By Howard Risher
May 25, 2011The call for cuts in federal pay by former Minnesota Gov. Tim Pawlenty in his first speech as a Republican presidential candidate this week cemented the prominence of government salaries and benefits as a political issue. His speech was in Iowa, not Washington. His comments were significant also because his experience in state government would not have outfitted him with direct knowledge of the federal pay program or agency staffing. He clearly saw the topic as a way to gain voter support.
Pawlenty's comments were notable also because he offered no evidence to justify his proposal. There are clearly reasons to question the federal pay program, but the true problems have yet to be adequately studied or possible solutions evaluated. His proposed changes would be "putting the cart before the horse." If implemented, they would affect both current employees and federal government's "brand' as an employer for years to come.
This is not to suggest federal employees should not be asked to accept a decrease in current or future income. As Pawlenty argued, government employees "shouldn't get a better deal than the taxpayers." That view has been expressed in different ways at all levels of government. But first let's determine how much better or worse that "deal" actually is.
Federal pay is an easy target. It seems like there are new proposals and rumors about it daily. The result is a highly divisive argument that could go on and on until the facts and goals are more clearly articulated. The closest to a real debate was the hearing chaired by Rep. Dennis Ross, R-Fla., in March, but that was more a series of speeches, not a sharing of ideas or information.
The list of proposals to reduce federal employment costs has grown steadily -- freezes, cuts, increased pension contributions, reduced benefits. But no evidence has been presented to justify the need for change. Proposals are floated with no assessment of the impact on agency operations.
Opponents argue that the impact of cuts on the federal workforce could trigger other problems that offset savings -- but those possibilities have not been explored beyond guesses as to what may unfold. Both the savings and the "costs" must be determined.
The Case for `Total Compensation' Planning
Despite the frequent criticism from politicians, there has been no substantive effort to defend current employee pay or benefits. The only solid information is the series of annual reports released by the Federal Salary Council and the Pay Agent showing federal employees are badly underpaid. Those conclusions have never been accepted by Republican or Democratic presidents, however. Federal benefits have not been systematically studied for more than a decade.
In every other sector, employers rely on salary surveys to compare the pay levels of "benchmark" or generic jobs common to employers competing in the same labor markets. Numerous industry-specific surveys are conducted for that purpose. Employers conduct annual "market" analyses to determine needed adjustments to remain competitive.
Increasingly, as the dollars spent for benefits have increased, companies have looked more closely at "total compensation" -- base salaries, cash incentives, paid time off and benefits -- to determine whether their compensation package is in line with market levels. At the executive level, stock ownership is also included. The "compensation" is the net cost of the benefits after accounting for employee contributions.
A more sophisticated, but less "transparent," approach looks at the value of benefits, using actuarial methods, based on salary, age, years of service and other demographic factors. That highlights the obvious -- benefits are more valuable to older workers. When federal benefit values were analyzed in the late 1990s, the results showed the package was roughly 5 percent higher for older employees. That difference was incorporated in the planning for the 1990 Federal Employee Pay Comparability Act.
The reality is that federal employees are provided with a more comprehensive and by some standards generous total compensation package than is provided by the typical employer. Paid time off is more generous -- and the public is aware of that. Employees are also provided a broader array of insurance coverages, though here the level of required contributions is a key issue in market comparisons.
Government retirement benefits also are more generous, but then again the majority of U.S. employers provide no such benefits. Comparisons that include the millions of mom-and-pop businesses are by any standard misleading. It is also important to highlight the differences in federal careers. The typical employee in the private sector moves from employer to employer several times over a career, and in some respects "starts over" each time. Federal employees can move between agencies and continue to accrue benefits so differences at retirement should not be surprising.
House Budget Committtee Chairman Paul Ryan, R-Wis., recently stated that federal pay and benefits "need to be reformed in line with the private sector". Surveys by The Washington Post and others confirm the public agrees government employees should be paid in line with market levels.
All of this points to the need for a comprehensive analysis comparing federal total compensation levels with employers in the private sector. It needs to account for employee contributions and also for the differences in career patterns. It is not as simple as comparing salaries, but the methodology is well-established.
What Is the Relevant Labor Market?
A key issue is reaching agreement on the labor market(s) relevant to this comparison. Normally employers define their human capital needs, and plan the pay and benefits package to enable them to compete for that talent. That strategy, however, is precluded for Title 5 agencies.
Market analyses are planned to understand compensation levels and practices across companies competing for the same talent. Because industry knowledge is often important, companies tend to rely on industry-specific surveys. That's the reason, for example, hospitals rely on surveys of competing hospitals, banks rely on banking surveys, etc. The larger employers in the Washington area would be a good starting point.
Significantly, the Federal Wage System is based on the market logic. Pay levels for "blue-collar" jobs are determined by local supply and demand. Accordingly, FWS wage schedules have been created for 132 local areas across the country. It's consistent with private sector practice.
The adjustments to the General Schedule system, however, are based on a different rationale, one unique to public employers. GS salaries have never been aligned with market rates. "White-collar" salaries are locked into the internal hierarchy of jobs as defined in Title 5. The Office of Personnel Management has had no reason to conduct a market analysis and is actually precluded from adopting the textbook approach for using survey data. That's vintage 1949 thinking.
That is the core issue that triggers the criticism and perpetuates this debate. It means very simply that some jobs can be overpaid relative to the market while others are underpaid.
But that's not the only reason federal pay levels appear to be high. First, the overall American workforce is not at all similar to the federal workforce. Federal employees are relatively better educated, older and concentrated in "knowledge" occupations. The think tank reports arguing federal pay is too high are based on comparisons that include data from industries like retail, hotels and construction that have little in common with government. Second, the Bureau of Labor Statistics insists on basing its surveys on the broadest possible samples of employers. They actually include employers that have a single employee. Census Bureau data show almost 80 percent have fewer than 10 employees. But they are nonetheless included in the "population" for BLS surveys.
Those "mom-and-pop" businesses are not in the same market for talent as federal agencies. It makes no sense to include those employers in pay analyses. Larger employers are known to pay higher salaries. OPM Director John Berry has commented on this problem.
Some have argued government jobs are never comparable. A number are clearly unique. Border Patrol agents are an obvious example. But some are directly comparable to jobs in state and local government. Others may involve somewhat different duties but require the same education or training, which means federal agencies are competing for talent with industry. And still others are directly comparable -- nurses, editors, chemists, statisticians -- the list goes on. Credible comparisons are possible for most federal jobs.
The Facts Are Needed
It appears to be inevitable that Congress will modify federal pay and/or benefits. See the recent headline -- "Democrats still won't rule out cuts to federal pay and benefits."
This is not uniquely a federal issue. It has been heatedly debated at the state and local level as well. It is also a global issue.
But it would be a grave mistake to reach agreement on those changes without assessing the anticipated impact. A realistic time frame to assemble the facts is six months or less. Precipitous actions could make it difficult to recruit qualified people for critical occupations -- border guards, nurses, cybersecurity -- for years. Agency performance depends on committed employees.
Howard Risher is an independent compensation and performance management consultant. He was the managing consultant for the studies leading to the 1990 Federal Employees Pay Comparability Act. He is the author or co-author of five books, including Planning Wage and Salary Programs (WorldatWork Press, 2009).
By Howard Risher
May 25, 2011