By Kurt Metzger
The Detroit Free Press ran a page-one story last Sunday entitled Is Michigan really the 'comeback state' yet?, by John Gallagher and Susan Tompor. The article was designed to look at the upcoming gubernatorial race between Rick Snyder and Mark Schauer, with a view of the economy since 2006 at its core.
What grabbed my attention were the six paragraphs that accompanied the story.
These six graphs were tied to six Economic Indicators that were being used to measure Michigan’s rebound from the recession and, by extension, Governor Snyder’s past four years in office.
The headline above the six charts was the following: “Economic Indicators Say Michigan Has Improved. But We Might Have a Long Way to Go.”
This is a great summary of the data as presented in the graphs. There is a major problem, however, in that the data in three of the charts are INCORRECT or MISLEADING. In each case, the correct data paint a much dimmer picture than what the article presents. Let me present the facts.
The Free Press says that Gross domestic product is “on the upside’ and that it “has surpassed prerecession levels.” There are two major problems with the chart presented. The first is that it indicates the data are in “billions of dollars” when it should say “millions of dollars.” The second is that, while the numbers are correct as shown, they are reported in current dollars and, thus, do not account for inflation adjustment. The chart below adjusts the years 2006-2012 to “chained 2013 dollars.” As one can see very clearly, while we have been on the upswing since 2009, we are still well below pre-recession totals.
Figure 1. Gross Domestic Product for Michigan, 2006 – 2013 (in chained 2013 dollars)
Let’s move on to median household income. This indicator was the one that got me doing this research. I have done a number of articles and presentations on trends in income and poverty (covered next) in Michigan and our local areas. I knew that the trend reported in the article was incorrect.
Median household income was another indicator “on the upside,” accompanied by the statement, “The income figure of $50,015 in Michigan has now surpassed prerecession levels, although not by much.”
The first problem, after checking with the author, is the source used. It was verified that it was from the Census Bureau’s Current Population Survey.
While this is a great source for national level data, its small sample size (just over 60,000 households) makes subnational data subject to significant sampling error. The American Community Survey, with a current sample of three million households, provides much more accurate annual state, metro area and large county and city data. In addition to the survey choice, which produced artificially high values with significant error, the chart that is used does not inflation adjust its numbers.
Figure 2 corrects both of these problems.
What a different picture these numbers portray! Not only have we not reached precession levels, we have continued to drop every year between 2006 and 2011, just showing the first gain (statistically insignificant however) between 2011 and 2012. We are now 13 percent below where we were in 2006!
Figure 2. Median Household Income in Michigan, 2006 – 2012 (in 2012 dollars)
The third and last area of dispute is with the data on poverty. Though not included as an indicator “on the upside” or “on the downside,” the text accompanying the charts states, “Michigan’s poverty rate has returned to about where it was before the recession.” The chart shows a Poverty Rate of 13.7 percent for 2012 , although it labels it 2013 (a year for which data are not yet available). Let’s look at Figure 3 for the real poverty rate values.
It is obvious that poverty in Michigan in 2012 is much higher than the Free Press reports, just experiencing its first decrease (1/10 of a percentage point) in 2012. The chart in the article has poverty decreasing since 2009!
I have written about the increase in Detroit’s poverty rate (highest in the country), in addition to the growing suburbanization of poverty, for years and the Brookings Institution has recently quantified that growth for metropolitan areas across the country. According to the Annie E. Casey Foundation, Michigan’s rank among all states in terms of poverty is tied for 33rd, based on all persons in poverty and tied for 36th when it comes to the percent of children (under 18 years) in poverty.
Figure 3. Poverty Rate in Michigan, 2006 – 2012
I am pleased to say that the charts depicting Michigan’s Population and Professional and Business Service Jobs check out perfectly with the datasets I reviewed. I chose to accept, and not research, Office Vacancy Rates. All reports regarding vacancy rates in the suburbs and city have been positive of late. In addition, we are seeing increased convention traffic and increasing hotel occupancy rates.
These are definitely positive indicators for Metro Detroit. I cannot speak to other parts of the state, however.
My career has always been about “better (accurate) data make for better decisions.” I encourage the use of data as a true measurement of outcomes – be they politically tied or programmatically tied. We must continue to advocate for open data in government and for the continuous measurement of the agencies and programs we fund.
Update: The Free Press addressed the issue Sunday. Click here to read the paper's explanation.
Kurt Metzger has spent more than 35 years promoting data accessibility in southeast Michigan through his work with the Census Bureau, Wayne State University, the United Way for Southeastern Michigan, and as founding director of Data Driven Detroit. He is currently mayor of Pleasant Ridge.