U.S. Office of Personnel Management - Ensuring the Federal Government has an effective civilian workforce

Locality Pay Surveys

Under FEPCA, we must use salary surveys conducted by the Bureau of Labor Statistics (BLS) to set locality pay. Commencing with the 1996/97 surveys, BLS implemented a new survey design for its salary surveys. The new survey program, called the National Compensation Survey (NCS) program, was used in all BLS salary surveys started after September 1996. NCS uses probability sampling of occupations within survey establishments, rather than a fixed job list with detailed job descriptions, as had been used in the past.

The new survey process was not immediately accepted for use in the locality pay program. In fact, the Federal Salary Council recommended that the original NCS methods not be used to set Federal pay. The Pay Agent also concluded that certain major aspects of the NCS program would have to be improved before it would be prudent to use NCS data for making pay comparisons under the locality pay program. In 2002, Pay Agent and BLS staff implemented three of the five planned improvements in the NCS program, and the Federal Salary Council recommended that we begin to phase in the use of NCS data to set locality pay. Since the 2005 report (for locality pay in 2007), we have used only NCS survey results for the locality pay program.

Four of the five NCS improvements are fully incorporated into surveys used this year:

1. The linkage of Federal and non-Federal jobs by developing a crosswalk between General Schedule occupations and the Standard Occupational Classification (SOC) System to permit weighting data by Federal employment.
2. The development of methods to identify and exclude survey jobs that would be graded above GS-15 in the Federal Government.
3. The development of an econometric model based on survey data to estimate salaries for jobs not found in the probability samples.
4. The development and implementation of better methods for grading supervisory jobs selected by probability sampling.

BLS continues to phase in the last improvement, which is the use of a four-factor job grading system with job family guides, as it replaces a portion of its establishment sample each year. BLS replaces all of its State and local government sample at the same time approximately every 10 years, and the private industry sample is replaced over a 5 year period. This improvement will be completed in all survey establishments in surveys conducted in 2010 and delivered in 2011. It is designed to improve grade leveling under the NCS program. All of the improvements are described in the 2002 Pay Agent's Report to the President.

Industrial and Establishment Size Coverage

As required by FEPCA, BLS salary surveys used for the locality pay program include the collection of salary data from private industry and State and local governments, which have large numbers of workers, especially in certain occupations that are unique to government functions. Before 1991, BLS surveys for the pay comparability process covered only private sector goods-producing and service-producing industries.

BLS delivered two sets of data this year, data covering establishments with 50 or more workers and data covering all establishments with one or more employees. For establishments with 50 or more workers, BLS surveyed a total of 14,659 establishments. In the 30 separate metropolitan locality pay areas (excluding Raleigh), BLS surveyed 8,122 establishments. The Rest of U.S. (RUS) locality pay survey covered 182 areas, including 77 additional metropolitan areas, 22 micropolitan areas, and 83 non-metropolitan counties or county clusters. A total of 6,537 establishments were surveyed in the RUS area. The Raleigh survey was discontinued in 2004, but is being reinstated during BLS' six-year transition to a new sample of areas.

BLS surveyed a total of 21,362 establishments in surveys covering establishments with one or more workers. In the 30 separate metropolitan locality pay areas (excluding Raleigh), BLS surveyed 11,262 establishments. A total of 10,100 establishments were surveyed in the RUS area.

The number of areas surveyed in the Rest of U.S. locality pay area increased from 118 to 182 areas. The NCS program is undergoing a six-year transition from a sample of areas based on the Office of Management and Budget (OMB) December 1993 metropolitan area definitions to a new sample of areas based on the December 2003 OMB area definitions. The NCS program is phasing in new metropolitan and micropolitan areas as defined by OMB and county clusters defined specifically for the NCS; at the same time, some areas under the December 1993 OMB definitions are being phased out of the sample.

The industry scope of the surveys includes private goods-producing industries (mining, construction, and manufacturing); private service-providing industries (trade, transportation, and utilities, information, financial activities, professional and business services, education and health services, leisure and hospitality, and other services); and State and local governments. Agriculture, forestry, fishing and hunting, and private households were excluded.

