(Note: The following is testimony delivered by Dr. Gary L. Wolfram, a Hillsdale College professor of political economy, before the Michigan House Taxation Committee on Jan. 31, 2006)
Mr. Chairman, members of the Committee, I have been asked to give my view of the state of Michigan’s economy. In short, Michigan's economy is doing poorly and this is not due to a slowdown in the national economy, but rather shortcomings that are specific to Michigan. In the absence of reductions in the cost of producing goods and services in Michigan, the economic decline will continue.
In discussing the state of Michigan’s economy, one must inevitably resort to numbers. But as Will Rogers once said: "There are lies, damned lies, and statistics." For example, in her State of the State address, Gov. Granholm asserted that 99,000 more people are working in Michigan than when she took office. This was countered by the Michigan Republican Party that Michigan had been losing jobs during the governor’s tenure. When trying to wend one's way through political arguments, it is useful to have common sense, thoughtful observation, and look for a pattern of facts. That is what I will attempt to do here.
I will look at Michigan's employment numbers, unemployment numbers, personal income, real gross state product, housing prices and migration data. If we can find a consistent pattern in the data, then we have a pretty good picture of what the economy looks like and what it is trending towards. So let us take a look at employment numbers first.
The governor's assertion that there are 99,000 more people at work today than when she took office is based upon what the Bureau of Labor Statistics calls the Household Survey of employment. This survey includes agricultural workers, self-employed, unpaid family members and private household workers. It is based on a smallish sample of between 1,500 and 2,000 households in Michigan. The numbers generated by this survey can be found at the BLS Web site and are in both seasonally adjusted data and non-seasonally adjusted data. If one uses the seasonally adjusted data, in January of 2003 there were 4.696 million employed in Michigan, and the latest figure, for December 2005, is 4.795, or 99,000 more employed.
If one uses the establishment survey, which counts non-farm payroll jobs, and uses the seasonally adjusted numbers, 4.445 million jobs in January of 2003, and 4.362 million jobs in December of 2005, one finds a loss of 83,000 jobs. The establishment survey is less comprehensive in what it counts as jobs, but it is from a much larger sample of about 15,000 establishments in Michigan.
If one takes the sum of annual job losses, which is an average rather than a monthly snapshot, one gets a sum of job losses in 2003, 2004, and 2005 of more than 160,000 jobs. What is the right answer?
The establishment survey clearly will have a lower number of jobs because it doesn't include the self-employed or farm employment. Since one is looking at trends rather than levels, in order for the household survey to be correct, there must have been a substantial increase in the number of self-employed and agricultural workers in the past three years for the household survey to give a better picture of what has happened to employment. Common sense will cast some doubt upon this. In addition, the employment sample is from a survey that is nearly 10 times larger and is certainly more accurate in what it measures. Thus, my opinion is that there has been a net job loss, at least in the tens of thousands over the last three years.
If one simply looks at the employment numbers one must choose one methodology over another and reasonable people may agree or disagree on this. But when one looks at a number of different measures of economic activity, a clear picture emerges of a faltering Michigan economy in the midst of economic prosperity in the rest of the country. For example, Michigan is the only state to have job losses on a year over year basis for each month of 2005. This is not consistent with an expanding Michigan labor market.
If we compare Michigan’s employment picture with its surrounding states, a pattern emerges as well. I ran a simple trend regression for total non-farm employment, total goods producing employment, total private employment and total manufacturing employment for the five states in the area: Michigan, Indiana, Illinois, Ohio and Wisconsin. The regressions were monthly employment for the various categories for the months beginning January 2003 and ending December 2005. A negative coefficient on the variable x, which is simply a time trend, indicates a consistent decline in employment.
In both total private employment and total non-farm employment, Michigan was the only state with a negative coefficient. Every other state has a positive trend line. In both total goods producing and manufacturing employment, we find Indiana and Wisconsin had a positive trend, and Michigan had the largest negative trend of the other three states.
