Monday, April 11, 2011

Election Rumination

In the aftermath of last weeek's Supreme Court vote, observers differed as to whether it reflected a significant Walker backlash. My initial impression was that there is little evidence of that. Having looked closer at the numbers, I still think that's the case although, in the end, a final judgment may require numbers we don't have.

One line of argument is that Kloppenburg and Prosser should never have been close. But all that tells us is that the race was transformed into a partisan referendum. With or without an anti-Walker backlash, a partisan referendum will be close because Wisconsin is a sharply divided state. The fact of the matter is that neither candidate had any significant name recognition going in (very few people know who is on the state supreme court)and, once a judicial election becomes salient (i.e.. people start to pay attention), incumbency isn't the advantage that it might be in other types of races. Up until recently, of course, the races rarely became salient. This one became more salient than most.

One argument looks at differences from the February primary to the April general but that is not all that helpful. Turnout in April was wll over three times what it was in February. It was a completely different electorate in April, i.e.. different people were voting,and there is reason to think that the differences were not neutral between Prosser and Kloppenburg.

The nature of those differences was central to the fears of Prosser's supporters coming into last Tuesday. It looked like a perfect storm was aligned in Kloppenburg's favor. In a relatively low turnout election, you had a well organized special interest in an almost unprecedented state of excitement. That special interest is well known for its ability to run the ground game, i.e., get out the vote. (Conservative candidates in spring judicial elections, for example, hate running when there is a race for the Superintendent of Public Instruction.) If additional turnout was comprised disproportionately of members of public employee unions, the race could be flipped from Prosser to Kloppenburg even in the absence of a more general anti-Walker backlash.

To some extent, that is what happened. Dane County may have set a record for turnout in a spring judicial election. It turned out at approximately 82% of its November 2010 turnout and at almost three times the level for the Butler-Gableman race. It went 73% for Kloppenburg. Turnout was up in other counties as well and three counties with heavy student populations (Portage, Eau Claire and LaCrosse) were up as sharply as Dane. While Dane was around 10% of the vote in November, it was 12% in April. Had I known only this, I would have thought Prosser was dead.

What got in the way of the Kloppenburg win was that increased turnout in other counties, while not as strong as Dane, was not limited to disgruntled public employees. The collar counties around Milwaukee County did not come out like Dane - just around 67% of November's vote and a little over twice the numbers for Butler-Gableman - but they came out and went as heavily for Prosser as Dane went for Kloppenburg.

The other factor is that turnout in Milwaukee was up only as much as in the state generally. It was not whipped into the frenzy that gripped Dane and did not go as heavily for Kloppenburg (about 57% as opposed to 62% for Barrett) as one might have expected.

Without Dane, Prosser wins with 52.6% - closer than it might otherwise have been but hardly indicative of a statewide Walker backlash. Of course, one might just as easily back out the heavily Republican WWOzCo. When you do that and take out the populous and heavily Democratic Milwaukee, Prosser wins by around ten thousand votes.

One could say this is evidence of an anti-Walker move. If you back those five partisan counties out of the November vote, Walker still wins handily. So Kloppenburg did significantly better (a bit over five percentage points) in the remaining 67 counties.

The problem with such a conclusion is that, even though turnout in the Supreme Court race was heavy, it was still only about a third of the November turnout. It would be interesting to know how the composition of the electorate differed. If turnout was more heavily composed of union members and Democrats, the backlash story gets more complicated.

A counter-narrative is that conservatives took everything the left has to give and survived. The budget repair bill will become law and Prosser was reelected to the Court. The recalls may change that narrative but they may also confirm it.


[Redacted] said...

Without Dane, Prosser wins with 52.6%...

Saying someone wins by X number of votes if you exclude this or that county is a specious argument, and you know it. It's just as useful for me to say that Barack Obama would have won Wisconsin in a landslide if not for Waukesha, Washington, and Ozaukee counties. But, guess what? Like it or not, those counties are part of Wisconsin. And given that Supreme Court justices not named Butler ordinarily win in a walk (look at Chief Justice Abrahamson), even your phony 52.6% is a lot closer than one would expect in an off-year election.

In any event, I wish you would stop with the lowest common denominator conservative trope about how things would be different if only Dane County didn't exist (I know you didn't say that, but it's implied). It happens to be home to a growing biotech industry that brings jobs and revenue to our state - normally things that people of your stripe appreciate. Unless that area has a liberal tilt in elections. I'd expect such false distinctions and hand-wringing from Charlie Sykes and Mark Belling, but it is beneath you.

illusory tenant said...

Observes the Shark, somewhat incoherently: "It was a completely different electorate in April [compared with February]." But see: "It would be interesting to know how the composition of the [April] electorate differed [from that of November]."

So you've divined how April vs. February differed, but you have no idea how April vs. November differed?

In light of this fundamental discrepancy in your reasoning, one might be moved to wonder: If it's invalid to compare the April demographic with the February demographic, then why is it any less invalid to compare the April demographic with the November demographic? Because January (which marked the official commencement of Walker policy implementation) occurred in the meantime. You cannot discount that factor, nor the rising awareness of the Fitz Van Walker administration's (and especially its supporters') apparent ... skepticism, shall we say, toward the rule of law.

As for your picking and choosing among counties, which [Redacted] is correct to explode, if Milwaukee County. (41.6%) had turned out even as few more voters as would total two-tenths of a percentage point below the State average (43.2%), your pal Mike Gableman would be rolling out the RedState carpet for his new colleague Justice Kloppenburg on 1 Aug.

David said...

Verdict: Inconclusive.

Time (and results) will tell.

George mitchell said...

Pretty grumpy material from redacted and tenant. I don't have an objective analysis of the vote. I am content to be delighted with the result.

Anonymous said...

You focus on Dane County, but consider the outcome if you took out the tree nut-bag counties instead -- Waukesha, Washington, and Ozaukee.

The result: Kloppenburg in a landslide.

Dane is as Democratic as the three nutbag counties are Republican. And this is somehow a knock on Dane County?

[Redacted] said...

Pretty grumpy material from redacted and tenant.

I'm not being grumpy, but calling out the professor on his use of the ridiculous, Palinesque "real American/Wisconsinite" division that doesn't carry any rhetorical water. Implying that residents of Dane County aren't "real Wisconsinites" is not only wrong, but offensive. And I don't care much liberals complaining about W-O-W Counties either. We're all Wisconsinites. Period. The end.

Also, I can't help but note that you're commenting on an election while your handle is the name of one of the guys who went to prison over Watergate. Surely it's a coincidence, but I find it a hilarious one nonetheless.

illusory tenant said...

Thanks, Mr. Mitchell. I had been aware that numerals and facts had a clear liberal bias but was heretofore unapprised of their emotional content. Anyway, not having a dog in the fight, I'm not particularly grumpy at the present outcome. Perhaps you missed this. I recommend it.

Geo Mitchell said...

Redacted, my "handle" is my name. Check your watergate history, btw.

illusory tenant said...

"So you could argue that Kloppenburg lost the election not in Waukesha, but in Milwaukee County, which failed to deliver her the votes she needed." -- Chas. Sykes

Well I'll be a medium wave howler monkey's uncle. I only been saying that since February 16.