As U.S. federal government shutdowns have increased in frequency and length, their related dysfunction may impact the federal government's ability to retain talent. This fall, SPA Assistant Professor Keunyoung (Eli) Lee, along with coauthors William Resh (Georgia State University), Yongjin Ahn (Korea University), and Weijie Wang (Texas A&M University) published a piece in Public Administration Review examining the impact of 2018-2019 U.S. federal government shutdown on the U.S. federal civil service labor market.
Read the Full PDF

History of U.S. Government Shutdowns
Over the last four decades, United States federal government shutdowns, in which the federal government curbs agency activities, discontinues non-essential operations, and furloughs non-essential workers while maintaining essential functions, have become more numerous and acute. As of January 2025, the United States has experienced 21 government shutdowns, during the administrations of Presidents Carter, Reagan, George H.W. Bush, Clinton, Obama, and Trump. The most recent shutdown (as of this writing) occurred during President Trump's administration, extending for 35 days from December 2018 through January 2019.

Shutdowns may follow from disagreements on contentious issues, limited or expired authority for the Treasury to borrow, and the design of the Congressional Budget Act. They produce wide-ranging impacts across jurisdictions, demographics, and policies—the Congressional Budget Office estimates that the five-week partial shutdown between December 2018 and January 2019 delayed $18 billion in federal spending and lowered the real GDP of the first quarter of 2019 by $8 billion.

For example, public services suspended during the 2018–2019 shutdown negatively affected healthcare for the vulnerable, SNAP benefits for hundreds of thousands of people, commodity prices, the value of the U.S. dollar, and American stock markets, due to the uncertainty shutdowns impose on domestic and international investors.

Internally, shutdowns can impact employee morale, intrinsic motivations, types of employee exit, general policy implementation inputs, and the economic consumption behaviors of furloughed employees. Government employees react to exogenous shocks, including budget crises, changes in political environments, and administrative reforms. Political principals may reorganize agencies in reaction to a shutdown, via mass transfers of talent to other agencies or reductions in force (RIF). Employees may leave through their own discretion by transferring from one federal agency to another, retiring, or resigning. Each form of turnover impacts an agency differently, as a loss of existing capacity, and may be differently influenced by a government shutdown.

Types of Federal Employee Turnover
Research has linked federal employees' turnover intention and actual turnover with demographic factors such as age and job tenure; organizational factors such as employee empowerment and participation, flexible work schedule, and workplace culture and personal factors (e.g., satisfaction with job/pay/promotion opportunities). Little work has examined how the political environment, such as government shutdowns, affects federal employees' actual turnover.

Voluntary Versus Nonvoluntary Turnover
Type of employee turnover—voluntary vs. nonvoluntary—depends on whether the employee or the employer, respectively, initiates the decision to terminate the employment relationship. These two types of turnover are affected by different factors and are likely to have different consequences. Voluntary turnover is likely a result of employees' balancing of the desirability of the current job, or “push” factors, and the availability of alternatives, or “pull” factors. While nonvoluntary turnover could be beneficial, voluntary turnover often costs institutional knowledge and skilled workers, negatively affecting organizational performance.

In the context of federal employee turnover, the three major forms of voluntary turnover are retirement, quitting, and transfers to other agencies. Nonvoluntary turnovers include mass transfers to other agencies, reductions-in-force (RIFs), and terminations or removals.

Impacts of Government Shutdowns
During shutdowns, a large number of federal employees are furloughed or required to work without pay, temporarily but greatly reducing their income. Related uncertainties increase stress and lower morale, potentially causing them to look to the private sector or retire. Its effect, however, could vary across types of turnover.

A government shutdown signals political dysfunction, potentially acting as an external shock, “jar[ring] employees toward deliberate judgments about their jobs.” Shutdowns dampen the career outlook for career civil servants and may hasten exit by marketable mid-career or senior employees. Junior civil servants may have to reconsider their career prospects in the federal government, especially if their skills are in demand in the broader labor market.

