Yes, Minnesota Beats Wisconsin
But is its better economic performance due to more liberal policies?
During his first campaign for Wisconsin governor, Scott Walker famously promised he would create 250,000 private sector jobs by the end of his first term in January 2015. As is well known, Wisconsin fell far short of Walker’s goal. In the next four years, Wisconsin only added 140,000 private sector jobs.
Today, more than 3 years late, Wisconsin finally seems poised to hit the 250,000 jobs mark. By comparison, Wisconsin’s neighbor Minnesota, with a slightly smaller population, crossed that line in January of 2017.
Walker’s plan for job growth reflected conventional conservative dogma. It included tax cuts, particularly at the high end, looser environmental regulation, cuts in education funding, and weakening public and private sector labor unions. However, Wisconsin avoided the fiscal crisis experienced by some other conservative-dominated states–such as Kansas, Louisiana, and Oklahoma–which drastically cut taxes in the belief that increased economic activity make up the difference.
Over the same years that Walker has been Wisconsin governor, liberal Democrat Mark Dayton has been governor of next-door Minnesota. This coincidence gives rise to two questions: (1) which of the two states faired better economically, and (2) to what extent can any difference in results be attributed to each state’s policy as opposed to other differences between the two states?
If one accepts Scott Walker’s assumption that job growth is by far the most important measure of effective policy, there is no question that Minnesota has been more successful since 2011. The graph below compares the percentage private sector job growth since the two governors took office.
A recent report by David Cooper of the Economic Policy Institute examines the comparative statistics on the two states. Besides job growth, on almost every measure, including the growth in wages and household income, reducing the wage gap between men and women, reducing poverty, expanding health insurance coverage, and economic growth, Minnesota outperforms Wisconsin. Cooper’s comparisons join previous ones (here, here, and here) but offer much more detail.
Ironically, Minnesota’s better economic performance is tacitly confirmed by Walker’s spokesperson. In a recent article describing Cooper’s study, the Journal Sentinel’s Craig Gilbert quotes the Walker response. Walker administration spokeswoman Amy Hasenberg makes three points:
- That Wisconsin’s latest unemployment rate is lower, at 2.9 percent in March, than Minnesota’s, at 3.2 percent.
- “Wages are up, and more people are working than ever before under Governor Walker’s leadership.”
- “Our business and tax climates have also dramatically improved from where they were in 2010.”
It is telling, I think, that only the first point is a comparative statistic. Before getting to the difference in unemployment rates, let’s look at the other two arguments.
The second point above typifies an argument that Walker commonly makes: comparing today’s numbers to those at the depth of the recession. These statements could be made about almost every state and tell us little about the Wisconsin economy — and offer no rebuttal to the idea that Minnesota has done better.
As to the third point, it is certainly true that with Walker’s ascension, Wisconsin got a boost in various ratings of business and tax climate. But these ratings do not seem to translate into economic performance.
For example, the table below shows Minnesota’s and Wisconsin’s ranks from the 2018 Rich States Poor States report published by the conservative American Legislative Exchange Council (ALEC). The Outlook score reflects ALEC’s agreement with each state’s policies. Clearly it prefers Wisconsin’s policies to Minnesota’s. Yet when it comes to actual performance, the positions of the two states are reversed.
The only actual data point cited by Walker’s spokesperson is the slightly lower unemployment rate. The next graph compares the two states on four measures of unemployment: the official rate (people who are unemployed and actively seeking a job), U-4 and U-5 (which include people who want a job but are not searching actively, and U-6 (which includes part-time workers who want full-time employment. It is striking how little difference there is in the current numbers for the two states.
However, as the next graph of three unemployment indices shows, this convergence is recent. For most of this period, unemployment, however measured, took longer to come down in Wisconsin than in Minnesota. This means that more Wisconsinites than Minnesotans are suffering the after-effects of long-term unemployment. Using Census data, the EPI calculates that in 2017, 22.4 percent of Wisconsin workers who were unemployed had been unemployed for at least 26 weeks, compared to 12.6 percent in Minnesota.
Over the whole period, both states lost more residents to other states than they gained. However, Wisconsin lost about twice as many as Minnesota. This suggests that one possible factor in Wisconsin’s low unemployment level is that some number of the unemployed have decided to leave Wisconsin for greener pastures.
The Economic Policy Institute makes a compelling case that on practically all fronts Wisconsin has underperformed Minnesota over the past seven or eight years. The challenge is to make the next leap—to conclude that the cause is Governor Dayton’s more liberal policies.
To his credit, Cooper acknowledges this limitation: “Precisely analyzing the causal impact of a particular policy requires sophisticated statistical methods … This report does not provide that level of causal analysis.”
Similarly, a skeptic of the view that Wisconsin’s and Minnesota’s differing policies account for their differing success could point to several other differences between the two states. These include the fact that Minnesota’s major university and capital are both located in its largest metropolitan area, Wisconsin’s greater dependence on manufacturing during a period when American manufacturing took a beating, Milwaukee’s proximity to Chicago, and a population in the Twin Cities metropolitan area that’s more than twice as large as Milwaukee’s.
It is also notable that Minnesota’s superior performance started before Governor Dayton’s inauguration in 2011. It continued during his first two years even as a hostile Republican-dominated legislature rejected most of his proposals.
Arguably Walker’s best defense would be to adopt the widespread consensus among economists that government policy has a limited influence on the economy. (See this opinion piece by an economist at the University of Minnesota arguing that Minnesota’s better outcomes mostly reflects the differences between the two states, and that policy differences probably play a minor role.) However, it seems unlikely that Walker would take this advice; he is much more comfortable insisting, despite the evidence, that his policies created jobs in Wisconsin.
Would “sophisticated statistical methods” put to rest the question of the relative role of policy versus that of inherent state difference? Theoretically yes, but only if the political and logistical obstacles to randomly selecting two samples from the Wisconsin and Minnesota populations could be overcome, one of which would be administered using Walker policies; the other using those adopted by Dayton. In other words, such an analysis is unlikely in the foreseeable future.
What conclusions can be made from comparing Wisconsin and Minnesota outcomes? The answer probably depends on where one lands on the question of how much of the difference in outcomes is due to public policy and how much reflects inherent state differences. Consider the two extreme points. If public policy dominates, the conclusion is that the Dayton set of policies lead to better outcomes than the Walker set of policies. Conversely, if the performance gap is due to other differences between the two states, a state can safely choose the policy that leads to a more just society, with no negative impact on the economy. Either choice doesn’t help Walker’s cause.