between primary and general election results, with Democrats doing a lot better in the general. This was even more pronounced in 2012. That occurred partially because of the additional voters who registered online between the primary and general election. There’s no such change like that on the horizon. I think the Democratic improvement from the primary to the general won’t be nearly as big in 2014 as it was then. I’d want to see increased Democratic ballots for the November election before drawing that conclusion.