Welcome to the jobs numbers game

OK, THIS could make your head hurt, but it's important stuff.

It's about job numbers.

Politicians, including embattled, poll-suffering Gov. Corbett, always promise jobs, tout jobs or point to jobs as evidence of leadership.

So even 18 months before the guv is on the re-election block, those seeking to unseat him are pointing to state job numbers.

This is common politics, and almost always dicey.

You'll hear the following: We're losing jobs; we're doing better than many states; ever-changing jobs data in big states is statistically insignificant to a state's overall economy; monthly jobs numbers are seasonally unadjusted snapshots that don't reflect real progress; and governors can't do all that much - it's Obama's fault.

Maybe grab an Advil or two?

The Pennsylvania number getting attention at the moment is from the W.P. Carey School of Business at Arizona State University. It ranks states by "job growth" each month, using data from the U.S. Bureau of Labor Statistics.

We're ranked 49th, ahead of Wyoming, the "Cowboy State."

Needless to say, few in the guv's camp are yelling yippee-ki-yay!

"It sounds bad," says Corbett press aide Kelli Roberts, "when you're not getting the facts behind it."

When you do, she suggests, the ranking is "meaningless."

I should add that when the same rankings were issued for the same month by the same business school in 2011, the year Corbett took office, we ranked 13th.

No other state in the region - not New Jersey, New York, Delaware, Maryland or Ohio - fell so far in the rankings during the same period.

But Roberts and others argue that the ASU data is not seasonally adjusted and only compares job growth during two months, last March and this March: a snapshot that shows 1,800 fewer jobs out of more than 5.7 million.

So I went directly to the U.S. Labor Department, where Gary Steinberg of the Bureau of Labor Statistics directed me to current jobs data that is seasonally adjusted.

He notes that BLS does not rank states. But the adjusted data is actually worse. It shows that in the March-to-March comparison, Pennsylvania is down 6,000 jobs and is the only state in the region with fewer jobs than last year.

Delaware gained 500; Ohio gained 2,000; Maryland gained 38,000; New Jersey gained 54,000; New York gained 86,000.

Hang on. There's more.

I also went directly to ASU and Lee McPheters, an economics research prof at ASU's W.P. Carey School, who compiles the data in question.

"It is correct that we do a snapshot," he tells me.

But ASU also does a rolling 12-month comparison, arguably a better way to look at the numbers.

Over the past 12 months, no state lost jobs and Pennsylvania added 32,000 jobs, McPheters says.

That sure sounds better. But Pennsylvania still ranks low (45th) with a job-growth rate of 0.6 percent. That's well below the national rate of 1.6 percent and below all our neighbors, including New Jersey and New York, which have higher unemployment rates.

Bottom line is that if you're running against Corbett, you use the one-month-to-one-month comparison. If you're interested in a little more fairness, you use the 12-month comparison. And if you're the incumbent, you use a longer comparison.

"What we look at is, 'Are we adding jobs?' And over the course of the governor's tenure," says Roberts, "we've added roughly 100,000 jobs which puts us around 11th [among states]."

So heads up. There are multiple ways to use data.

And we now could get into how and why job-growth rates and unemployment rates measure different things and are often disconnected. But I have a headache.

 


Blog: ph.ly/BaerGrowls

Columns: ph.ly/JohnBaer