APPENDIX C EXAMINATION OF NON-FEDERAL PAY IN SELECTED COLA AREAS The Office of Personnel Management prepared this appendix specifically for inclusion in this announcement. It has not been approved or endorsed by the plaintiffs or their counsel. It has also not been approved or endorsed by the U.S. Postal Service. SPECIAL ADVICE TO MEMBERS OF THE CLASS IN CARABALLO: This appendix compares non-Federal pay to Federal pay in the COLA areas in the same manner that is used to compare non-Federal pay to Federal pay in the continental U.S. under the Federal Employees Pay Comparability Act of 1990 (FEPCA) (5 U.S.C. chapter 53, subchapter I). Please Note: The proposed settlement in Caraballo would not extend the locality pay provisions of FEPCA to the COLA areas. Only Congress could enact legislation to extend FEPCA's locality pay provisions to the COLA areas. The use of FEPCA's methodology in this appendix to compare non-Federal pay to Federal pay in the COLA areas has no effect whatsoever on COLA rates or the locality pay program. The Bureau of Labor Statistics (BLS) conducted the surveys used for these analyses in 1996. Therefore, the conclusions and tables in this appendix apply only to the period during which BLS conducted its surveys. The conclusions and tables that would result from the use of any other time period or methodology are unknown. EXAMINATION OF NON-FEDERAL PAY IN SELECTED COLA AREAS I. How Did the Research on Non-Federal Pay Evolve? Congress asked OPM to study the pay practices of other employers in the COLA areas. The Bureau of Labor Statistics (BLS) and the Office of Personnel Management (OPM) jointly studied the largest cost-of-living allowance (COLA) areas in terms of Federal white-collar employment. BLS surveyed the areas and collected non-Federal salary data, and OPM analyzed these data within the context of the Federal Employees Pay Comparability Act of 1990 (FEPCA) (5 U.S.C. chapter 53, subchapter I).[Footnote 1] Under FEPCA, the Government pays most Federal white-collar employees in the 48 contiguous States and Washington, DC, rates of pay under the base General Schedule plus a locality-based comparability payment. The purpose of this payment is to set Federal salaries at a level comparable to non-Federal pay in the locality. Inherent in FEPCA is the concept that the Federal Government should have a neutral influence by paying rates that are close to the local labor market. The Government bases locality-based comparability payments on prevailing non-Federal pay within a geographic area. Thirty-two locality pay areas exist within the United States. All but one cover large metropolitan areas. For example, using Office of Management and Budget Metropolitan Statistical Areas (MSAs), the Government defines the Washington, DC, locality pay area as the Washington-Baltimore Consolidated Metropolitan Statistical Area (CMSA). The one area that does not relate to a local major metropolitan area is the Rest of U.S. (RUS) area, which covers areas within the 48 contiguous States that are not part of one of the other locality pay areas. Federal white-collar employees at duty stations outside the 48 contiguous States (i.e., in the COLA areas, other U.S. territories and possessions, and in foreign countries) do not receive locality-based comparability payments. Instead, they receive rates of pay under the base General Schedule. They also may receive an additional amount of pay based on relative living costs and, in some cases, on conditions of environment and other factors, such as political unrest. In order to carry out Congress' request, OPM examined whether it was feasible to conduct comprehensive salary surveys in the COLA areas and what the impact might be if the locality pay provisions of FEPCA were extended to the COLA areas. The Memorandum of Understanding (MOU) provided funds for this research. The research was a two-step process. First, BLS collected non-Federal salary data in selected COLA areas. Next, OPM compared these data with Federal salaries in the same areas. II. What Were the Results of the BLS Salary Surveys in Alaska, Hawaii, and San Juan? In 1996, BLS conducted Occupational Compensation Surveys (OCS) in Alaska, Hawaii, and in the San Juan-Caguas-Arecibo CMSA in the Commonwealth of Puerto Rico. BLS did not survey salaries in Guam or the U.S. Virgin Islands because it did not believe it would be possible to collect a sufficient quantity of statistically reliable data to allow Federal and non-Federal pay comparisons for these areas. BLS conducted special surveys in the COLA areas for the purposes of this report and the MOU process. BLS tabulated and published the results of these surveys in a series of bulletins in early 1997. [Footnote 2] In addition to the Statewide data for Alaska and Hawaii and the San Juan CMSA data, BLS also tabulated and published