Occupational Coverage

Under the NCS program, BLS uses random sampling techniques to select occupations for survey within an establishment. The occupations are selected and weighted to represent all non-Federal occupations in the location and, based on the crosswalk published in Appendix VII of the 2002 Pay Agent's report, also represent virtually all GS employees. OPM provided the crosswalk between GS occupational series and the Standard Occupational Classification (SOC) system used by BLS to group non-Federal survey jobs. OPM also provided March 2007 GS employment counts for use in weighting survey job data to higher aggregates. (BLS completed delivery of the most recent NCS surveys in July 2008, before March 2008 GS employment data became available.)

Matching Level of Work

In the NCS surveys, BLS field economists cannot use a set list of survey job descriptions because BLS uses a random sampling method and any non-Federal job can be selected in an establishment for leveling (i.e., grading). In addition, it is not feasible for BLS field economists to consult and use the entire GS position classification system to level survey jobs because it would simply take too long to gather all the information needed. This would also place an undue burden on survey participants.

To conduct grade leveling under the NCS program, OPM developed a simplified four-factor grade leveling system with job family guides. These guides were designed to provide occupational-specific leveling instructions for the BLS field economists. The four factors were derived and validated by combining the nine factors under the existing GS Factor Evaluation System. The factors were validated against a wide variety of GS positions and proved to replicate current grade levels.

The job family guides cover the complete spectrum of white-collar work found in the Government. BLS has been using the guides in its ongoing surveys and roughly 47 percent of the data this year are leveled under the new approach.¹ Fully implementing the new leveling system will take 3 more years because of BLS' data collection cycle, where BLS conducts detailed interviews when establishments are added to the survey sample. A new government sample was completed in July 2007, and new private industry sample members will be completed by July 2011. See Appendix IV of the 2002 Pay Agent's report for a summary of the BLS data collection cycle. Appendix VI of the 2002 Pay Agent's report contains the job family leveling guides.

Jobs above GS-15

For the NCS program, it was necessary to develop generic instructions for identifying white-collar jobs in the random surveys that would be graded above GS-15 (above the highest grade in the General Schedule) if they existed in the Federal Government so that the data could be excluded from pay gap measurements. BLS developed and tested the guidance with assistance from OPM. Appendix V to the 2002 Pay Agent's report explains the process for identifying these jobs in the NCS program.

Grading Supervisory Positions

Grading supervisory jobs presented another problem for the NCS program because the Government does not use the FES approach to grade supervisory jobs. OPM occupational classification specialists suggested an approach based on the highest level of work supervised. Under the this approach, BLS grades the highest level of work supervised using the appropriate four-factor leveling guide, not the supervisory job itself, and then adds one grade for a first-level supervisor, two grades for a second-level supervisor, and three grades for a third-level supervisor.

Missing Data

While BLS surveys all white-collar jobs under the NCS program, it does not find all jobs at all work levels in each survey area. This is a serious problem with the NCS program because survey results and pay disparity measures can vary considerably based on which jobs are included. The Pay Agent asked BLS to develop an econometric model to provide estimates for jobs not found in NCS. The model is described later in this report and in Appendices II and III.

Establishment Size

BLS delivered data for both establishments with 50 or more workers (large establishments) and all establishments, i.e. including establishments with as few as one employee (small establishments). Establishments with no employees (single entrepreneur owners) are not covered by the surveys. Since locality pay began in 1994, we have used only data from large establishments in the locality pay program.

BLS defines an establishment as an economic unit that produces goods or services, usually at a single physical location, and is engaged in one or predominately one activity. BLS defines a firm as a legal business, either corporate or otherwise, and may consist of one establishment, a few establishments, or even a very large number of establishments. Hence, large firms can have small establishments if there were fewer than 50 employees at the site. Therefore, the pay practices at "small establishments" reflect the pay practices of large and small firms. BLS estimates there are 4.7 million firms in the United States with fewer than 50 employees that employ about 29 percent of full-time workers and only 248,000 firms with 50 or more workers that employ 71 percent of full-time workers, so most of the small establishments BLS selects to survey are likely also small firms.

Table 1.
Pay Gaps and Percent Modeled Data by Establishment Size

The pay gaps in Table 1 including data from small establishments are actually higher, on average, than those limited to data from establishments with 50 or more employees. The typical pattern in the data was for pay levels to be lower in small establishments than in large establishments for lower graded jobs but higher for higher graded jobs. BLS found many relatively high pay management and physician jobs in higher grade levels in small establishments. Because we use Federal employment to weight the data and there are many higher graded Federal jobs, using data from small establishments increases the pay gaps.

