Agencies need to consider alternative personnel systems

It’s hard to decide where to start. Human capital management has been on GAO’s High Risk list for more than 20 years. Recent GAO reports have focused on several agencies where staffing shortages and challenges are causing operational problems. Staffing problems have also been discussed in recent Government Executive columns. Last week it was FEMA’s problems. A search for columns discussing staffing challenges found close to 5,000.

A key point is that staffing problems exist in very different federal agencies and involve very different job series, from VA nurses to prison correctional officers. Starting in the 1980s agencies solved staffing problems by replacing the General Schedule with an alternative personnel system. But for some reason the options have not been considered for years. 

Now COVID and the pandemic have changed the country’s labor markets along with worker expectations. The heavy retirements – roughly 100,000 each year – and declining pool of young workers suggest the workforce problems will get worse. Simply stated, the GS system’s rigid structure, the annual adjustments, and related workforce practices are barriers to competing for talent in the “new world of work.”

McKinsey captured the problem in a recent column, although the focus was the private sector:

“  . . . a management system based on old rules -- a hierarchy that solves for uniformity, bureaucracy, and control -- will no longer be effective. Taking its place is a model that is more flexible and responsive . . .”

They argue leaders “should do nothing less than reimagine the basic tenets of organization.”

A new model for government is still to be developed but history and the current problems make it clear the GS system is not the answer.

Today’s Tight Labor Markets Add to the Problems

Labor markets across the country are historically tight. Hiring in March “soared past expectations.” Employers in every sector have experienced worker shortages. While the number of vacant jobs has steadily declined, the total in January remained close to 8 million. There were 900,000 vacant jobs in the public sector. That’s more than double the total at the end of 2010. 

In the most recent BLS report, there were 6.5 million unemployed – only 0.8 job seekers for each vacancy. Prior to 2018 there were always more unemployed workers than job vacancies – at times there were five or more unemployed workers per job opening.

But there are more relevant statistics. In March the unemployment rate for college graduates was 2.1%, with only 1.3 million unemployed.  

Added to that, the population in their prime working years, 25 to 54, who are currently employed is close to an all-time high. In early 2000 it peaked at 81.9%; it’s fluctuated since then, hitting a low in 2020, but today it’s back to 80.7%.

Simply stated, people who want to work have jobs. Enticing others to join the workforce would require aggressive recruiting. Furthermore, the population data for those 24 and younger – future workers – show a steady decline relative to the total (except for 9/11 babies). That will further exacerbate the recruiting problem.

The decline in the youth population is reflected in today’s college enrollments. There are 1.5 million fewer college students than in spring 2019. Enrollments are down in a long list of majors, from healthcare to engineering to public administration. That’s prompted a growing number of employers to drop degree requirements. Information/technology majors are one of the few that are increasing.  

The GS System is a Barrier to Change

It’s often forgotten but the General Schedule was actually created in 1923. The Classification Act of 1949 locked in consistent salary administration, simplified the classification system, and established equal pay for substantially equal work – but that’s the problem, salary administration is controlled by statute.  

In the late 1940s workforce management was very different. Workers started with an employer and stayed with the same organization until they retired. There were no salary surveys; the planners of pay systems focused narrowly on the internal ranking of jobs. Only the worst performers were denied increases.

Now surveys are conducted in virtually every industry. Employer associations, consulting firms, professional groups and even a few journals conduct surveys. All report job-specific competitive pay levels for commonly defined “benchmark” jobs (e.g., entry level civil engineers). Limited pay data are available to job seekers on websites like Indeed and Glassdoor, making it easy to find better paying opportunities.

The Federal Salary Council report from February shows market pay levels are significantly higher than GS salaries, ranging from 109 percent in San Jose-San Francisco to 37% for Rest of the U.S.  On average, GS salaries are 59% below market pay levels – and compared with prior years it’s getting worse. 

The reports combine pay data for all GS levels and all job series into a single value for each locality area. The gaps for high demand, high pay occupations are no doubt worse. For reasons that are not clear, current BLS surveys do not report market pay levels. They did when the Federal Employee Pay Comparability Act was enacted in 1990 but BLS changed their survey methodology a few years later. 

That’s very different from the use of survey data in other sectors. Pay for performance is another difference. It’s effectively universal for white collar employees in other sectors. Private sector pay systems have more flexibility to respond to emerging competitive practices. Newer issues like DEI/pay equity and “hot [higher order thinking] skills” are beginning to influence pay practices. The changes in jobs triggered by AI is an added factor. The new work environment forced employers to begin rethinking their practices.

Leaders in the National Academy of Public Administration concluded years ago that government also needs to rethink the GS program. Less than a year after passage of FEPCA, a NAPA team produced a report recommending the GS system be replaced. NAPA has released several subsequent reports advocating its replacement. In a 2021 column, NAPA President Terry Gerton and Janet Hale wrote: 

“[Government] is limited by rigid and outdated hiring, pay and performance policies and practices. Public managers and employees also struggle to adapt to the rapidly changing nature of work. . . . we face a significant risk that soon many public agencies and organizations will not have the workforce capabilities necessary to achieve their critical missions and provide services to the public.”

That risk has proven to be true.  

There Are Proven Alternatives 

The program changes started only a few years after the Classification Act was passed. In 1955 incentive awards were added to recognize outstanding performers. A few years later quality step increases were introduced to reward the better performers. In 1978 the Civil Service Reform Act created the Senior Executive Service and extended the use of performance incentives and ‘merit pay’ for managers.

Merit pay for managers eventually failed but that was because of a basic mistake. In hindsight, the mistake could have been avoided – managers eligible for merit increases earned the same as those continuing with step increases, thus minimizing any incentive to accept the new policy. It was too early for merit pay.

The more important change in light of current staffing problems was the authorization of Demonstration Projects, starting with China Lake in 1980. It gave agencies the chance to test ideas to address workforce problems; a common thread was the switch to pay for performance – and the change gained acceptance. Agencies also tested the idea of pay banding, a pay model where employees are rewarded for their skills. 

A related development was the legislation in 1989 that allowed the FIRREA agencies, most notably FDIC, OCC, NCUA and SEC, to develop their own pay systems. Congress recognized that the GS system “could impede these agencies’ ability to recruit and retain employees critical to meeting their organizational missions” [from a GAO report]. A few years later FAA, GAO and IRS also developed new pay systems.  One of the more recent is the intelligence community.  

Three agencies known to have workforce problems that could benefit from new pay systems are the National Park Service, the Bureau of Prisons and FEMA. Each have experienced a serious drop in employment and have staff in major cities as well as isolated rural areas. None are “white collar.”

The VA developed a slightly different model. Under Title 38 the primary medical specialists (physicians, nurses, etc.) are paid under separate schedules based on local surveys.

Developing separate pay programs for select occupations could help address specific workforce problems. Those working in border patrol or at our ports are possibilities. Workers in the new Wildland Firefighter series are another possibility. The goal should be to attract needed workers, not comply with decades ago regulations.

The alternative program models are proven and well documented. Agencies understand their staffing problems, they also understand the costs and risks of not addressing the problems. Experience with alternative pay systems over the 40+ years since China Lake proves the changes can gain acceptance and contribute to better workforce management. But there is still strong resistance to new systems even when the need is readily apparent. A case in point is the years lost trying to develop a separate pay system for government’s almost 100,000 technology specialists.  

OPM’s FY 2022-26 Strategic Plan does not mention the General Schedule. A century after it was created, it’s time for a rigorous review.