Data revisions shed new light on job growth in region's metros, say Cleveland Fed researchers

Some previously reported job losses were actually job gains, and vice versa

On March 21, the Bureau of Labor Statistics (BLS) released the benchmarked metro-level Current Employment Statistics from the State and Metro-Area Employment, Hours, and Earnings survey (CES/SAE). According to Federal Reserve Bank of Cleveland researchers Joel Elvery and Christopher Vecchio, this revised data shows how unreliable the initial metro-level CES/SAE estimates can be.

For example, according to the initial estimates, Cleveland lost 7,200 jobs from September 2012 to September 2013. But the revised data show that Cleveland actually gained 7,400 jobs over that period. Cincinnati and Columbus, meanwhile, gained substantially more jobs than originally estimated:  17,800 vs. 6,700 and 25,900 vs. 12,900, respectively.

Employment growth was revised downward in other metro areas in the Fourth Federal Reserve District. Pittsburgh, for example, gained 2,400 jobs, not the 20,000 jobs suggested by the initial CES/SAE estimate. (See spreadsheet for all metro areas in the US).

Elvery and Vecchio compared employment growth for a dozen Fourth District metros with the 1.9 job growth percent for U.S. metros as a whole. They found that job growth in Columbus and Lexington exceeded that for metros as a whole, while Cincinnati, Toledo, and Wheeling came in just under 1.9 percent. Jobs grew in Akron, Cleveland, Lima, and Pittsburgh, but less than half as fast as for US metros as a whole. Dayton and Youngstown showed no change in the number of jobs, and Erie lost 0.3 percent of its jobs.

For more details, see below and read Annual Revisions to Metro Jobs Data Shed New Light on Job Growth in Fourth District Metros

Initial estimates of metro-area employment numbers not always accurate, say Cleveland Fed researchers

A more reliable alternative? The Quarterly Census of Employment and Wages (QCEW)from the BLS

Policymakers, businesses, and the public are all interested in knowing how many jobs their region has gained or lost. Unfortunately, the initial estimates of metro-level Current Employment Statistics from the Bureau of Labor Statistics’ State and Metro-Area Employment, Hours, and Earnings survey (CES/SAE) are subject to large revisions. Federal Reserve Bank of Cleveland researchers Joel Elvery and Christopher Vecchio recommend an alternative that, while not as timely, is far more accurate.

Elvery and Vecchio compared initial CES/SAE estimates of job gains and losses to two other data series -- final CES/SAE estimates and the Quarterly Census of Employment and Wages (QCEW) -- for six metro areas and four states in the Fourth Federal Reserve District. Their results show that it can be difficult to accurately gauge the magnitude of changes in employment—or even the direction of changes—from the initial monthly CES/SAE estimates. Looking at monthly data for Cleveland, for example, from January 2006 to March 2012, the researchers say the initial CES/SAE disagreed with the final CES/SAE on the direction of job changes in 21 percent of the months. However, the QCEW disagreed with the final SAE in only 4 percent of the months.

Elvery and Vecchio say the survey design and size for the initial CES/SAE produce too small a sample to generate accurate estimates for metro areas. The QCEW, on the other hand, is based on an actual count of employment, making it more accurate.

Read Which Estimates of Metropolitan-Area Jobs Growth Should We Trust?