The President's Pay Agent
BLS has implemented three of the five improvements designed for its National Compensation Survey (NCS) program:
(1) Problems associated with random selection of survey jobs.
Progress: BLS has designed an econometric model that is used to estimate salaries for jobs not randomly selected in a locality survey. NCS program data used for this report include modeled data when survey data were not available.
(2) Matching Federal and non-Federal jobs.
Progress: OPM formed an interagency working group that developed a crosswalk between Federal job classifications and the Standard Occupational Classification system, which BLS uses in its surveys. OPM staff made a few improvements designed to better match certain jobs, and BLS used the crosswalk and March 2004 GS employment data to weight the NCS job data used in this report.
(3) Excluding randomly selected jobs that would be classified above GS-15 in the Federal Government.
Progress: BLS developed methods for identifying and excluding non-Federal jobs that would be classified above GS-15 in the Federal Government. These jobs were excluded from data delivered to the Pay Agent for use in the locality pay program.
Two other improvements are now being introduced into the surveys, but will not begin to be reflected in the survey data until 2006. These are:
(1) Assigning GS grades to randomly selected survey jobs.
Progress: OPM designed and tested a four-factor evaluation system for use in the surveys, and BLS successfully used the new approach in field tests. OPM also developed 20 job family grade leveling guides that cover the range of work under the General Schedule and provide occupation-specific information for use in the surveys. BLS has begun to phase the new approach into its surveys. This improvement will take 5 years to fully implement because BLS conducts detailed job leveling interviews only when it first adds an establishment to its surveys and replaces only 1/5 of its establishment sample each year. However, data used for this report continue to rely on the original NCS use of the primary standard and the 9 factors of the Factor Evaluation System.
(2) Assigning GS grades to randomly selected survey jobs with supervisory duties.
Progress: BLS identified survey establishments where supervisory jobs were surveyed, discussed new collection procedures with its staff, and tested a new method of grading supervisory jobs based on grading the highest level of work supervised. BLS has completed field testing of the new procedures and has begun to use the new approach in its surveys. However, data used in this report exclude 2nd and 3rd level supervisors.
The last two improvements in NCS surveys will begin to affect data delivered in 2006. We encourage BLS and Pay Agent staff to expedite completion of these last two improvements in the NCS program.
Establishments with Fewer than 50 Employees
BLS has expanded its surveys to cover establishments with fewer than 50 employees. Data currently delivered to the Pay Agent exclude these small establishments, but BLS could include them in data deliveries commencing in 2007. The Federal Salary Council plans to study this issue, and we will consider the Council's recommendations before making any changes.