A government shutdown could also produce some unintended financial consequences (e.g., increased savings) that pull civil servants to stay. It only temporarily reduced employees' income (they received back pay after the shutdown); and during the shutdown period, furloughed workers increased home production (e.g., cooking at home rather than dining out; opting out of child daycare services). In addition, government shutdowns can influence nonvoluntary turnover via RIF, termination, and mass transfer. Agencies might even update their contingency plans by transferring employees to exempted or excepted positions, redefining essential functions of the federal government.

Differential Impacts by Types of Employees
The 2013 shutdown led to an increase in voluntary separations among federal employees, as instability drove them to seek more stable employment. Shutdowns therefore act as catalysts for federal employees, especially those with skills transferable to the private sector, to pursue more stable opportunities outside the government. Further, the impact of the 2013 shutdown on voluntary separations varied significantly by age and tenure. Longer-tenured employees are more likely to remain with an organization due to accumulated benefits, loyalty, and a sense of obligation. Less tenured federal employees, more sensitive to career instability, may be less willing to endure shutdowns.

Contractual status should also influence the impact on voluntary separations, particularly among nonpermanent or contingent employees. They may face a greater degree of job precarity and, therefore, interpret shutdowns as signals of expendability, leading to higher voluntary separation rates. In turn, voluntary exit may be diminished among permanent employees who have stronger job protections and more entrenched incentives. Hierarchical level might also matter: lower-grade employees, who typically earn less, had higher separation rates following the 2013 shutdown than did higher-grade employees.

Nonvoluntary separations may also increase as a function of the shutdown. Agencies may increasingly rely on contracted private firms to fill gaps left by departing employees. This increased reliance on contract labor might lead agencies to streamline their workforce, potentially increasing nonvoluntary separations among nonpermanent employees.

The Study
Lee’s paper exploits the comparative basis provided by the intervention of 2018–2019 shutdown to understand impacts to the civil service labor market. Due to the timing of appropriations, some agencies shut down while others did not. The team leveraged personnel data from Office of Personnel Management (OPM) Enterprise Human Resource Integration (EHRI) (2005-2022) and the Employment Cube.

The team used two-way fixed effects difference-in-differences models of shutdown status and the percentage of workers furloughed in an agency to capture the “intensity” of that shutdown. They subset their samples by contractual status (permanent or nonpermanent), length of federal service, and hierarchical general schedule (GS) status. Analysis focused on overall separations, voluntary separations, and nonvoluntary separations, particularly quits (leaving federal service before retirement, perhaps the most troubling sign) and voluntary retirements.

Results
Lee and coauthors found that the 2018-2019 government shutdown influenced total and specific types of separations among federal employees, causing substantial disruptions, particularly through voluntary exits. Shutdowns significantly increased voluntary separations, particularly among less-tenured and lower-grade employees, while nonvoluntary measures disproportionately affected non-permanent staff.

Contrary to expectations, the analysis found no simple linear relationship between the percentage of furloughed employees and separation rates: the effects of furloughs on turnover may depend on factors beyond their scale, such as peer dynamics or perceived job insecurity. Agencies with higher furlough rates may create clearer expectations around roles, mitigating some negative effects on employee retention.

These results provide critical insights for policymakers and agency leaders aiming to mitigate the adverse effects of future shutdowns on workforce stability. Understanding the nuanced responses of different employee cohorts is essential for developing effective retention strategies and ensuring organizational resilience.

Key Shutdown Outcomes (see PDF version)

• Total separations in affected agencies rose by 0.532 percentage points, a 19% increase over mean total separations (2.81%).
• Voluntary separations increased by 0.285 percentage points, showing that employees in affected agencies were more likely to leave.
• Quit rates rose by 0.172 percentage points, a 17% increase over the mean (1.02%).

