Discretionary Spending

Multiple Budget Functions

Reduce the Size of the Federal Workforce Through Attrition

CBO periodically issues a compendium of policy options (called Options for Reducing the Deficit) covering a broad range of issues, as well as separate reports that include options for changing federal tax and spending policies in particular areas. This option appears in one of those publications. The options are derived from many sources and reflect a range of possibilities. For each option, CBO presents an estimate of its effects on the budget but makes no recommendations. Inclusion or exclusion of any particular option does not imply an endorsement or rejection by CBO.

Billions of Dollars 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2019-
Change in Spending  
  Budget authority 0 -0.6 -2.0 -3.4 -4.3 -4.7 -4.9 -5.0 -5.2 -5.3 -10.3 -35.4
  Outlays 0 -0.6 -1.9 -3.3 -4.3 -4.7 -4.8 -5.0 -5.2 -5.3 -10.2 -35.2

This option would take effect in October 2019.


In 2017, the federal government employed about 2.1 million civilian workers, excluding Postal Service employees. About 43 percent worked in the Department of Defense or Department of Homeland Security, and roughly 17 percent were employed by the Department of Veterans Affairs. The rest worked in agencies that provide a variety of public services—regulating businesses, investigating crimes, collecting taxes, and administering programs for the elderly, poor, and disabled, for example. The largest costs that the federal government incurs for its civilian employees are for salaries, future retirement benefits, and health insurance.


This option would prohibit selected federal agencies from hiring more than one employee for every two workers who leave, until the number of federal civilian employees at agencies the President allowed to be affected was reduced by 10 percent. Agencies would be limited in their ability to replace those employees with contractors or to increase compensation for new hires because their appropriations would be decreased accordingly. The President would be allowed to exclude an agency from the requirement to replace every two workers with one worker under certain conditions—because of a national security concern or an extraordinary emergency, for instance, or if the performance of a critical mission required doing so.

Effects on the Budget

This option would reduce the deficit by $35 billion from 2019 through 2028, the Congressional Budget Office estimates. CBO arrived at that figure by combining estimates of the reduction in hiring with estimates of the average cost of compensating a hire. About two-thirds of the federal civilian workforce would be exempt from the requirement, in CBO's estimation, leaving an affected workforce of about 700,000 and a total reduction in that workforce of about 70,000. Given recent rates of employee turnover, the government would reach that total by hiring about 21,000 fewer employees in each year through 2022 and about 6,000 fewer employees in 2023. By the end of 2020, CBO expects, the average cost of compensating an employee would be about $72,000 for his or her first full year of employment. Thus, if employees are hired at roughly the same rate throughout the year, the amount spent on them would be reduced by about $800 million in the first year after enactment of this or a similar option. The deficit would fall by a smaller amount—$600 million—because about one-fifth of employees are paid from fees their agency collects for providing certain services, such as customs fees and patent registration fees. CBO expects that decreasing the number of people providing those services would reduce those collections by an equal amount. By 2028, the reduction in the deficit would grow to $5.3 billion as the effects of reduced hiring on the size of the workforce accumulated.

A large source of uncertainty in this option's estimate of savings over the next 10 years is CBO's estimate of the portion of workers who would be exempt from this requirement. To determine that number, CBO examined data from the two most recent government shutdowns. On the basis of the number of employees who continued working during those shutdowns, CBO estimates that about two-thirds of the federal civilian workforce would be exempt from this requirement. However, it is unclear whether the President would respond to this option the way past Presidents responded to temporary shutdowns.

A second large source of uncertainty is the portion of nonexempt workers whose compensation is provided from agency collections. Depending on which agencies the President chose to exempt from this requirement, the fraction of the reduction in budgetary authority that represented a decrease in offsetting collections would vary. That is because the employees whose compensation is funded by such collections are unevenly spread across agencies.

Alternative approaches that set more or less stringent limits on the fraction of departing workers that agencies could replace might lead to the President's exempting more or fewer agencies from the limit. For example, the savings from a hiring freeze (in which case agencies are not allowed to hire any employees for every two who leave) would be less than twice the savings generated by this option if the freeze constrained the ability of more agencies to perform critical missions.

Other Effects

An argument for this option is that some agencies could continue to provide crucial services with a smaller workforce by operating more efficiently and by eliminating services that are not cost-effective. The number of management and supervisory positions has increased in many agencies as the workforce has aged, and research suggests that, in some cases, the additional layers of management hamper performance. This option could encourage agencies to reduce the number of managers and supervisors through attrition as people in those positions retire over the next few years. Research also suggests that federal workers in the roughly 40 percent of jobs that do not require a college degree earn more than their counterparts in the private sector. If private-sector compensation is indicative of the value of those positions, then the savings generated by trimming that part of the workforce would exceed the value of the services that those jobs produce.

An argument against this option is that trends in federal employment suggest that the federal workforce might already be under strain from previous cost-cutting measures and that further reductions could impair the government's ability to fulfill parts of its mission. The federal civilian workforce has grown little over the past 20 years, whereas both the number of people the government serves (as measured by the population of the United States) and federal spending per person have grown substantially. Moreover, the workforce at most agencies has shrunk, and the modest growth in the total number of federal civilian employees largely reflects hiring for the Department of Homeland Security (which was established in 2002) and the Department of Veterans Affairs (which increased the volume of services it provides to veterans). Workforce reductions at those or other agencies would probably reduce the quality and quantity of some of the services provided and could have other negative effects, such as increasing the amount of fraud and abuse in some government programs. Lastly, because this option would be phased in as workers left their positions, federal agencies would have little control over the timing of the workforce reductions.