How the GS System Fails Federal Workers and Everyone Else

If government could start with a clean page to plan a new pay system, it would never consider the General Schedule program model.

The Yankees’ Yogi Berra was right: “It’s like déjà vu all over again.” In December 2017, the government’s Pay Agent—the entity composed of the Labor secretary, the director of the White House Office of Management and Budget and the director of the Office of Personnel Management—reported “major methodological concerns” that undermine the credibility of the annual pay gap analysis calculated by the Federal Salary Council. The report noted the analysis “fails to reflect the reality of labor market shortages and excesses,” requires a single, average pay gap calculation covering diverse occupations and labor markets, and ignores the value of benefits.

This year’s delayed report cited the same problems using essentially the same statements. The question is: Will next year’s report repeat the same problems? With four of the seven members of the Federal Salary Council representing unions, which have opposed changes to the methodology used to calculate the gap, next year’s report could be a carbon copy.

The fact is the concerns are solidly justified. The methodology has been pieced together over two decades, with statistical bandaids added and replaced over time. It is best seen as a Rube Goldberg contraption that I defy any member of the Council to describe without notes. If government could start with a clean page to plan a new pay system, it would never consider the General Schedule program model.

The differences in this country’s local and regional labor markets are vast. Pay levels for similar jobs can vary by 50 percent or more from high pay to low pay areas. Very different labor markets exist but are ignored within locality pay areas. The Washington-Baltimore-Arlington locality area, for example, extends for roughly 180 miles from areas of West Virginia, dominated by low pay industries, to the shorelines along the Chesapeake Bay in eastern Maryland. It defies common sense to consider that area as a single, homogenous labor market. Likewise, arguing that the pay differentials across the country can be summarized in a single percentage is simply not credible.

A core problem is that the Bureau of Labor Statistics no longer conducts surveys designed to report the prevailing market pay for specific jobs. At one time BLS did conduct a so-called benchmark survey, reporting market rates nationally and in each locality area, but it was terminated in the mid-1990s.  

That’s when the problems started. No president has ever agreed to close the “pay gap” and there is no reason to believe that is going to change. Current BLS surveys were not planned for this purpose, and after trying for two decades, there is no reason to think complex statistical models will produce credible answers.

Existing BLS surveys are useless in studying the pay for jobs requiring specific skills. That’s also true for understanding the pay of managers or executives. The data are also useless for studying pay levels in specific types of organizations (e.g., nurses working in hospitals vs nursing homes).  

Congress Must Intervene

There is no other employer in the country that would rely on the current methodology. No other employer uses the BLS surveys for this purpose. The universal practice is to rely on benchmark surveys.

But the barrier is that the Federal Employee Pay Comparability Act of 1990 created the Federal Salary Council and requires the use of BLS survey data. When the law was enacted, nothing like the current methodology was ever anticipated. (I was involved in the planning meetings.)  New legislation will be required before this problem can be solved.

Both of the recent Pay Agent reports refer to a 2017 report from the Congressional Budget Office that concluded federal employees are paid more than similar private sector employees, especially when benefits are included. However, that report also relies on data (Current Population Survey) collected for a different purpose and focuses on broad, general employee data (e.g., age, education level). The data fail to provide relevant information for comparing jobs requiring specific skills, college majors or relevant job experience.  

For salary planning, the CBO methodology is even more useless than the FEPCA dictated approach. No employer would consider adopting it. It is completely divorced from the reality of labor markets.

The value of federal benefits is a relevant consideration (although never considered in drafting FEPCA). Again, however, CBO’s broad brush comparison is misleading. Yes, federal retirement benefits are more generous. But credible comparisons should reflect government’s older workforce and longer years of service. More importantly, the comparison should focus on employers that have something in common with the government’s workforce—not the millions of mom-and-pop businesses that provide little more than legally required benefits. Benefits are normally evaluated across employers competing for talent.

Both the CBO and BLS surveys fail to report income from performance incentives or from stock ownership. For executives, managers and many professionals, ignoring the income from these common pay components significantly understates their total compensation.

For reasons that are unstated, the Pay Agent report commends “the Council’s continuing evaluation of analytic methods used in the locality pay program, most notably the Council’s plan to undertake a thorough review of the program’s salary survey methodology.” Realistically no one on the Council has the experience to evaluate the current methodology or recommend an alternative approach.

A better answer would be to rely on experts from the consulting and academic worlds to both assess the current methodology and make recommendations for the future. Their independence and objectivity is important. A similar group played a role in the planning for FEPCA in 1990. It is essential to start with agreement on the facts.

A Credible Approach

The textbook approach to using survey data starts with staffing and workforce planning. Employers need to identify the skills needed for future operation, to assess the availability of those skills in the labor market, and to analyze pay survey data to plan the compensation needed to hire and retain employees with essential skills. That is the accepted basis for salary planning.  

None of that is reflected in the administration of the General Schedule pay system. Workers in the post World War II years, when the GS system was adopted, stayed with the same employer for their careers. The focus of wage and salary management was on maintaining internal pay relationships—internal equity. It was not until the 1950s that pay surveys were introduced with an initial focus on executive compensation. Today, with labor shortages and high demand for specific skills, a more flexible, responsive pay system is essential.

At this point, after all the years of reported pay gaps, the biggest hurdle is likely to be convincing employees that “fair pay” can no longer be defined as it was in 1949, and that a new approach to planning and managing salaries will be better for everyone. That will require leadership that to date has been missing. If employees believe the planning process is contrary to their career interests, it will trigger costly declines in engagement.

The new Pay Agent report concludes with a succinct, well stated argument for replacing the GS system:

“It is important to appropriately compensate personnel based on mission needs and labor market dynamics. The existing GS compensation system fails in this regard. The president’s budget . . .  [proposes] “to realign incentives by enhancing performance-based pay and slowing the frequency of tenure-based step increases.

“Ultimately, we believe in the need for fundamental reforms of the white-collar federal pay system. We believe it is imperative to develop performance-sensitive compensation systems that will contribute to a government that is more citizen-centered, results-oriented, and market-based. We need to empower federal agencies to better manage, develop, and reward employees in order to better serve the American people.”

That is consistent with the pay philosophies in well-managed, successful organizations. It should be the basis for compensating federal employees.