How Mitch Daniels’s anti-Keynesian policies saved Indiana

Mitch Daniels’s Indiana is an unusual outpost of fiscal discipline in the Midwest. Compared to its neighbors, Indiana stands out with its AAA rating and post-recession budget surplus. And now it looks like the governor whom the Hudson Institute’s Herb London once described as “viscerally parsimonious” has shown that fiscal sobriety is good not just for the taxpayer, but for the job-seeker.

In an original new study comparing the experiences of Indiana and Michigan during the recession, Dr. Mike Hicks and Kevin Kuhlman of Ball State University write:

Indiana and Michigan are remarkably homogeneous states, with similar populations, demographics, and industrial structure. They are adjacent and enjoy very similar patterns of historical development. The experiences of both Indiana and Michigan’s economies over the past 30 years have been similar. However, during this recession, the experience of the two states has diverged remarkably. This divergence is a subtle story of economic and fiscal conditions in both states.

The authors used a model to predict how Michigan would fare in the recession. The model perfectly predicted the 14.9 percent unemployment that Michigan experienced. The model also predicted Indiana would top 14 percent unemployment, but in actuality the Hoosier state peaked at 10.8 percent. Between 2007 and 2009, weekly wages in Indiana rose by $3.51 but dropped by $22.93 in Michigan.

Hicks and Kuhlman found that:

•             Changes in manufacturing didn’t explain the difference between the states. Indiana is actually more manufacturing intensive than Michigan.

•             Indiana reduced taxes during the recession while Michigan passed two tax increases.

•             Michigan got more stimulus money than Indiana, yet performed worse. Even though unemployment was 40 percent higher in Michigan than in Indiana, each new stimulus job cost $70,000 more in Michigan than the Hoosier State.

•             Indiana managed to keep its bond debt low. It has about one-third the amount of bond debt per capita as Michigan, an important signal that taxes won’t need to rise.

Daniels inherited a bankrupt state in 2005 and immediately launched a host of initiatives to bring Indiana’s fiscal house in order. The anti-Keynesian diet that followed—cost-cutting accompanied by low taxes and low levels of borrowing—resulted in the right kind of economic stimulus. A good lesson for other states.

Streeter is a Distinguished Fellow for Economic and Fiscal Policy at the Sagamore Institute and a Nonresident Transatlantic Fellow at the German Marshall Fund.

6 thoughts on “How Mitch Daniels’s anti-Keynesian policies saved Indiana

  1. Indiana’s weekly avg wages rose by a WHOLE WHOPPING $3.51 in 2 years. WOW!!!! OH, WOW!!!
    What is Indiana’s weekly avg wages? Low, low, low. Be so proud of Indiana’s wages. We are getting closer to undeveloped countries’s wages. YEA!!!!
    Get real.
    Bankrupt? Really? Possible, in debt. But bankrupt. Really?

  2. Ryan Streeter is making a lot of assumptions about why Michigan and Indiana differed in their unemployment rates during the Great Recession, which are not supported by the report he references. The Ball State University report doesn’t make any claims with respect to Mitch Daniel’s leadership, John Maynard Keynes, or anything else like that. All the report says is the econometric model developed by Hicks and Kuhlman to forecast unemployment rates in Michigan and Indiana did relatively well since 1978, but, beginning in late summer of 2007, when the Great Recession was just starting, the model’s ability to forecast unemployment rates in Indiana weakened considerably, and that the authors believe four factors potentially play a role in explaining why.

    These four factors are: First, the structure of the automobile industry. Second, fiscal and debt issues (the report mentions the less favorable tax climate in Michigan and concerns about future tax increases, but interestingly, says nothing about cuts in state and local government). Third, the relative effect of foreclosures. Fourth, the efficiency of job creation from stimulus spending as reported to the Office of Management and Budget (OMB) (Michigan received $3.9 billion more in federal stimulus money than Indiana, but $70,000 more had to be spent for each job the stimulus money created in Michigan).

    Models like this one assume the relationship between different variables remain the same over time and in instances where those relationships change, the model deviates from reality. I think that’s what happened here. If you look at the chart on the bottom of page 2, which compares forecasted unemployment to actual unemployment, the model appears to work relatively well… except during every recession we’ve ever had. Check the recession of the early 1980′s to confirm this. Things change in a recession and apparently the model fails to fully capture what drives unemployment rate changes during such times. Any wonder there’s no consensus among economists on when the economic downturn will truly end?

    Unlike in past recessions, where the model forecast higher unemployment for both Michigan and Indiana than actually occurred, that didn’t happen this time. Instead, the Michigan forecast showed slightly lower unemployment than actually happened and the Indiana forecast showed signficiantly lower unemployment than actually occurred. In other words, something changed during the Great Recession that likely requires the model itself to be revised. It’s a lot more complicated than sim[ly blaming Keynes or lionizing Daniels, as the op-ed writer would like us to do.