data for the Anchorage, Alaska, and Honolulu, Hawaii MSAs. [Footnote 3] Approximately 79 percent of all COLA recipients work in these three MSAs, and more than 93 percent of all COLA recipients work in the States of Alaska and Hawaii plus the San Juan CMSA. The Anchorage, Honolulu, and San Juan metropolitan areas have populations of approximately 250,000, 870,000, and 1,200,000, respectively. BLS analyzed the results of its special surveys and compared salary levels in the COLA areas with mainland salary levels. BLS published these analyses in a series of articles in 1998. [Footnote 4] In brief, BLS found that average pay in Alaska tended to exceed average mainland pay significantly. BLS found the opposite in the San Juan CMSA, where average pay tended to be significantly lower than mainland pay. In Hawaii, BLS found varying results depending on occupation and skill level. White-collar pay in Hawaii tended to be fairly similar to, but generally a bit lower than, mainland pay. III. What Were the Results of OPM's Pay Comparability Analyses? BLS did not compare Federal pay with non-Federal pay in the COLA areas. Although FEPCA requires BLS to collect locality pay data, OPM has the responsibility for analyzing the data to determine locality-based comparability payments. Therefore, OPM compared the BLS non-Federal salary data with average Federal salaries for the same locations. OPM applied the same process it uses for calculating locality payments for employees in the 48 contiguous States. (The Annual Report of the President's Pay Agent describes the process for calculating locality pay. [Footnote 5]) In brief, the locality pay process requires comparing non-Federal pay to the salaries of Federal employees in jobs equivalent to those surveyed by BLS. OPM uses various mathematical techniques to refine the comparisons and to provide greater weight to the jobs and groups of jobs that Federal employees typically occupy. The final result is a single percentage, which is the pay disparity for the locality. The pay disparity is the difference between non-Federal pay and Federal pay expressed as a percentage of Federal pay. OPM used the BLS data to compute pay disparities for five locations: State of Alaska; Anchorage, AK MSA; State of Hawaii; Honolulu, HI MSA; and San Juan-Caguas-Arecibo, PR CMSA. Table A shows the results of these analyses. For comparison purposes, the pay disparities for the San Francisco, CA, Washington, DC, and RUS locality pay areas are shown for the same time period. Table A Pay Disparities in Various Geographic Areas as of March 1996 Geographic Area 1996 Pay Disparity 1996 Pay Disparity Without COLA With COLA* State of Alaska 62.42 29.94 Anchorage, AK MSA 59.95 27.96 State of Hawaii 16.39 -3.00 Honolulu, HI MSA 16.09 -5.23 San Juan, PR CMSA -3.53 -12.30 San Francisco, CA CMSA 48.38 NA Washington, DC CMSA 31.16 NA RUS 24.50 NA * Based on actual COLA rates payable in March 1996. Disparity does not account for the exemption of the COLA payment from Federal income taxes. All other factors affecting taxes being equal (e.g., filing status and tax deductions), the after-tax pay disparity would be lower. The table shows that the pay disparity without COLA is substantial in Alaska, somewhat less than the lowest mainland rate (RUS) in Hawaii, and nonexistent in Puerto Rico. For Alaska, the pay disparity suggests that, if the Federal Government were not paying COLA or locality pay, it might encounter significant recruitment and retention problems. For Hawaii, the relatively small pay disparity without COLA means that with COLA, Federal pay exceeds prevailing local labor market rates. The same is true in Puerto Rico. OPM next considered what the locality-based comparability payments might have been in these areas had these areas been covered by FEPCA's locality pay provisions. As stated above, FEPCA's locality pay provisions do not apply in the COLA areas. Only Congress could enact legislation to extend locality pay to the COLA areas. The purpose of OPM's analyses was only to simulate what might have been the locality pay amounts in these areas. The simulated locality pay amounts were 19.57 percent for the State of Alaska and 6.78 percent for the State of Hawaii and San Juan, PR. OPM used three factors to calculate these simulated amounts. The first factor was the 1996 pay disparities without COLA shown in Table A. The second factor was FEPCA's provision for reducing the disparity to not more than 5 percent in each locality, which is referred to as "the disparity to close." [Footnote 6] OPM calculated the disparities to close for the States of Alaska and Hawaii and for San Juan, Puerto Rico. The results are shown in the "1996 Disparity to Close" column of Table B, below. The third factor that determines the size of locality payments is FEPCA's provision for a 9-year phased-in