September 14, 2010 | WebMemo on Labor
The Heritage Foundation has conducted two separate studies that both reach the same conclusion: Federal employees are paid substantially more than comparably skilled private sector workers. Defenders of the federal pay system, including the U.S. Office of Personnel Management (OPM), have mischaracterized The Heritage Foundation’s analyses by suggesting they ignore skill differences between the public and private sectors, resulting in an “apples to oranges” comparison. On the contrary, Heritage has carefully accounted for skill differences, always comparing apples to apples.
While federal employees do earn more partially because they are more skilled than the average private sector worker, controlling for skills does not eliminate the federal pay premium. Depending on the methodology employed, the average federal employee receives as much as 22 percent more in wages than an equally skilled private sector worker. Including both wages and benefits, overpaying federal workers costs taxpayers approximately $40–50 billion per year.
The Heritage Studies
Heritage analyst James Sherk recently published a detailed comparison of federal and private pay in a report from The Heritage Foundation’s Center for Data Analysis. Separately, Heritage analyst Jason Richwine worked with Andrew Biggs of the American Enterprise Institute on a similar study, the results of which were published in The Wall Street Journal. The studies employ different datasets and methodologies, but they draw similar conclusions: Federal workers are overpaid in both wages and benefits at substantial cost to taxpayers.
Federal Pay Debate
Defenders of the federal pay system have responded by arguing that these studies ignore differences in skills and occupations. OPM Director John Berry summarized this view in a recent statement:
The Cato Institute and USA Today stories quoting Cato staff (and similar statements from the Heritage Foundation) look only at gross averages, including retail and restaurant service workers and other entry-level positions that reduce private sector average pay in comparison to the Federal average, which does not include many of these categories in its workforce. The Federal workforce today is highly specialized.
This critique mischaracterizes The Heritage Foundation’s research on federal pay, which expressly controlled for education, experience, and other observable characteristics to make an apples-to-apples comparison between federal employees and private sector workers.
Controlling for Skills and Occupation
Regression analysis allows researchers to break pay differences between groups down into two portions. One is explained by differences in observed characteristics (such as education and experience), while the other contains differences these characteristics do not explain. The unexplained difference in pay estimates how much private sector workers would earn with their measured skills if they worked for the federal government.
Both the Sherk study and the Richwine–Biggs study used regression techniques to compare the wages of federal employees and private sector workers, controlling for variables such as age, education, marital status, race, gender, size of the metropolitan area, and several others.
Federal Employees Earn More Than Comparable Private Workers
Table 1 below shows the results from the Sherk study. A portion of federal workers’ higher cash pay comes from their higher skills, but the government pays the average federal employee 22 percent more per hour than accounted for by observable skills. Restricting the analysis to workers in identical occupations in both sectors still shows a federal hourly wage premium of 19 percent in the Sherk study. In their own analysis, Richwine and Biggs found a premium of 12 percent.
Including the value of non-cash benefits such as health care and pension benefits adds to the federal advantage in total compensation. An apples-to-apples comparison shows that the federal pay system gives many federal workers significantly more compensation than they would get in the private sector. The total premium costs taxpayers $40 billion (according to Richwine and Biggs) or $47 billion (Sherk) per year above market rates.
Not an Isolated Finding
This is not an isolated finding. Academic research consistently shows that the federal government pays more than the private sector. Liberal and conservative economists alike find the same thing. Alan Krueger, the current Assistant Secretary of the Treasury for Economic Policy in the Obama Administration, wrote one of the seminal research papers on federal pay. He compared wages of similar private sector and federal workers and also examined how the wages of workers who joined the federal government changed. In both cases, federal employees earned higher pay.
Not All Federal Workers Are Overpaid
Not all federal employees are overpaid, of course. The government’s seniority-based pay scale divorces compensation from individual performance. Many highly skilled federal workers earn no more (and perhaps even less) than they would in the private sector. Consequently, Congress should not uniformly reduce the pay of all federal employees. Instead, Congress should expand outsourcing to the private sector and replace the General Schedule with a performance-based pay system tied to market compensation. Congress should also bring federal benefits in line with market standards.
Reform the Federal Pay System
Defenders of the federal pay system, including the OPM, argue that federal employees earn higher wages because they perform more skilled work. This is only partially correct. Skill differences explain only a portion of the federal government’s disparate level of pay. After controlling for skills and occupation, the federal government still pays its employees substantially more than they would earn in the private sector. According to two separate Heritage studies, the current federal pay system:
Clearly, reform is needed.
Overtaxing All to Benefit a Few
Congress should not overtax all Americans to overpay workers in the federal civil service. Congress should bring equity to federal pay by replacing the General Schedule with performance-based pay, requiring federal agencies to compete with the private sector, and bringing benefits to market levels.
James Sherk is Senior Policy Analyst in Labor Economics, and Jason Richwine is Senior Policy Analyst in Empirical Studies, in the Center for Data Analysis at The Heritage Foundation.
James Sherk, “Inflated Federal Pay: How Americans Are Overtaxed to Overpay the Civil Service,” Heritage Foundation Center for Data Analysis Report No. 10-05, July 7, 2010, at http://www.heritage.org/Research/Reports/2010/07/Inflated-Federal-Pay-How-Americans-Are-Overtaxed-to-Overpay-the-Civil-Service.
Jason Richwine, “The Government Pay Bonus,” Heritage Foundation Commentary, July 6, 2010, at http://www.heritage.org/Research/Commentary/2010/07/The-Government-Pay-Bonus.
More specifically, the studies use data from different parts of the Current Population Survey (CPS), a monthly survey that polls a large sample of Americans about, among other things, how much income they earned “last week.” The CPS also publishes an annual demographic supplement that asks about income earned “last year.” Sherk used last-week data, while Richwine and Biggs used last-year data.
Richwine and Biggs used a single regression to estimate the average difference in pay. Sherk used two regressions—one for private workers and one for federal—to further analyze how the pay difference changes at different skill levels. For more details about the two-regression method (specifically, the “Oaxaca Decomposition”), see James Sherk, “Inflated Federal Pay,” Appendix A.
The set of control variables is slightly different in each study, owing to data availability and researcher preference.
This is a reproduction of Table 4 from the CDA report referenced in footnote 5.
Note that the unexplained wage difference is in addition to the explained difference in wages. A worker in the federal government earns 28.5 percent more per hour than the average private-sector worker because of his greater skills and earns an additional 22.0 more in addition to that because of the higher pay in government. The total federal premium is thus 1.285 * 1.220 = 1.5677, or about 57 percent more than the hourly wages of the average private-sector worker.
For an overview of this literature, see Robert Gregory and Jeff Borland, “Public Sector Labor Markets,” in Handbook of Labor Economics, ed. Orley C. Ashenfelter and David Card (Amsterdam: Elsevier, 1999), Vol. 3A, Chap. 31.
Alan B. Krueger, “Are Public Sector Workers Paid More Than Their Alternative Wage? Evidence from Longitudinal Data and Job Queues,” in Richard B. Freeman and Casey Ichniowski, eds., When Public Sector Workers Unionize (Cambridge, MA: National Bureau of Economic Research, 1988), at http://www.nber.org/chapters/c7910.pdf (August 25, 2010).