Border blues

Minnesota’s border counties experience the competitive disadvantage of high taxes and hostile regulations

For many, it’s just common sense that Minnesota’s high taxes and burdensome regulations weaken the state’s economy. Excessive taxes and regulations impose extra costs on businesses, which makes it more expensive to create and expand a business in Minnesota.

Nonetheless, many Minnesotans don’t see a problem with the state’s taxes and regulations. For them, high taxes allow the state to invest in the public infrastructure businesses need to grow and to support a high quality of life that makes the state attractive to employees. While these people might admit taxes can change behavior, they don’t believe any negative impact outweighs the positives.

The challenge for both sides of this debate is marshaling the evidence on the true impact of Minnesota’s tax and regulatory policy.

Despite an extensive library of academic literature on the connection between state-level policy and economic growth, the results of research often conflict and do not lead to firm conclusions. Comparing and measuring the impact of state policies is difficult because there are so many factors outside specific state policies that influence economic growth, such as a state’s climate, culture, natural resources, and proximity to markets.

To help control for these confounding factors, more and more researchers employ methodologies that compare differences along state borders. In general, the climate, culture and other factors outside a state’s control are similar in counties on either side of a border. With similar conditions, any economic differences are more likely attributable to government policies.

Using this methodology, the academic literature is uncovering a stronger relationship between state tax and regulatory policy and economic growth. Moreover, taking a closer look at employment growth along Minnesota’s border reveals growth is strongest where taxes and regulations are less burdensome.

Why study border counties

Identifying and measuring the degree to which state policies impact economic growth is much harder than people might expect. The main problem is that it is difficult to isolate state policy from all the other factors that influence economic growth. Economists use various statistical tools to account for variables like weather, population density, educational attainment and historical growth patterns. However, the results of any economic study can change dramatically depending on the methodology used to control for outside factors.

To understand just how difficult these questions can be, it is instructive to consider a critique of research on the impact of right-to-work policies. Right-to-work is a state policy that forbids compelling workers to join a union. Many people feel strongly that compelling unionization raises employment costs and thereby discourages growth in unionized industries. In support of this position, the states with the highest growth in manufacturing jobs are right-to-work states located mainly in the South. Simple economic regression analyses tend to confirm a strong connection between right-to-work and employment growth.

However, in a broadly cited 1998 paper, Thomas Holmes—an economist at the University of Minnesota—explained how these statistics reveal “little about the effects of state policy” because they “ignore a serious identification problem.” The statistics don’t account for how the “right-to-work states systematically differ in a number of geographic characteristics from the non-right-to-work states.” Holmes identified a number of factors beyond right-to-work policies that explain the growth of manufacturing in the South. The productivity revolution in agriculture freed more workers in the South to transition to manufacturing. Substitution of trucking for rail transportation diminished the value of Midwest rail networks to manufacturers. The lack of a historical union presence made it easy to pass right-to-work laws in the South. The advent of air conditioning made the South a much more attractive climate to live and work.

To solve these identification problems, Holmes compared the growth in manufacturing employment in counties on both sides of the border between right-to-work and non-right-to-work states. Here’s the benefit to this approach, according to Holmes:

At state borders, the geographic determinants of the distribution of manufacturing—for example, climate, soil fertility, access to transportation, and the level of agglomeration benefits—are approximately the same on both sides of the border. What differs at the border is policy. To the extent that the pro-business policies pursued by the right-to-work states have been a factor in the migration of industry, there should be an abrupt change in manufacturing activity at the border. In contrast, if the policies make no difference, there should be no abrupt change at the border.

In his study, Holmes used right-to-work as a proxy for a state having pro-business policies. Utilizing this border comparison approach, he found that “on average, the manufacturing share of total employment in a county increases by about one-third when one crosses the border into the pro-business side.” This result led Holmes to conclude “that state policies do matter.”

Further research confirms state policies matter

Though not the first to compare counties on either side of a state border in order to distinguish the impact of state policies from state characteristics that have nothing to do with policy, Holmes is largely credited with popularizing the approach. Over the past 20 years, research comparing border counties tends to confirm that state policies do indeed matter.

For instance, economists Randall Holcombe and Donald Lacombe compared border counties to study the impact of changes in state income tax rates. Over a 30-year period from 1960 to 1990, they found “states that raised their income tax rates more than their neighbors had slower income growth and, on average, a 3.4 percent reduction in per capita income.”

Another study of border counties by Holcombe and Lacombe shows “an increase in the generosity of Aid to Families with Dependent Children increased the incidence of female-headed households and reduced female laborforce participation in 1990.”

Recent work by Jeffrey Thompson and Shawn Rohlin for the Federal Reserve Board studies the impact of sales taxes at state borders. They find “sales tax changes have a detrimental effect on employment, payroll, and hiring in border areas, but that these effects are only present in counties with substantial levels of cross-border commuting.” In counties with a high level of cross-border commuting, employment declines by .34 percentage points following a one point increase in the sales tax rate.

Maybe the most dramatic results from this new series of research on border counties comes from economists Marcus Hagedorn, Iourii Manovskii, and Kurt Mitman. After Congress terminated the extension of unemployment benefits in December 2013, many states continued to provide benefits for varying amounts of time, which allowed these researchers to compare border counties based on whether the state continued these benefits. They found “changes in unemployment benefits have a large and statistically significant effect on employment.” Remarkably, they estimated a “cut in the benefit duration accounted for about 50 to 80 percent of the aggregate employment growth in 2014,” which amounts to 2.1 million people gaining employment due to the benefit cut.

A look back at research on Minnesota’s borders

As already noted, Holmes was not the first to use a cross-border comparison. In fact, the Federal Reserve Bank of Minneapolis used a similar methodology back in 1979 to assess Minnesota’s business climate.

