Thursday, July 17, 2008

Making new numbers 

MPR is running a series of stories on the "Minnesota Slowdown". More on the stories in a moment, but first I want to call attention to their new "Job Pain Index." They define it as "a gauge of how hard it is to find a job."
A lower index number means it's more difficult to find a job because job seekers are facing less demand, or more competition, or both.
OK, that sounds very cool to me, as someone who has created some indexes himself for the St. Cloud area. So I'm eager to dig into this index and see what it is and how it works. And they give me details.
The Job Pain Index incorporates measures of supply and demand for workers, and total employment, the point at which supply and demand reach equilibrium.
I don't understand the last item. We don't really know what equilibrium employment is; what we observe in the labor market is where the amount of labor sold equals the amount of labor bought, which is true tautologically. We have some idea of full employment unemployment rates (I know, it sounds oxymoronic, but all labor markets will contain some workers who are literally between jobs and others for whom the skill-position mismatch is so severe that they cannot find meaningful work even at full employment. c.f. Greg Ip.) I would think conceptually that an indication of job pain would be measured by things which increase the supply of labor -- meaning workers face more competition from other workers -- or a decrease in the demand for labor -- fewer buyers.

The index contains four items:
The first item is certainly a good measure, but it is an indicator of increase supply of sellers of labor in the market, not a measure of demand. We use initial claims in the St. Cloud unemployment office as the one of our measures for the St. Cloud Index of Leading Economic Indicators, and it is used nationally both for the U.S. LEI index and as an indicator for the stock market. (Today's report.)

Help wanted is also a good measure that we use for St. Cloud (our measure comes from linage at the St. Cloud Times rather than an on-line series.) But this is a demand measure that rises when demand is rising, in which case job pain is falling. Perhaps they invert or flip the sign on this number; if not, it points in the opposite direction of job pain.

But an online series is rather skewed in its representation of jobs. Dave Senf at DEED studied online job postings. He finds, for example, that more than 35% of jobs advertised online are professional positions, but the share of professional jobs in the state is closer to 12%. Another 29% of ads online are in management, which is far less than 10% of the jobs out there. We have intentionally kept the newspaper data in St. Cloud because the shares of those jobs in St. Cloud are even less than for the Twin Cities. I think the measure is rather skewed.

I have fewer issues with the number of workers unemployed here, but we do know that workers may be encouraged to join the labor force and leave family-building and other non-labor activities when wages rise. That is not necessarily a measure of pain. But the logic of the index seems to equate pain with an increasing labor surplus. If that's the theory, the component is at least reasonable.

So two of the four items in the index are really suspect, one is good and the other OK if you buy the premise of the index. There already exists a series for each of the fifty states, created by the Philadelphia Federal Reserve, which includes some of the indicators they are using, except rather than some value-laden term like "job pain" they call the index a neutral "coincident indicator" series. The data and methodology are documented there far better than the MPR series, and if one is looking for a piece of data to use I would recommend the Philly series.

About the rest of the Minnesota Slowdown project: I like the idea of the story, but note that it did not include any "economic lookouts" from Central Minnesota. This happens to be the one area that has faster employment and population growth longer-term than the Twin Cities. But the whole project seems to hinge on a story that makes the housing industry's misfortunes cause the state economy to sink before the national economic slowdown/recession. It's the premise the state's economists have used for a year now; weird that with so much pain the government keeps getting more revenue than they expect, isn't it?

Labels: ,