    So what changed for Indiana to have less unemployment than the model forecast? Was it federal stimulus spending (factor four)? Assuming OMB’s numbers of 15,116 jobs created in Michigan and 10,974 jobs created in Indiana are accurate, this only accounts for a differential between the two states of 4,142 jobs (score one for Michigan). Was it the abnormally large amount of pressure on state and local governments to trim spending (factor two)? According to the U.S. Census Bureau, 10,887 government workers lost their jobs in Michigan between 2007 and 2009 while Indiana actually added 10,502 government jobs (score one for Indiana). In other words, federal stimulus spending was more than offset by austerity measures in Michigan, that, on net, left Indiana ahead of Michigan by 17,247 jobs. In states the size of Michigan and Illinois, though, 17,247 jobs fails to move the unemployment rate by even 0.1%. Indiana did go on to reverse course and cut government jobs in 2010, but that can’t explain a deviation starting back in 2007.

    What about the tax changes in Michigan and Indiana (factor two)? The Tax Foundation rankings the study uses to compare the favorability of the corporate tax climate of the two states between 2007 and 2010 shows Michigan improving from 50th to 48th. Indiana gets worse and then gets better, going from 22nd to 23rd and then to 21st. Likewise, the Tax Foundation’s overall tax climate index shows Michigan improving from 23rd to 17th then falling to 21st and finally improving to 17th. Indiana fell from 12th to 13th to 14th and finally improved to 12th. If the argument is Indiana’s business climate is better than Michigan’s, it is. And Mitch Daniels was probably a better governor than Jennifer Granholm. But it’s a big stretch to argue the tax climate improved so much — in real or relative terms — to explain any significant deviation in unemployment rates between the two states when the Tax Foundation identies little improvement. Neither state saw a big influx of new businesses during the recession because few businesses were starting up or expanding anywhere. (Score little for anyone. Times are tough for the Midwest.)

    I’m inclined to believe a combination of the last two factors best account for the difference — the structure of the auto industry (factor one) and the relative effect of foreclosures (factor three). Michigan has been in a recession for nearly a decade, largely driven by a wholescale restructuring of the auto industry, which culminated in the near bankruptcy, bankruptcy, or bailout of Ford, GM, and Chrysler. The Ball State University report largely glosses over all this. And the op-ed fails to mention the auto industry entirely. While Hicks and Kuhlman’s model incorporates 3-month lagged auto sales to help forecast unemployment when auto sales do decline, this in no way can account for such a precipitous drop in automotive sales volumes as was experienced during this time period. Sales dropped from 17 million units in 2006 to 10.6 million units in 2009, which was unprecedented. Nor can the model possibly account for the revolutionary change this brought to the industry — plant closings rather than simply reduced overtime, reduced hours, and layoffs, which were the historical norm when times were tough.

    This time there were eight plant closings in Michigan to rationalize capacity. Only one occurred in Indiana. There were layoffs at twenty-seven plants in Michigan. Only five plants had layoffs in Indiana. And this only includes the Big Three. It ignores all the job losses at tier one, tier two, and tier three auto suppliers who were forced to close their doors when the auto plants they served closed. 221,000 automotive jobs were lost in Michigan between 2000 and December of 2007 with another 70,000 jobs lost during the recession. The job losses in Indiana were incredibly light compared to Michigan’s because such a large portion of the reduction in plant capacity was concentrated in Michigan, not Indiana. This is something the model could not possibly have forecast. And the job loss disparities are made even greater because they don’t include all the anciliary, non-automotive jobs that also disappeared when the plants closed, whether that be the waitress at the corner diner that provider lunch or the lawn service crew that kept the grass trimmed.

    Historically, the Hicks and Kuhlman model overprojected job losses during recessions. This time Michigan was near the forecast because it’s job losses were much worse than normal. And as entire communities were devasted by factory closings, many people couldn’t afford their mortgages, which is why Michigan, which has 50% more residents than Indiana had 2.5 times the number of foreclosures. And the problem is further compounded by the 6-month redemption period on residential foreclosures in Michigan, which means banks can’t even try to sell foreclosed homes for six months after the foreclosure is complete. All this does is further depreciate home values and lead to a deterioration in the home stock. There are plenty of reasons Indiana has been more fortunate that Michigan during the Great Recession and they have little if anything to do with dead economists.

  3. let’s see if this “distinguished” fellow has the stones to address real criticism — doubtful, since life is so much easier when you can boil down huge amounts of economic data into a “my team is better than your team” talking point.

  4. Hilarious…Indiana is still the poorest state in the Midwest. Has the highest poverty rate in the Midwest. Michigan cut its budget from 2000 to 2010 by far more than Indiana did. And it cut its taxes over that period by far more than Indiana.

    To look at 2007-2009 and call that an economic analysis is ridiculous.

    Here’s the real question: What state has the lowest unemployment rate and the highest per capita income in Midwest? Minnesota, the state with the highest taxes — and highest percentage of college graduates living in it. What’s Gov. Daniels done to attract more college graduates? Nothing.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>