reduction of pay disparities. FEPCA provided that for the first year of locality pay (1994), the reduction factor was 20 percent (i.e., 20 percent of the disparity to close). In subsequent years, FEPCA provided that an additional one-tenth of the remaining disparity would be closed. Under this timetable, the pay gaps were to be reduced to within 5 percent of the local market pay rates by January 2002. Except for 1994, the FEPCA timetable has not been followed. Congress and the President have delayed implementing locality payments for various reasons. As a result, the phase-in factor as of January 2000 is approximately 33.5 percent of the disparity to close, instead of the 80 percent provided by the FEPCA timetable. OPM thus applied the 33.5 percent phase-in factor to the pay disparities to close for the three COLA locations. The results are shown in the "2000 Simulated/Actual Locality Payment" column of Table B, below. (Note: The column heading is "Simulated/Actual" because the locality pay values are simulated for the three COLA locations, while the actual values are shown for San Francisco, Washington, DC, and RUS, which are included for reference.) Table B Simulated Locality Payments in COLA Areas and Actual Locality Payments in Selected Areas as of January 2000 Geographic Area 1996 Disparity Disparity to Close 2000 Simulated/Actual to Close Updated to 2000** Locality Payment State of Alaska 54.69 58.41 19.57 State of Hawaii 10.85 13.51 6.78** San Juan, PR 0.0 0.0 6.78** San Francisco, CA, CMSA 41.31 44.51 15.01 Washington, DC, CMSA 24.91 26.93 9.05 RUS 18.57 20.64 6.78 * OPM updated to 2000 the 1996 disparity-to-close values for the States of Alaska and Hawaii and for San Juan, PR, to account for pay changes. To update Federal pay, OPM used the General Schedule adjustments since 1996. To update the BLS data, OPM used the change in the nationwide Employment Cost Index (ECI), White-Collar Workers, Less Sales, published by BLS. The ECI measures the change in non-Federal employer payroll costs over time. OPM used nationwide ECI data because BLS does not publish ECI information for the COLA areas. ** The Pay Agent's current policy is that no locality pay rate will be lower than the RUS locality pay rate. The simulated results for Hawaii and San Juan indicated that locality pay for those areas would have been less than the RUS rate. Therefore, the RUS rate is shown in the table above. In summary, the research discussed in this appendix shows that BLS was able to collect a sufficient quantity of statistically reliable data on non-Federal salaries in certain COLA areas to allow a comparison with Federal salaries in those areas. In addition, the research shows that if locality pay were extended to the COLA areas, the only area that would have a locality-based comparability payment greater than RUS would be the State of Alaska. ----- FOOTNOTES 1. In March 1997, BLS and OPM briefed the Safe Harbor Working Group on the results of the surveys and OPM's analyses. The survey publications previously cited and OPM's analyses were provided to the Working Group at that meeting (MOU-170). OPM briefed the Working Group again in July 1999 on locality pay issues and provided updated materials at that time (MOU-549). This appendix synthesizes and documents the verbal and written information provided at those meetings. 2. U.S. Department of Labor, Bureau of Labor Statistics, Occupational Compensation Survey: Pay Only, State of Alaska, Bulletin 3085-32, July 1996; State of Hawaii, Bulletin 3085-37, August 1996; San Juan-Caguas-Arecibo Consolidated Metropolitan Area, Puerto Rico, Bulletin 3085-44, October 1996. 3. U.S. Department of Labor, Bureau of Labor Statistics, Occupational Compensation Survey: Pay Only, Anchorage, Alaska, Metropolitan Area, Bulletin 3085-30, July 1996; and Honolulu, Hawaii, Metropolitan Area, Bulletin 3085-34, August 1996. 4. U.S. Department of Labor, Bureau of Labor Statistics, Compensation and Working Conditions, "How Do Wages in Alaska Compare to Wages on the Mainland?" Fall 1998, pp. 18-24, Fall 1998; "How Do Wages in Hawaii Compare to Wages on the Mainland?" Summer 1998; pp. 31-40; "How Do Wages in San Juan Compare to Wages on the Mainland?" Winter 1998, pp. 23-29. (These documents were provided to the Safe Harbor Working Group in MOU-495, 496, and 497.) 5. OPM publishes the Annual Report of the President's Pay Agent--Report on Locality-Based Comparability Payments for the General Schedule. The President's Pay Agent consists of the Director of OPM, the Director of the Office of Management and Budget, and the Secretary of Labor. The Pay Agent's report describes in detail the methodology OPM applies in its analyses. 6. For example, if the pay disparity for a locality pay area is 30 percent, the disparity to close for that area is 23.81 percent (i.e., 1.30 / 1.05 = 1.2381).