Shocking Revisions to Pittsburgh Employment Numbers

Shocking Revisions to Pittsburgh Employment Numbers

Jobs numbers released on March 18th that incorporated a data re-benchmarking by the Department of Labor and Industry and the Bureau of Labor Statistics paint a dramatically different picture of the Pittsburgh region’s employment situation than earlier data had portrayed. Establishment payroll employment was revised downward for late 2012 and all of 2013 with the downward revisions getting larger as the year progressed. For December 2013, the estimate of non-farm employment has been reduced by 20,700 jobs compared to the number originally reported just a month ago. The downward revisions wiped out all the gains that had been previously reported for 2013.

 

Bear in mind that through most of 2013, the employment reports for the Pittsburgh Metro area were showing the region as the statewide leader in job growth with widespread gains across several industry groupings.  With the new data, the Pittsburgh area is just an also ran with job growth below the state’s very anemic pace. Indeed, the revised December 2013 jobs count is 3,100 under the re-benchmarked figure posted for December 2012.  What’s more, all the strong 15,000 to 20,000 year-over-year increases registered in the summer and early fall of 2013 have completely disappeared in the revised data.

 

Months Measured in 2013

Total Non-Farm Jobs (Original Data)

Total Non-Farm Jobs

(Re-Benchmarked Data)

Difference

October

1,188,800

1,169,600

-19,200

November

1,190,800

1,171,500

-19,300

December

1,183,400

1,162,700

-20,700

Private sector employment (non-farm less government) took a major hit as well with the December 2013 number coming in 21,600 jobs below the figure reported a month ago.  Again the pattern was similar to the non-farm data with downward revisions moving higher throughout the year eliminating what was originally thought to be very good growth.

 

Revisions by industry groups were uneven with some taking big hits, some small hits, and a few showing upward changes. The goods producing sector was down, but only marginally, by a total of 2,400 jobs. Construction was taken down by 1,300, Manufacturing a statistically insignificant 200, and Mining by 900 jobs. The decrease in Mining, while small in absolute terms, does represent a nearly ten percent reduction in that industry sector.

 

Obviously, the big losses had to come from the service producing sectors.  Trade, Transportation and Utilities jobs numbers underwent downward revisions throughout the year that basically eliminated the initially reported employment gains. By December 2013, the lowered estimates had removed 4,000 employees from payrolls in the sector. Wholesale trade accounted for almost 2,500 of that reduction.

 

Finance jobs were lowered throughout the year but took their biggest hit in the summer months when original reports showed jumps of around 4,000 jobs in June, July and August. All those gains are now gone according to the newly revised figures.

 

Meanwhile, downward revisions in the estimates of employment in the Professional and Business services sector were relatively modest for most of 2013 but the reduction of 4,000 jobs in the December count essentially wiped out the originally reported gains for the twelve month period ending in December 2013.   Within this sector, the Professional and Technical grouping was relatively hard hit as half of the growth reported originally for 2013 was taken away in the revised estimates.  Administrative and Waste Services employment also experienced a pullback in its jobs count with the December 2013 number lowered by 3,400, leaving the employment total below the December 2012 estimate. However, the Management of Companies and Enterprises portion of this sector did have its job totals revised upward with each month now showing about 1,000 more jobs than initially reported.

 

In something of a surprise, the perennially strong Educational and Health sector suffered the largest setback with the amount of the reductions increasing throughout 2013 to over 9,000 for the last four months. The revisions took away all of the gains that had been originally reported during the year. Indeed, for December 2013, employment in the sector was actually down slightly from the figure posted in December 2012.

 

On the upside, sectors including Information, Leisure and Hospitality, and Government had their jobs totals revised upward. The biggest winner was Leisure and Hospitality with a pickup of 2,600 jobs in the newly re-benchmarked December 2013 number.  The other sectors’ increases were quite small. Obviously, the gains from upward revisions were not nearly enough to offset the large negative revisions in the rest of the industry groups.

 

Without question, if the re-benchmarked numbers are correct, and they usually stand up over time, they are giving a completely different and far less optimistic portrayal of the Pittsburgh region’s employment situation from the one depicted in 2013’s originally reported jobs numbers.  It should also be noted here that the two other Pennsylvania metro areas (Philadelphia and Harrisburg-Carlisle) for which re-benchmarked data are available showed modest upticks in employment over the earlier reports.

 

Moreover, the Household Survey of employment and unemployment corroborates the weakness in jobs growth in the region. The report for January 2014 shows the number of people reporting themselves as working fell slightly from a year earlier and on a seasonally adjusted basis was also down by over 2,000 from December 2013. Meanwhile, the number of people in the labor force dropped sharply, falling a seasonally adjusted 22,400 over the last twelve months. Thus, with the unemployment count falling by a similar number, the unemployment rate tumbled from 7.6 percent to 6.0 percent—a quite remarkable decline in the unemployment rate but for the wrong reason.

 

Throughout 2013 the difference in signals about employment gains between the Household survey and the Establishment survey were suggesting that one or the other or perhaps both sets of employment estimates might have to be revised. The unusually large gains in the Professional and Technical category were also problematic, as we pointed out, because so much of the growth was concentrated in sub-sectors too small to be reported separately and thus there was no way of knowing or verifying what was propelling the Professional and Technical sector growth.

 

What does all this mean? It means our understanding of the regional economy and forecasts of future growth will have to be re-examined.  The spate of stories about how well the Pittsburgh region has fared in terms of employment compared to other regions will certainly have to be looked at again. Perhaps other metro areas in other states will find their jobs numbers were not as good as first thought—or maybe some, like Philadelphia, will see better jobs numbers.

 

To conclude, it is important to note that Pittsburgh’s recovery from the recession through 2012 was very strong as private sector employment climbed to an all-time high following a dramatic surge in job formation beginning in mid-2010.  The re-benchmarking did not push the 2013 employment total below the 2012 total but did effectively eliminate the gains we thought had occurred in 2013.