In the early 1970s, Minnesota raised taxes while surrounding states lowered them. Minnesota also spent more, per dollar of personal income, than surrounding states. As the study explained, that level of spending might be okay if businesses received a good value for the services: “What matters is whether or not [businesses] want the services they are getting at the price they are paying.”

The study went on to investigate whether “the amount of taxes in Minnesota has seriously increased costs for businesses without providing desired public services.” If yes, they expected to find that businesses expanded more in the 1960s when taxes were lower and that, in the 1970s, more mobile businesses located in neighboring states where taxes were lower. And that is exactly what they found.

The study focused on employment as the best indicator to assess business behavior at the time. “When employment grows in an area or an industry,” as the study explains, “it indicates that existing firms have done well enough to add to their work forces, that new firms have been attracted to the market, or both.”

What did they find? After Minnesota taxes “increased so sharply” between 1969 and 1976, employment grew by 25 percent along the state’s western border, but on just the other side of the border, employment grew by 40 percent. Similarly, employment grew by 15 percent in Minnesota on the southern border but by a substantially larger 36 percent on the Iowa side. To the east, Minnesota growth was similar to Wisconsin, which was attributed to stronger growth on the outskirts of the Twin Cities. Altogether, employment grew by 26 percent inside Minnesota’s border and by 36 percent outside.

Looking specifically at manufacturing jobs, the difference was even more pronounced. Along the entire border, manufacturing jobs declined by three percent on the Minnesota side while growing by 32 percent outside Minnesota.

That was then. What about now?

Minnesota employment not keeping pace with Dakotas and Iowa

All of this prior research demonstrates the value in comparing the economic performance across counties along Minnesota’s borders. With similar labor pools, transportation networks, climates, and natural resources, any economic differences are more likely attributable to state policies.

Like the old Federal Reserve Bank of Minneapolis study, the figures represented here compare employment growth on either side of Minnesota’s border to gauge the behavior of businesses. The figures are based on data from the Bureau of Labor Statistics Quarterly Census of Employment and Wages. These data report the number of wage and salary workers, which excludes self-employed workers, including farmers. The time frame was simply chosen because that is the data easily accessible through their website. Following Thomas Holmes’ methodology, a border county is defined as any county within 25 miles of the border.

Unfortunately, a look at employment growth between 2001 and 2015 reveals a similar growth pattern to the one the federal reserve reported back in the 1970s. During that time employment grew by eight percent on the Minnesota side of the border but by a more robust 18 percent in the Dakotas, Iowa, and Wisconsin. As shown in Figure 1, the Minnesota side of the border supported over 30,000 more jobs in 2001, but by 2011 Minnesota lost this lead and has been losing ground ever since. As of 2015, there are 13,460 more jobs on the other side of Minnesota’s border.

Figures 2 and 3 take a more detailed look at employment growth in the service-producing and goods-producing industries. Minnesota falls short across all them.

The growth advantage on the other side of Minnesota’s border is almost entirely due to strong growth in North and South Dakota. Figure 4 shows 28 percent growth in North Dakota compared to four percent growth in Minnesota and 25 percent growth in South Dakota compared to zero growth in Minnesota.

As employment in North Dakota has grown from the oil boom, there appears to be little spillover across to the Minnesota side of the border. The only apparent obstacle between North Dakota and Minnesota is state policy.

In making comparisons to South Dakota, counties on the Minnesota side should not be expected to grow nearly as fast because there is just no comparable population center like Sioux Falls. But zero growth on the Minnesota side reflects a poor business climate and clearly shows Minnesota is not competing well against the South Dakota side.

The counties along the Minnesota and Iowa border might be the most comparable. Most of the counties have lower population densities and none of them host a major city. By this comparison, Minnesota still loses out. In fact, employment growth in Minnesota counties along the Iowa border declined by three percent between 2001 and 2015, compared to a one percent increase across the border. Iowa’s superior performance is remarkable when considering there is one major difference between the transportation network on either side of the border. Most of the Minnesota side has an Interstate cutting through it, which should attract more economic activity. Nonetheless, employment is declining in this area of Minnesota.

Most of the employment on the Minnesota side of the border exists next door to Wisconsin, and, here, Minnesota employment is growing and growing a bit faster than Wisconsin. This is good, but this is largely a reflection of the border’s proximity to the Twin Cities, which undermines the usefulness of the comparison.

Moreover, the Minnesota and Wisconsin comparison is less informative in terms of comparing state policies because both Minnesota and Wisconsin share a more similar policy mix. Both are relatively high tax states. Based on the Tax Foundation’s measure of state and local tax burden for fiscal year 2012, Wisconsin’s tax burden ranked fourth highest in the country while Minnesota’s ranked eighth. Also, through much of that time, both Minnesota and Wisconsin were non-right-to-work states.

Overall, this review of employment growth along Minnesota’s border shows stronger growth on the side of the border with lower taxes and less burdensome regulations. The tax burdens in Iowa, North Dakota, and South Dakota rank 31st, 33rd, and 49th respectively.

The conclusion here is identical to that made by the Federal Reserve Bank of Minneapolis in 1979:

Businesses are facing a cloudy future in Minnesota: uncertain and at least a little unattractive in comparison to neighboring states. Policymakers thus need to be concerned about the competition of these states if they want businesses to stay in Minnesota and contribute to its continuing prosperity.

This apparently means lowering tax rates.

Peter Nelson is vice president and senior policy fellow at Center of the American Experiment, where he has worked since 1997. his policy work focuses on health care, energy, and state tax and budget issues. He is a graduate of Wheaton College and the University of Minnesota Law School.