Among permanent employees:

• There were significant increases in overall (0.208 percentage points) and voluntary separations (0.184 percentage points): even employees with greater job security were impacted by the shutdown.
• Quit rates rose by 0.087 percentage points.
• Voluntary retirement was +0.056 percentage points, a 7.5% increase over the mean rate (0.75%), but the effect was not statistically significant.

Among non-permanent employees:

• Total separations increased by 0.354 percentage points over that in unaffected agencies, a 33% increase over the mean (1.08%).
• Nonvoluntary separations rose by 0.280 percentage points (p<0.1), a 51% increase over the mean (0.54%), suggesting that agencies may have initiated terminations.

Among employees with 10 or more years of service:

• Voluntary separations increased by 0.118 percentage points (p < 0.1) for all employees and by 0.129 percentage points (p < 0.05) for permanent employees, a 12.7% increase over the mean (1.01%).
• Nonvoluntary separations were not significantly affected: longer-tenured employees are more insulated from employer-initiated actions.
• These findings highlight the resilience of longer-tenured employees while pointing to increases in voluntary exits.

Among employees with less than 10 years of service:

• The shutdown’s effects were more pronounced, particularly among those with nonpermanent contracts.
• Quit rates rose by 0.121 percentage points, a 22% increase over the mean (0.54%), underscoring the sensitivity of less-tenured employees to workplace instability.
• Nonvoluntary separations increased by 0.242 percentage points (p<0.1), a 48% rise over the mean (0.503%).

Among higher-grade employees (GS-13 and above):

• Voluntary separations increased by 0.107 percentage points (p<0.1), a 16% increase over the mean (0.67%): even top-level staff are not immune to shutdown-induced stress.
• Quits increased by 0.073 percentage points, a 48.3% increase over the mean (0.151%).
• Nonvoluntary separations showed no significant changes, likely reflecting the higher protections and specialized roles associated with these positions.

Among lower-grade employees (GS-12 and below):

• Effects were more pronounced.
• Voluntary separations increased by 0.075 percentage points (p<0.1), an 8.2% increase over the mean (0.91%).
• Nonvoluntary separations, while positive, were not statistically significant.

Conclusion
Government shutdowns, increasingly a hallmark of U.S. politics, significantly affect the economy and the consistent delivery of public services. Lee’s paper helps explain these disruptions to the operational capacity of federal agencies and the civil service labor market, challenging some preconceived notions and offering new insights.

Federal government shutdowns impact the extent of voluntary and non-voluntary separations across different employee cohorts, causing critical vulnerabilities in government operations. Understanding these differential impacts will help public sector managers implement proactive retention measures, especially in anticipation of shutdown periods. Integrating political economy concepts into strategic human resource planning will help blunt these impacts.

Prospective employees may view shutdowns as indicative of broader instability, deterring talent from entering public service. Shutdowns also test intrinsic motivation among current employees, diminishing engagement, morale, and agency innovation and productivity, and increasing reliance on high-cost contractors to bridge staffing gaps.

Recommendations
Strategic workforce planning is essential to mitigate these challenges. Agencies must prioritize retention strategies for at-risk groups, and emphasize the meaningful public service mission of federal employment. Policymakers must also explore ways to insulate agencies from the disruptive effects of political gridlock.

Limitations
The team’s use of DiD models offers robust insights but also comes with limitations. First, the assumption of parallel trends between affected and unaffected agencies may not fully capture unobserved variables, such as agency-specific policies or external economic shifts. Second, the quantitative approach, while illuminating patterns in separations, cannot capture the complex motivations behind employees' decisions to leave. Third, their analysis primarily captures immediate and short-term effects of shutdowns; longer-term impacts, such as delayed retirements or subsequent waves of resignations, warrant further investigation.

Future Research
To address the above limitations, the paper suggests that future research incorporate qualitative perspectives, including interviews or ethnographic studies with affected employees. Such approaches could provide deeper insights into the mechanisms driving separations, such as morale, peer influence, or the perceived erosion of federal job stability. Additionally, longitudinal studies should explore lagged effects of shutdowns on turnover and agency performance to show long-term consequences.