Before I begin, please note that all of this involves a lot of back-of-the-envelope math and more than a few assumptions made along the way. The number of registered voters in Harris County, and thus in Houston, will certainly change between now and November. And of course, every election is different, with candidates and exogenous conditions having an effect on who does and doesn’t vote. So keep all that in mind as we proceed.
We have an open seat Mayoral race this year, and history says those are our biggest ones. For context, here are the Harris County turnout numbers from the last three open Mayoral races:
Year Votes Registered Turnout
===================================
2015 268,872 979,401 27.45%
2009 178,777 935,073 19.12%
2003 298,110 955,205 31.21%
I should note that in 2003 there was the Metro referendum, and in 2015 there was the (unfortunate and infuriating) HERO repeal referendum, in addition to four Harris County ballot propositions. Having a high profile referendum on the ballot helps drive extra turnout. We will have a couple of Houston referenda on the ballot this November so the higher end is more likely to me, but there is a range and a lower end is possible as well. Unlikely, to my mind, but I want to be transparent here.
Now let’s look at Houston voter registration numbers. I’m limiting myself to Harris County, which accounts for about 97% of Houston election totals. I don’t have this data for every year, but we can do with a sample:
Year Houston Harris Share
===================================
2003 955,205 1,786,767 53.46%
2009 935,073 1,881,112 49.71%
2015 979,401 2,054,717 47.67%
2018 1,082,506 2,307,654 46.91%
2019 1,085,813 2,329,277 46.62%
2022 1,133,155 2,543,162 44.56%
As noted, there will be a different voter registration figure this fall, likely a bit higher for each but with more growth in the non-Houston part of Harris County. This is close enough for our purposes.
So what does this suggest for 2023 turnout? The main thing I want to point out is just that we have more voters in Houston now than we did in 2015. There hasn’t been nearly as much growth in Houston as there has been in the non-Houston parts of Harris County, but it’s still up over 150K from 2015. As such, if we have the same rate of turnout as we did in 2015, we would see over 311K Houston voters from Harris County, or more than a 40K increase from 2015. The Fort Bend and Montgomery share will likely be in the 5-6K range total, but even without that we’d easily have a new high total for a Houston election. If we have the same turnout rate as we did in 2003, we’re looking at over 353K voters from Harris County. That’s an enormous increase over 2015. We only need 26.3% turnout in 2023, based on the 2022 registration numbers, to equal the number of voters from 2003.
What then is the argument for getting the higher end of turnout this year? There will be a couple of referenda on the ballot, and maybe possibly an HISD bond referendum. There’s already a lot of money in the campaigns, which if nothing else should mean plenty of advertising. The argument that compels me is just simply that the last four even-number elections we’ve had have featured historically high turnout. We broke the record for total numbers of voters in any election in 2016, then shattered it in 2020. We destroyed the record for turnout in a non-Presidential year in 2018, and then even though turnout was down relative to it in 2022, it was still easily the second-highest off-year total, both in absolute numbers and percent of registered voters. We’re used to higher levels of turnout now. Why shouldn’t that continue in the first open-seat Mayoral race since then?
Like I said, I’m making some assumptions here, any or all of which could be wrong. My point is simply that it would not take anything like a historic turnout percentage to produce a record number of total voters, just simply because there are more voters now. Another way of looking at this is that there will be a lot of people who haven’t voted in a prior Mayoral election but will be participating this year, both because of the increase in registrations and the usual comings-and-goings over the years. Campaigns maybe ought to take that into account.
