Can Governors Do Much About Economy?

In a recent Data Wonk column, I suggested that the Wisconsin economy under Scott Walker has performed very similarly to what one might have expected if a Democrat had been governor, “that the Walker economy looks very much like an extension of the Doyle economy.” In response, commenter Vincent Hannah suggested:

That’s really interesting. So both are failures then right? I mean conservatives incessantly bash Doyle and the state’s economic performance while he was governor, but if it’s no worse than the state’s performance under Walker, then neither is a good governor right?

Let’s examine Mr. Hannah’s provocative suggestion. If one accepts my contention on Democrat Jim Doyle and Republican Scott Walker, does that mean that a change in governors will have no effect on Wisconsin’s future?

A possible starting point is to apply the principles of experimental design and hypothesis testing to the question: is the Wisconsin economy measurably different because of a switch in political parties? In statistics a hypothesis has two parts: a “null” hypothesis, which can be viewed as the “nothing changed” hypothesis; and the alternative hypothesis, which is adopted if the null is rejected.

The next step is to design a test for the hypothesis. Under ideal conditions, an experiment can be designed in which samples can be randomly selected from a larger population. That’s seldom practical in assessing public policies. Neither legislators nor governors are likely to be enthusiastic about researchers telling them they have been randomly selected to adopt a particular policy.

Instead, public policy evaluations are mainly observational studies, using existing available data. States may be compared to other states or to themselves over different time periods The challenge comes because states vary in ways that might color comparisons to each other. Similarly, the outside environment can radically change over time. To compare the Doyle and Walker administrations, for example, requires an adjustment for the fact that nationally the Great Recession struck during the Doyle administration while Walker entered office during the recovery from the recession.

A Walker supporter might formulate the problem this way: rejecting the null hypothesis means the Wisconsin economy has improved under Walker’s policies more than could be expected given national conditions. Conversely, a Walker critic might formulate it in the opposite direction: rejecting the null means the economy is worse because of Walker, again taking into account national conditions. Let’s look at each:

Alternative Hypothesis 1: Walker has improved the Wisconsin economy

The fundamental strategy taken in by Walker and his fans in support of this hypothesis is to ignore the national context. Typically this involves comparing recent data with data from the heart of the recession.

A secondary Walker strategy is to pick and choose which data to emphasize each month, taking advantage of the monthly variability of the data, depending on which puts Wisconsin in the best light. Thus Wisconsin’s absolute job growth might be compared to one neighbor state, while its percentage growth is compared to another state.

Another strategy is to claim credit for things in which Wisconsin excelled before Walker took office. Examples include a low unemployment rate and a well-funded state retirement system.

Finally there is the approach of ALEC, which I analyzed last week, which praises Walker’s economic policies, ranking the state 14th best in this category — even though it ranks the state  41^st in economic performance, meaning its own analysis shows these policies have been ineffective.

Alternative hypothesis 2: Walker made the economy worse.

In early 2010, a year before Walker took office, both Wisconsin and US employment bottomed out and resumed growing. During the next year, employment grew faster in Wisconsin than in the average state. Once Walker took office, Wisconsin employment growth fell to a level below that of the average state, where it has remained up to now. This change led to a belief among some Democrats that Walker’s policies caused the slowdown.

However, if one goes back to the early years of Doyle’s term as governor, one sees a pattern very similar to that under Walker: growth in Wisconsin employment lagging the national growth. In both eras, there is a continual slow decline over time in Wisconsin’s share of US employment.

By way over-promising the number of jobs he would bring to Wisconsin, Walker gave credibility to the perception his policies helped create Wisconsin’s job growth gap. However, this gap supports the notion that Walker’s policies have had no impact, not that they caused the lag in growth.

It is widely recognized that governors get too much credit when times are good and too much blame when times are bad. It might be more productive to look at the other factors holding Wisconsin back. Some, such as cold winters or the absence of mountains or ocean beaches, are inherent. Others, like a stodgy, backward-looking business community, children locked in poverty, or an under-average number of college graduates can, over a long run, be susceptible to public policy.

Although they may not account for Wisconsin’s monthly employment figures, Walker’s policies come at a cost to Wisconsin:

• There is an opportunity cost. The money lost by state taxpayers through the manufacturing and agriculture tax credits or through Walker’s refusal to expand Medicaid, for instance, could have yielded more than a billion dollars to address other state needs. For example, these funds could be used to replace lead water lines or create temporary jobs.
• The continued adherence to the idea that prosperity will result from reducing taxes on wealthy people represents the triumph of ideology over evidence. This notion continues to rear its head despite more extreme failures in Kansas and Louisiana, and the failure of the Bush tax cuts to create a booming economy. It is never healthy when ideology trumps evidence.
• By bypassing the market, government handouts like the Manufacturing and Agriculture tax credit waste public resources. One criticism is they may simply subsidize spending that would have occurred in the absence of the credit. They also help create a business culture that looks to government for help rather than better meet customer needs.
• In a dynamic market economy, some businesses grow while others shrink. A recent examination of store closing announcements from retail chains estimates that at least 50,000 employees will lose their jobs as a result. This loss is almost as much as the total number of coal miners. Yet this potential retail job loss has gotten nowhere near the same attention and sympathy as the decline in coal mining jobs. Why the obsession with coal miners? According to a report issued earlier this year by the US Energy Department, there are more people working in solar energy and wind energy than mining coal. But Wisconsin’s policies seem much more rooted in nostalgia for declining technologies and industries than those that are growing.

Wisconsin needs to move beyond nostalgia politics and nostalgia economics if it hopes to build an economy that better serves its residents.