It was in early January when the headlines and newscasts echoed a familiar dirge -- that federal labor statistics showed that the Cleveland-Elyria-Mentor area had lost more lost more jobs the previous November than any region in the country, and was the only one out of the 37 biggest metropolitan areas to do so.
Those grim assessments stemmed from the Current Employment Statistics (CES) report, which is issued each month by the federal Bureau of Labor Statistics (BLS).
And they particularly rankle the business development and regional marketing folks at Team NEO, a semi-private extension of Ohio’s job creation and promotion arm that serves the northeast region.
“Greater Cleveland is not losing jobs,” argues Jacob Duritsky, its managing director of research.
Duritsky acknowledges that Team NEO has used the monthly figures in its quarterly economic reviews. But he says after so many dour snapshots, they dug into the BLS’ methodology… and now will no longer use them.
“The current employment statistic surveys about 6 percent of all U.S. establishments," says Duritsky. "So a very, very small percent of businesses in the country are even surveyed. At the national and state level, the data fairly good… but once you get down to the metro level, there are very small samples with very high error rates.”
A leading economist at Cleveland State University is also dumping the monthly CES data. Ned Hill, Dean of CSU’s Levin College of Urban Affairs, says the monthly BLS unemployment stats appeal to those wanting a quick and accessible number to throw into a news story or press release, and not always for the best reasons.
“They aren’t worth commenting on because they’re statistical garbage," says Hill. "I’ve had journalists as well as political copywriters come in, and -- can you imagine -- seeing all these bad numbers, and how it can be used against elected officials? It’s the rush to be immediate, it’s causing us to make mistakes.”
Hill says a more accurate measure comes months later, called the Quarterly Census of Employment and Wages (QCEW).
“This isn’t a sample, this is a census that covers 98 percent of all employment in the United States.”
Both Cleveland State and Team NEO say they’ll work with the QCEW figures, even if they aren’t immediate. A review showed that a reported loss of nearly 7000 jobs between June 2012 and June 2013 actually was later adjusted to a gain of nearly 9000 jobs when the BLS’s quarterly census data was available.
This isn’t news to Joel Elvery. He’s a researcher with the Federal Reserve Bank of Cleveland, and has long championed the quarterly reports in gauging the Greater Cleveland area’s job outlook.
“And so actually the average revision’s larger than the average decline. Which suggest that if the revisions follow the same pattern they’ve been on, then after the revisions, we’ll find that we’ve had a job gain and not a job loss.”
Recently, the Plain Dealer addressed the issue in an editorial, embracing the adjusted findings that painted a rosier jobs outlook for Greater Cleveland.
But the monthly CES releases have defenders, including Cleveland economic researcher George Zeller.
“These figures are extremely reliable," he says. "The fact that they are revised is not unusual, it happens every year. It’s gonna happen again this year. It’s going to be a slight upward revision but not nearly enough of an upward revision to show that we’ve regained the jobs we previously lost.”
Zeller says outside the numbers debate, the main point he sees is that Greater Cleveland is still struggling to recover fully from two major recessions that happened in the previous decade, and its rate of recovery still trails that of the nation. Or to use a sports analogy….
“If the Browns are in last place, and they win a game and they’re still 10 games out, are they now in first place?" asks Zeller. "No, that’s an improvement, but we have to compare ourselves to everybody else. And I think that’s where this leap to judgment, that these numbers are no good, is ill-founded.”
Paul LaPorte of the BLS office in Chicago says the Bureau stands by the methodology of its monthly reports, but since they’re based on a sample, errors are to be expected.
He says a new benchmark employment report for all of 2013 is due to be released March 17th.