November 14, 2006
How much are those endorsements worth?

One often hears that newspaper endorsements are not worth much in elections these days. But how much is "not much"? I think we can at least get a data point from this year's campaign.

There's a clear bifurcation in the nine statewide races other than the Governor's race, which is sui generis. Let's look at them by each group. The first group is the Republican high performers:

U. S. Senator

Kay Bailey Hutchison REP 2,658,657 61.68%
Barbara Ann Radnofsky DEM 1,554,324 36.06%
Scott Lanier Jameson LIB 97,582 2.26%

Total Votes Cast 4,310,563

Lieutenant Governor

David Dewhurst REP 2,512,197 58.18%
Maria Luisa Alvarado DEM 1,616,945 37.45%
Judy Baker LIB 188,956 4.38%

Total Votes Cast 4,318,098

Attorney General

Greg Abbott REP 2,553,610 59.50%
David Van Os DEM 1,598,378 37.25%
Jon Roland LIB 139,525 3.25%

Total Votes Cast 4,291,513

Comptroller of Public Accounts

Susan Combs REP 2,542,917 59.44%
Fred Head DEM 1,584,325 37.03%
Mike Burris LIB 150,996 3.53%

Total Votes Cast 4,278,238

Every GOP candidate here got at least 2.5 million votes, turnout was at least 4,278,000 votes, and every Democrat got less than 38%. These were the highest-profile GOP candidates, all of whom spent money on TV advertising (though I for one never saw a KBH ad; she probably needed to advertise the least, however).

Now let's look at the other half:

Commissioner of the General Land Office

Jerry Patterson REP 2,314,965 55.08%
vaLinda Hathcox DEM 1,720,985 40.95%
Michael A. French LIB 166,935 3.97%

Total Votes Cast 4,202,885

Commissioner of Agriculture

Todd Staples REP 2,304,494 54.75%
Hank Gilbert DEM 1,759,507 41.80%
Clay Woolam LIB 144,970 3.44%

Total Votes Cast 4,208,971

Railroad Commissioner

Elizabeth Ames Jones REP 2,267,047 54.03%
Dale Henry DEM 1,751,844 41.75%
Tabitha Serrano LIB 177,315 4.23%

Total Votes Cast 4,196,206

Justice, Supreme Court, Place 2

Don Willett REP 2,124,205 50.93%
William E. Moody DEM 1,876,845 45.00%
Wade Wilson LIB 169,777 4.07%

Total Votes Cast 4,170,827

Presiding Judge, Court of Criminal Appeals

Sharon Keller REP 2,343,845 56.64%
J.R. Molina DEM 1,794,432 43.36%

Total Votes Cast 4,138,277

Here, every Democrat exceeds 40%, partly due to getting at least 100,000 more votes than the top performer in the other group, and partly because every Republican gets between 200,000 and 400,000 fewer votes than their counterparts. I'd mostly credit that effect to the lack of TV exposure in these races - nobody was able to get on TV and really persuade some low-information folks to push the button for them. This I think is a better reflection of the base vote for each party. It's largely on par with the 2004 results, when the downballot statewide races were as obscure.

The two that stand out are JR Molina and Bill Moody. Both were on the ballot in 2002 (and again in '04 in Molina's case), so they might have had a bit of name ID advantage over their peers. Molina also benefitted from not having a Libertarian on the ballot to potentially suck up anti-Republican votes.

Then there's Bill Moody. What's the one thing he had in this race that nobody else did? A long list of newspaper endorsements from all parts of the state. He wound up with an average of 133,000 more votes than his peers in this group who were also in a three-person race, while Don Willet got 171,000 fewer votes. That's a 7.5 point swing towards Moody, which you can see in the percentages - he got a bit more than three points more than his peers, and Willet got about four points less than his. Perhaps there were other factors in play here - maybe Moody's walk across Texas generated more positive press for him than I'm crediting him with - but based on this, given that advertising was not a major factor in this race, I'd say the answer is that all those endorsements were worth about seven points. And that's not too shabby.

Posted by Charles Kuffner on November 14, 2006 to Election 2006 | TrackBack

That's an interesting analysis, Kuff. I suspect newspaper endorsements wouldn't be worth as much in the top-tier races, where the candidates are better known and advertising plays a bigger role. But as Texas becomes less red and more purple, endorsements could start playing an important role in these down-ballot races.

Molina also benefitted from not having a Libertarian on the ballot to potentially suck up anti-Republican votes.

Given what I know about Libertarian philosophy, I doubt Libertarian candidates "suck up" more anti-Republican votes than anti-Democratic ones. Molina may have benefited, but probably didn't gain a net benefit.

In theory, Libertarians agree mostly with liberal Democrats on non-economic issues such abortion, gays, civil liberties, and (most well-known) drug policy, and they agree mostly with Republicans on economic issues such as taxes and business regulation (although a big exception to this rule is guns). But my experience has been that they are far more forgiving of the heresies they percieve on the Right than the ones they perceive on the Left. Remember, when Ron Paul joined a major party, he chose the GOP.

Individual candidates will of course vary, but I suspect the presence of a Libertarian on the ballot usually ranges from a wash to actually helping the Democrat.

Posted by: Mathwiz on November 14, 2006 12:51 PM

Mathwiz - Some cursory analysis so far indicates that the Libertarian vote was higher than usual this year in Texas. My hypothesis is that there were a certain number of normally-Republican voters who were willing to join in on the anti-GOP wave of this year, but were not willing to vote for a Democrat. That's what I meant in my remark about Molina. I should have been more specific.

Posted by: Charles Kuffner on November 14, 2006 1:40 PM

Bifurcation? Sui generis? Wow, I thought I read this to escape from law school!

Posted by: Justin on November 14, 2006 9:36 PM