Tuesday, January 30, 2007

Marginal or inframarginal? 

That's at the basis of the question of whether JOBZ tax breaks create jobs.
The JOBZ program, launched in 2004, has given qualifying firms exemptions from sales, income and property taxes for up to 12 years. Of the nearly 300 firms getting the tax breaks, 13 moved to Minnesota from other places, according to the state. The bulk of the firms in the program expanded existing operations in Minnesota or relocated from a taxed zone to a tax-free one.

An analyst from the Department of Revenue said this month that the exemptions could cost state and local governments more than $100 million through 2015.

The state says businesses that entered the program since 2004 have committed to creating 4,236 jobs and retaining 9,131 in outstate Minnesota.

But some who have scrutinized the program wonder how many jobs have been created and how much credit the program deserves.

"There is really no way at this point to figure out if these jobs would have taken place in the absence of the program," said Laura Kalambokidis, a University of Minnesota economist who has been studying JOBZ.
Some of the debate over JOBZ has been over the wages paid by these firms, but that's not the real issue. If you agree to take a job at the wage offered, you've revealed that the money earned on that job is worth more to you than what you could earn elsewhere, and worth more than if you were at leisure. The wage requirements in JOBZ are an attempt to make the jobs created worth the amount of tax breaks paid. (For example, there's got to be a very big wage differential over existing jobs for the $336,000 per job subsidy John Palmer reports on an electric plant in Quebec.)

What's more important is whether the tax breaks actually increase jobs. One local economic development official refers to JOBZ in this story as "the extra dollop or two of whipped cream on top of the pudding." The question is whether the extra dollop of whipped cream induces more pudding to be consumed -- is the JOBZ money changing the marginal cost of hiring another worker? If it is, then you have the question of elasticity, or "how much does a $1 property tax break lead to hiring of an additional worker?" (Attention students: Seminar topic.) John's comment on the long-run Phillips curve, as he acknowledges, doesn't apply if you've changed the marginal revenue produced by a unit of labor. But if JOBZ is just a transfer to the capital owner without any change in labor's marginal value, then there would be theoretically no effect on the amount of labor hired. Moreover, if there is an effect that causes JOBZ-favored firms to increase wages then all wages go up, even at the firms not gifted by JOBZ, as good workers are bid away to the place where their marginal revenue product is higher.

Worth a read: Prof. Kalambokidis' 2003 summary of the effects of enterprise zones. She's a skeptic; I don't agree with every point she makes, but on balance I share her skepticism. See also this MN House research report from 2005.