Including data from small establishments increases the number of non-Federal employees represented by the data since about 29 percent of non-Federal workers are employed in small establishments. It also slightly reduces our reliance on modeled data, with about 1.7 percent more Federal employees represented by survey data rather than modeled results. After reviewing the data from small establishments in 2007 and 2008, the Federal Salary Council concluded we should begin using the data from all establishments, small and large, for locality pay in 2010. We agree with the Council's recommendation and the remainder of this report is based on data from all establishments.

State and Local Government Resampling

BLS replaces all of its State and local government sample at the same time approximately every 10 years. This is different than its sample replacement for private industry where 1/5 of the establishments are replaced each year. BLS believes that more frequent but gradual sample replacement is not necessary in State and local governments because "the Government sector is more stable in terms of new establishments coming into existence or establishments going out of business. Also, response rates are higher within the Government sector..."

OPM staff noted some substantial variations in this year's data for average salaries by occupational category and grade level. We use PATCO (Professional, Administrative, Technical, Clerical, and Officer) categories to group occupations. Many of these variations are in the Officer category where much of the data is for jobs common to State and local governments. (The Officer category includes jobs such as correctional officer, border patrol agent, police, and fire protection.) For example, OPM staff noted the following changes in average salary between the data used in 2007 and the data used in 2008 for the Officer category.

• Denver: GS-5 plus 25 percent
• Los Angeles: GS-6 plus 27 percent
• Los Angeles: GS-7 minus 24 percent
• Memphis: GS-7 plus 21 percent
• Miami: GS-6 plus 30 percent
• New York: GS-4 plus 30 percent
• Portland: GS-5 minus 21 percent
• Sacramento: GS-4 plus 31 percent
• Sacramento: GS-6 plus 79 percent

BLS attributed these changes to randomly selected jobs in State and local governments rotating in or out of the sample. In some cases, a single job was identified as causing most of the change. These changes are far too large to be due to actual changes in salary levels in the locality. (As a point of comparison, the ECI increased at an annual rate of 3.2 percent across all jobs from March 2007 to March 2008.) The variability in survey results may be due to survey samples that are too small compared to the range and variability of salaries found within an occupational group and grade level, coupled with replacement of the entire State and local government sample at one time.

The Federal Council has recommended that BLS samples should be increased to improve the surveys and this is another indicator we need larger samples to enhance the credibility of the survey results. Since BLS has already pulled a new government sample for the next 10 year cycle, BLS' sample rotation for State and local governments won't reveal instability of the pay measures for another 10 years. Nevertheless, we agree with the Council that it would be desirable to increase survey samples. However, under current budget limitations, it appears that BLS' sample in locality pay areas will be reduced by about 9 percent. Increasing the NCS sample in existing locality pay areas must also be viewed in light of the Council's other desire to increase the number of separate locality pay areas, which is a competing goal for scarce survey resources. At present, there are no funds available to increase survey samples or conduct locality pay surveys in additional areas.

Other Variations in Survey Results

OPM staff also identified and asked BLS about several other large swings in the survey results. These included a 23 percent decrease in the average salary for GS-5 Clerical employees in Huntsville where BLS' response indicated "a high-paid, high impact job in private industry was abolished due to company restructuring"; a 22 percent increase in the average salary for GS-7 Clericals in Seattle where BLS' reply was that "a relatively low-paid job no longer contributes, since the local government schedule rotated out". These are additional indicators that our sample size may be too small to produce stable estimates, at least for some occupational categories/grade levels.

Effect of Incentive Pay on the Rest of U.S. Pay Gap

Another substantial change discussed by OPM and BLS involved a 45 percent increase in the estimate for the GS-12 administrative category in the Rest of U.S. (RUS) locality pay area. This increase was unusual because it involved the RUS area, which includes the largest sample since it is a composite of many surveys. Based on information provided by BLS, the estimate increased by 45 percent mainly because it was derived in part from sampled data for a job in one of the many surveys conducted for the RUS locality pay area that received uncommonly high earnings (base salary plus incentive pay) of more than $1 million.