Next, one should look at the unemployment figures, which are based upon the household survey. Michigan's unemployment rate has consistently been among the highest in the country, well above the national rate. The state's unemployment rate hit an historic low of 3.1 percent in March of 2000. The rate climbed during the recession of 2001 and 2002, moving above the national average in early 2001. But while the rest of the nation has been in recovery, Michigan's unemployment figures show that it has not. If one looks at monthly unemployment rates for 2005, the national unemployment rate has ranged from 5.4 percent to 4.9 percent, with the December rate being 4.9 percent. Michigan's rate, on the other hand, has ranged from 7.5 percent to 6.1 percent, with the current rate at 6.7 percent.
One can also look at Indiana and Minnesota for comparison. Indiana is a Midwestern state with a higher percentage of its jobs in manufacturing than Michigan. Its unemployment rate in 2005 has ranged from 5.7 percent to 4.8 percent, with a current rate of 5.5 percent. Minnesota, another Great Lakes state, has had an unemployment rate from 4.4 percent to 3.6 percent with a current rate of 4.1 percent. Clearly, Michigan stands out in its high rate of unemployment.
If we look next at state personal income, the pattern continues to emerge. Looking at percentage change in personal income from quarter to quarter for the last five quarters for which data is available, that is 2004:III through 2005:III, the change in Michigan was .7, 1.5, .1, 1.4, and .6. The national rates were 1.2, 3.0, .6, 1.1, and .7. That is, growth in personal income from one quarter to the next was lower in Michigan than the nation in four of the five quarters. Michigan ranked 47th in growth in personal income in the latest quarter, 2005:III. Per capita personal income numbers are significantly delayed, but the annual numbers are available for 2003 and 2004. In 2003, Michigan's per capita personal income was $102 above the national average, and in 2004 it was almost $1000 below the national average.
Real Gross State Product is a measure of the amount of production in the state in a given year. The percentage change in Real Gross State Product is available from the Bureau of Economic Analysis for the 2003-2004. Michigan ranked 50th among the states in percentage change in the amount of economic production, at 1.2 percent. This growth in output was barely a quarter of the growth in the nation, which was 4.3 percent. It was less than one third of Indiana’s 3.8 percent growth, and Minnesota’s 3.9 percent. The highest growth in state product was Nevada’s 9.3 percent.
Continuing the pattern, we can look at out-migration. State's that have an expanding job market do not usually have a net out-migration of its citizens. United Van Lines, the nation’s largest household goods mover, does an annual survey of migration from state to state. The latest survey was released this month, and Michigan was classified as a high-outbound state. For Michigan, 63.9 percent of its shipments in calendar year 2005 were outbound. This was second in the country to North Dakota's 67.8 percent outbound shipments. The lowest outbound rate was Oregon at 36.4 percent.
The Census Bureau numbers are consistent with the United Van Lines survey. Between July 2004 and July 2005, nearly 48,000 more people moved out of Michigan than moved in. During the same period the prior year, 37,000 more people left Michigan than migrated to the state.
The state of the economy is often reflected in housing prices. Areas that have an improving economy are likely to have an increase in demand that will bid up prices. If we look at the change in housing prices from the third quarter of 2004 to the third quarter of 2005, Michigan ranked 51st in the country, behind every state and the District of Columbia. Its 4.01 percent increase was about a third of the increase in prices in Montana, and right at one-third of the national average. At the recent International Home Builders Show in Orlando, three of the economists addressing the convention singled out Michigan as a state that will continue to struggle.
Things are not getting better in the state's housing market either. Median sales prices in southeast Michigan fell 4.3 percent in December. Total home sales fell by 11 percent while the number of homes listed for sale grew by 11 percent. The homes that sold in the region in December had spent an average of five weeks on the market longer than the prior year.
In summary, while one might find a statistic or two here and there that could be used to attempt to describe Michigan’s economy as either fair, or doing poorly because of the national economy, the preponderance of the evidence is to the contrary. Michigan’s economy is fairing poorly and is moving in the wrong direction, while the national economy has been and remains strong. The final quarter of 2005 was the 17th consecutive quarter of growth in inflation adjusted GDP at the national level. Michigan needs to look inwardly, discover why it is more expensive to produce here than elsewhere, and act to rectify the problem.
Gary L. Wolfram is an adjunct scholar at the Mackinac Center for Public Policy, a research and educational institute headquartered in Midland, Mich. Permission to reprint in whole or in part is hereby granted, provided that the author and the Center are properly cited.