Revisions Boost 2011 Job Gains Estimates
When the Pennsylvania Department of Labor and Industry (L&I) issued the January workforce information release for the Pittsburgh metropolitan statistical area (MSA), the household employment data indicated a slowdown in the local economy compared to the brisk pace of 2011, especially the strong growth posted in the fourth quarter. While this caused a minor stir in the media, of more importance was that January’s data reflected newly re-benchmarked household employment (persons reporting themselves as working) and labor force data for 2011. The updated benchmarks in the household survey data show a more robust labor market in the Pittsburgh MSA in 2011 than was originally reported last year.
To see the impact of the re-benchmarking of the 2011 data, a comparison to the original reported data is required which was retrieved from the U.S. Bureau of Labor Statistics’ website.
Interestingly, the benchmarking raised the household survey’s labor force and employment numbers significantly for each month in 2011 (revisions for 2010 were very slight at 50 or less for both labor force and employment). For example, the revised January 2011 labor force showed an increase of 9,400 over the pre-rebenchmarked figure, while the labor force for May 2011 was increased by slightly above 17,000. The smallest monthly increase in a re-benchmarked labor force occurred in August (up 6,400). For all of 2011, the average monthly upward revision was 9,900.
A similar scenario unfolds with the household employment data. In January 2011, the number of employed increased by more than 8,200 over the originally released data. The largest revision was in May (up 13,700) and the smallest in October (up 5,500). The average monthly change in 2011 was 8,300 over the old data.
Because the 2010 data were basically unchanged the year-over-year growth in both employment and labor force rose each month by the amount of the upward revision produced by the new benchmarks. For each month throughout 2011, the year-over-year labor force figure was much higher than the 2010 figure by an average of 9,400 people. The benchmarking process made that difference much greater at 19,300, a more than doubling of the original increase.
The household employment data show a similar pattern. For every month during 2011, the year-over-year increase to employment was much higher than in 2010 by an average of 19,200. The benchmarking process increased that amount to 27,600. In short, 2011 was much stronger than the original data showed.
However, there is a darker side to the newly benchmarked data. Consider the latest numbers for January 2012. Instead of the 20,000 plus year-over-year gains recorded in the final quarter of 2011, the newly revised data show a loss of around 500 persons in the labor force. Meanwhile, the newly benchmarked data place the employment gain for January 2011 to January 2012 at 4,900, a far cry from the monthly gains in 2011. In sum, while the newly benchmarked data for 2011 reveals stronger gains than earlier reported, 2012 started off on a much weaker note.
The employer payroll survey data corroborates the household survey data. The two employment measures of interest are total non-farm and private sector employment. The benchmarked establishment data also revised the jobs totals in both categories upward. Prior to benchmarking the average year-over-year increase in total non-farm employment for each month in 2011 was just less than 18,000. For total private sector data that average was a bit higher at 19,700. After the numbers were revised, the average year-over-year increase for total non-farm employment increased to just over 23,200 while the increase for total private sector jobs rose to 26,000.
Unlike the household survey data, the revisions to the employer survey data did impact 2010 figures as well as 2011 data. During the first six months of 2010 the adjustments to both total non-farm and total private job levels were less than 1,000. However, the revisions grew larger, reaching an increase of 6,000 in total non-farm jobs in December 2010 while private sector employment was raised by 5,600 in December.
Prior to the recent benchmarking, 2011 data for the household survey showed that the Pittsburgh MSA had made substantial improvements in the number of people employed as well as in the labor force. The benchmarking process resulted in a significant increase to the household survey numbers which was substantiated by the employer payroll survey. The benchmarking did not alter the fact that employment and labor force levels grew during 2011, just at a stronger pace than originally reported. Likewise revisions to employer survey data also showed stronger than previously reported gains. The size of these benchmark revisions reminds us that our current view of what’s happening based on monthly labor market reports may be too weak or too strong and is subject to change depending future revisions.