BLS excludes bonuses and other payments such as premium pay from the survey results used for the locality pay program. However, incentive pay, defined by BLS as payments for meeting job goals where the formula is clearly known by both the employee and the employer beforehand, is included in our estimates for any job where it's the practice of the surveyed establishment to determine pay based on a production driven formula. To the extent such payments were used in jobs surveyed, incentive payments have been included in BLS data used for setting GS pay since the 1970s. These payments are generally included as income for tax purposes, sometimes included as income for annuity computations, and generally not included as base pay for subsequent years. Employees under the General Schedule are eligible for bonuses but generally do not receive payments equivalent to incentive pay in the private sector.

While incentive payments have been included in the surveys for years, this is the first time a large swing in survey results has been attributed to incentive payments. Large fluctuations such as this one cause instability in the pay measures, and for 2008, would result in pay gaps in five locality pay areas (Cincinnati, Dayton, Indianapolis, Raleigh, and Richmond) below that for the RUS locality pay area. Some of the Council's Working Group members questioned whether such windfall payments should be included in the pay comparisons used to set Federal pay.

The Council asked BLS to review the data for other categories highly affected by incentive payments in this year's data. BLS reported that administrative GS-8 in Chicago increased 12.3 percent and clerical GS-3 in San Francisco decreased 16.9 percent mainly due to jobs receiving incentive pay. Both of these categories have very low weights in the pay gap calculations. BLS also summarized that 4.7 percent of the weighted workers in our GS to private sector job matches receive incentive pay.

OPM staff recomputed the RUS pay gap using the data supplied by BLS for GS-12 administrative jobs last year aged to 2008. The pay gap with the GS-12 incentive pay is 35.32 percent, but with last year's GS-12 administrative data aged to 2008 it would be 29.34 percent. This is a difference of 5.98 points mainly due to the effect that uncommonly high incentive pay in one surveyed area has on the GS-12 administrative category estimate.

Chart 1
Administrative Jobs in RUS
June 2006 and July 2007

Chart 1 shows the magnitude of the anomaly in the RUS data for the GS-12 administrative category in this year's survey compared to data delivered last year.

BLS also stated that it originally published a "less sales" occupations ECI to isolate the potential volatility of incentive payments on the ECI but revised that to a "less incentive data" ECI in 2006 because incentive payments were occurring in non-sales occupations. BLS states the way incentive payments are recorded in its database would make it difficult to exclude the payments, "base" salaries for jobs receiving incentive payments may be lower than otherwise would be the case, and that if the Council or the Pay Agent wishes to exclude incentive payments, it would be easier to exclude all the workers receiving such payments.

As pointed out by the Council, we have not discussed the suitability of including incentive payments since locality pay began in 1994. Likewise, there are no established procedures for dealing with outliers in the data. If we were to adopt such procedures for general use in the future, we would develop them after considering the Council's recommendations on the subject. In the meantime, OPM staff suggested using last year's data appropriately aged for the GS-12 administrative category in the RUS area in lieu of the current survey data influenced by high incentive earnings.

The Council recommended that we should use the data as delivered by BLS, including the incentive pay data. While the Pay Agent is pleased to accept the Council's related recommendation to further study incentive pay and outliers in the survey data, we respectfully disagree with the Council about including the incentive pay data for GS-12 administrative jobs found in the RUS survey this year.

The data and survey results are clearly influenced by an extreme outlier that represents salary levels that are more than ten times the typical salary found at the grade. Including this outlier would result in extreme fluctuations in the RUS pay gap from 2007 to 2008 and likely from 2008 to 2009 if the company making the payments is no longer surveyed or if smaller incentive payments are authorized in the future. Five separate locality pay areas have measured pay gaps below that for the RUS area if the data are included and the RUS locality rate authorized for 2010 would be substantially higher than otherwise warranted if these data are included. Such a higher locality rate for the RUS area would be at the expense of locality pay rates that could otherwise be approved in the other, generally higher paying, locality pay areas.

Therefore, we instructed our staff to replace the GS-12 administrative data for the RUS area with last year's data aged to March 2008 for the pay gaps included in the remainder of this report. These data and this technique were discussed with the Council at its meeting of September 30. While it might have been technically more correct to have BLS remove the data in question, rerun its pay model, and resubmit the data, such efforts would have been time consuming, have had an impact on BLS' workload, and were not available for discussion with the Council at its meeting of September 30, 2008.

The President will have the benefit of the Council's recommendations, which are shown in Appendix I, and include the incentive pay data as delivered by BLS. But, it is our recommendation to the President that the GS-12 administrative data for this year's RUS survey not be used in the pay comparisons.