Politico’s essay on conservatism in Mesa, Arizona is God-awful

So Joseph Cordes got me up this morning with this piece from Politico, which is really nice in the way it talks to residents and really awful in the way it talks about cities: Are Conservative Cities Better?

We start off with Mesa, Arizona. Yay, conservatism! The part where he covers the politicians bragging about themselves and their city is fine. Whatever. Standard bloviating.

Only, whatever. Here’s the first howler:

While it’s willing to make investments, Mesa is also lean in ways that more bloated liberal cities can’t boast. Take the City Council. Despite Mesa’s hefty population, council members are part-timers who have day jobs in fields from education to copper mining. City leaders also pay themselves considerably less than those in other cities do. Mesa City Council members make only $33,000 a year, and the mayor is paid only $73,000. (And those salaries represent the fruits of a big raise: Before last year, city councilmembers made less than $20,000 a year and the mayor earned only $36,000.) By contrast, as of 2012, in similarly sized Fresno, the mayor made $126,000; city council members brought home nearly $65,000. In neighboring Phoenix, meanwhile, the mayor makes $88,000 and city councilmen earn more than $61,000.

Read more: http://www.politico.com/magazine/story/2014/09/mesa-arizona-are-conservative-cities-better-111069_Page3.html#ixzz3DlvKrqfY

Mesa has 450,000 people in it. Phoenix has 1.5 million. And the higher salary of the folks in Fresno? Fresno in the original study has no policy preference; it’s governed by both, and it’s in 15 cities more conservative than the remaining 50 shown. Phoenix is among the top 25 most conservative in the rankings. He’s done some research, but he’s not thinking , nor is he really using the study that prompted him to start writing in the first place. These comparisons could actually be interesting: Mesa’s median home value is $270K, Fresno’s is $170k. Those Fresno salaries do seem pretty far out of line, but Fresno policy mix is where it is in ranking–more conservative than most cities.

The next howler:

As the great, Democratic-run cities across the country—Chicago, Detroit, Los Angeles—face fiscal calamity, America’s conservative cities are showing that there’s another way.

Read more: http://www.politico.com/magazine/story/2014/09/mesa-arizona-are-conservative-cities-better-111069.html#ixzz3DlwYy1ks

Ok, well, yes, those “Democratic Run” cities. Except for those are metropolitan regions, cherry picked for the fact that they have had money trouble, and yeah, they are solidly Democratic, but there are many solidly Democratic cities that are doing just ducky. Can I name some? Um, yes: When was the last time New York had a Democratic mayor? Then there’s San Francisco, Portland, Boston, Philadelphia, Seattle, Milwaukee, St. Paul, Minneapolis,etc.

Look, I don’t have to do that work. Politico is trying to make a story out of something that isn’t really a story. Why isn’t it a story? According to Tausanovitch and Warshaw’s original measure of conservative policy mix that prompted the story, there are 11 cities that are actually negative in the measure of conservatism in their ranking, and the distribution of the index scores is solidly positive. Now, they have data that goes past the rankings, but we have an index measure. On the ranking there are about 40 other cities, all leaning more liberal but ranked against each other. So generalizing that those liberal cities are ‘in trouble’ has no basis in these data.

I’ve read the original paper, it’s very interesting and the methods are promising, but what is currently posted is a draft. I’d wait for the full paper–the authors themselves haven’t even figured out what their data mean. They are trying to figure out if voter bases prompt municipal governments to reflect those in a policy mix. They’ve been able to show that there is a difference in federal and urban policy mixes, but I don’t see where they show the differences among cities are significant, and if that’s true, then we’ve established that people have differences in their preferences for federal policy but somewhat less variation in their preferences about local government policy, which theory leads us to suspect. The methods are cool, but the paper is sprawling, undisciplined thought piece right now, and they haven’t digested their results yet. But with that kind of imbalance in the groupings with a tortured index variable, it’s going to be very hard to show what the authors are seeking to show.

And none of it evaluates whether a city is well run or not, or whether a city is a fiscal trouble.