Minnesota and Wisconsin: a Natural Experiment

One of the things I teach here at Plymouth is a MA seminar on methodology and research design (for the MA in International Relations, but that’s another story). I’m in my tenth year on it, and have always enjoyed it, because I’m a bit of a geek when it comes to tradeoffs inherent in any design choice, and especially the lengthy philosophical conversations we have surrounding epistemology and ontology early in the term. Yesterday’s philosophical conversation surrounded whether or not I was right to ignore Special Branch’s invitation at an airport a few years ago to essentially spy on my students, which was remarkably germane to the seminar. We were discussing the ethical considerations involved in covert participant observation contrasted with the issues of reliability and validity encountered by overt p.o., and we ended up there.

The classic experiment as research design is rare in the social sciences, though at least in political science this has been changing quite  impressively in the past ten years or so; I was fortunate enough to serve as a discussant at the MPSA a couple years back where all four of the papers relied on experimental research design.

This piece ran in the NYT five days ago, written by a political scientist at Minnesota, and received some play nationally. I’m sure most LGM readers are aware of it. As it’s written by a political scientist, I tend to give it more benefit of the doubt regarding the validity of the comparisons being made. It has the superficial appearances of a natural experiment: two upper midwest states, similar political cultures, recently history, and recent voting patterns (each have voted Democrat every Presidential election since 1988), yet have taken divergent paths recently at the state level. We all know about Wisconsin and Scott Walker, but less trumpeted is Minnesota. Minnesota appears to be kicking Wisconsin’s ass:

Three years into Mr. Walker’s term, Wisconsin lags behind Minnesota in job creation and economic growth. As a candidate, Mr. Walker promised to produce 250,000 private-sector jobs in his first term, but a year before the next election that number is less than 90,000. Wisconsin ranks 34th for job growth. Mr. Walker’s defenders blame the higher spending and taxes of his Democratic predecessor for these disappointments, but according to Forbes’s annual list of best states for business, Wisconsin continues to rank in the bottom half.

Along with California, Minnesota is the fifth fastest growing state economy, with private-sector job growth exceeding pre-recession levels. Forbes rates Minnesota as the eighth best state for business. Republicans deserve some of the credit, particularly for their commitment to education reform. They also argue that Minnesota’s new growth stems from the low taxes and reduced spending under Mr. Dayton’s Republican predecessor, Tim Pawlenty. But Minnesota’s job growth was subpar during Mr. Pawlenty’s eight-year tenure and recovered only under Mr. Dayton.

Ideally, this could be turned into a proper study (if it isn’t already by someone somewhere), with precision on demographic variables to ensure the suitability of the comparison (which of course makes the design more akin to a quasi experiment), and then the various outcome metrics. I’d tackle it if I had the time, but that’s a precious commodity at the moment (and I have two new papers I need to write by April, neither of which have so much as a word applied to them beyond the seemingly good ideas that generate impressive sounding conference proposals). I’m intrigued, however.

Incidentally, Wisconsin also incarcerates at a much higher rate:

So, here’s the essential story (as detailed in the chart that appears after the jump): Wisconsin incarcerates many more people than Minnesota, while Minnesota puts many more individuals on probation.  The two states have about equal levels of crime, and Minnesota actually has a larger percentage of its population under supervision (that is, either incarcerated or on probation or parole release).  However, because incarceration is so much more expensive than community supervision, Minnesota’s corrections budget is much smaller than Wisconsin’s (about $99 per resident, versus Wisconsin’s $234 per resident).  Given the similarity of the two states’ crime rates, it appears that Minnesota’s probation-based strategy is delivering more bang for the buck than Wisconsin’s.

Ah, and Happy Thanksgiving to our American-based readers, from the original Plymouth. This is the 13th Thanksgiving I’ve experienced abroad. I’ve replicated it here in England a couple of times, participated in the annual Plymouth festivities a couple of times (yes, Plymouth marks Thanksgiving in its way, including a thing down at the Mayflower Steps), and tonight I’m going to a Thanksgiving Dinner hosted by the newish (American) Dean of Students here at the Enterprise University. So it’s almost the same. Without, you know, the four to five day weekend, or watching the Detroit Lions lose a football game.