City Election Turnout

Doc Murray on turnout predictions for the 2009 city election here in Houston:

... turnout in 2009 will be significantly below our 300,000 voters in 2003. How much lower? I think in the 225,000 to 250,000 range. This means campaigns have to carefully target mail, electronic messages, and phone bank efforts at a shrunken electorate. How to do that? Start with the 125,000 persons who voted in the 2007 General Election. If they came out when Bill White had no meaningful competition two years ago, they will very likely come back to the polls this year, if they are still on the city voter rolls. Then, try to figure out who the other 125,000 likely suspects are by a variety of means.

I noted earlier that the early polling certainly indicated that a drop in turnout would be among the possibilities of what to expect in November. In fact, I think it's about the highest certainty we can expect among all the variables.

I tend to agree with Murray's turnout model, but it's for reasons such as this that I actually miss George Strong's analysis. He's been spot-on in much of his city election commentary in prior years.

In reviewing some numbers from elections past, one thing that jumps out is that the turnout by Council District doesn't really move all that much from election to election. For instance, traditionally Republican District G will have an open seat election and possible runoff. All of the Mayoral candidates will give the district ample courting in the hopes of either earning a plurality in November that may turn into a majority in December, or at least avoiding a wipeout in the district. But the turnout in 2003 was 17.3% of the ballots cast in November 2003 and 18.5% in December 2003. In every election since 2001, it has been between 16.4% and 17.4% of the vote.

Where you tend to see major alterations in the turnout model are in non-Mayoral election runoffs. In 2005, District B went from 8.4% in November to 14.3% in the runoff, while District C went from 15.0% in November to 27.8% in December. Meanwhile, in 2007, District D went from 15.4% in November to 30.0% in December, while District E went from 12.0% in November to 24.9% in December.

In short, open seat Mayoral contests flatten the playing field. The three strongest GOP seats (A, E, & G) will garner a collective 39-40% in both November and December. The two African-American districts will represent about 24-25% in both elections. The two Hispanic districts will represent about 15% in both elections. And the two others (C and F) clock in at about 20% both times.

There's still an argument that the makeup of voters within each district changes from November to December and also changes depending on whether it's an open seat district race versus an incumbent running in a hot Mayoral year. I haven't seen a great deal of analysis on that yet, so I merely throw that out for consideration and discussion.

What I will suggest is that the emphasis in local elections to make a particularly concerted appeal to GOP voters may be a bit overblown. I think there's enough of an audience available that can be appealed to on the basis of what I'd call a "civic-minded" basis. And that universe probably would include a fair amount of people who vote Republican in even-numbered Novembers. But that's not the same as suggesting that a candidate has to take a particularly heavy-handed line on public safety issues, immigration, or whatever current GOP hot-button issue of the day that rightwing radio is chattering about. If we've learned nothing else from the Zaf Tahir campaign, I think that'd be on the short list of lessons learned.