Is the U.S. Civil Service really broken? What the research says about bureaucratic efficiency and reform

The Trump administration is pursuing sweeping changes to the federal workforce, aiming to increase political control over civil servants and reduce the size of the bureaucracy. The new Department of Government Efficiency (DOGE), led by Tesla CEO Elon Musk and other private-sector executives, has been tasked with streamlining government operations, eliminating diversity, equity, and inclusion (DEI) initiatives, and cutting what the administration describes as bureaucratic waste. Central to President Trump’s effort is an executive order that would reclassify tens of thousands of “policy influencing” career civil servant positions and make them political appointees.

To understand these efforts, we spoke with political economist Guo Xu, an associate professor at UC Berkeley Haas who studies the structure and efficiency of government bureaucracies. His work has examined historical civil service reforms, the partisan composition of the U.S. workforce, and the effects of political misalignment on government performance. Xu discusses the purpose of civil service protections and what the evidence says about their effectiveness, the potential consequences of the Trump administration’s efforts, as well as the debate about the existence of a so-called deep state.

Why do we have a civil service?

Civil servants are the backbone of the state—they make the government “work.” There is broad consensus among economists that every economy needs basic public goods to thrive. Civil servants—government employees—are those who implement the provision of these essential public goods. 

How did the establishment of the Civil Service change the U.S. government and the services it provides?

The evidence shows that the establishment of a professional civil service made the U.S. government more effective in providing goods more conducive to economic growth. Prior to the emergence of a civil service, public good provision was spotty. Government workers also served at the whim of politicians. They were appointed not based on merit but political loyalties. Constant turnover was the norm as presidents appointed their own supporters. The establishment of a civil service that selected its workers based on an exam and protected them from undue political firing not only improved the quality of the public sector personnel, but also provided the continuity needed to run a government.  My work on the Pendleton Act in 1883—a hallmark civil service reform in the U.S.—has shown how the introduction of merit-based selection and civil service protections reduced personnel turnover and improved the quality of a key public infrastructure: the postal service.

President Trump has a new executive order title “Restoring Accountability To Policy-Influencing Positions Within the Federal Workforce” (formerly known as “Schedule F”). What would this do and how many people will it affect?

Broadly speaking, civil servants can be either Political Appointees—appointed and fired at the discretion of the president—or Career Civil Servants—positions where the appointment and removal is insulated from political influence. Today, career civil servants make up more than 99.5% of the 2 million or so workers in the civil service. Schedule F is a proposal to reclassify senior-level career civil servants into positions that more closely resemble political appointments. There is no definite number of how many individuals Schedule F—if implemented—will ultimately cover, but it is most likely to cover tens of thousands of positions with substantial influence. 

Based on your research, what are the potential implications of reclassifying certain civil service roles—reducing job protections and increasing political appointments?

The first-order implication is that civil service positions will be less secure. Without civil service protections, government workers can be more easily fired. The selection of these workers will also be more discretionary. My historical research on the first-time introduction of civil service protections suggests that bureaucratic turnover is likely to increase hand-in-glove with presidential turnover. It is likely that this increased turnover will be disruptive for the continuity of policies and government programs, whose project horizon often extends beyond a single presidency. Unstable jobs are also less attractive to the best and brightest, potentially resulting in a brain drain. 

Your 2023 study on political ideology in the federal workforce was the first detailed look at the partisan composition of the civil service. Briefly describe what you did and what you found about the prevalence of Democrats versus Republicans in the federal bureaucracy.

In this study, we linked personnel records of federal government workers to voter registration records to examine the partisan composition of the civil service. A key descriptive finding of the paper was that civil servants were predominantly Democratic. Democrats made up just over half of the federal workforce during the 1997-2019 period (compared with about 41% of the U.S. population), while registered Republicans dropped from 32% to 26% during the period. An increase in Independents made up the difference. We also find that Democrat civil servants tend to remain in the federal government for longer. 

Does this ideological composition tend to shift with political cycles?

We find that the ideological composition shifts only among the political appointees, with Presidents preferring to appoint civil servants from their own party. What’s more important, however, is that we do not detect any partisan political cycles when we look at the 99.5% of career civil servants. What this means is that civil service protections work, shielding career civil servants from political interference. 

Your research indicates that political misalignment between federal employees and a presidential administration can lead to decreased performance and slowdowns. What did you find? 

Civil service protections decrease turnover by, as mentioned earlier, insulating civil servants from political interference. At the same time, however, this “firewall” makes civil servants harder for politicians to control. Without the ability to replace career civil servants, there will always be some civil servants who are politically misaligned with the president—they might disagree with policies or they could simply be demoralized. When we looked at procurement officers, we found that officers who did not share the same party affiliation as the president tend to incur larger cost overruns and delays. When we looked at HR surveys covering federal government workers, we also found that politically misaligned respondents were less motivated and less likely to identify with the overall mission of the agency. In this sense, civil service protections are a double edged sword: civil service protections can increase government performance by providing stability and continuity; however, this also means that not all workers will always agree with the president in office. This potential for mission misalignment might make the civil service less effective. 

So how might the recent firings of federal workers and proposed changes affect the overall efficiency and morale within federal agencies? And how might this affect public services?

The immediate impacts of the recent measures to downsize the civil service are clearly negative. Civil servants are currently facing heightened uncertainty and disruption. The downsizing is not only demoralizing but has already adversely impacted the delivery of public services, as we have seen in numerous news reports. The longer-term impact of any change to civil service protections is difficult to assess. Removing civil service protection can give politicians the carrots and sticks to motivate government workers, but it also opens the door for undue political interference and, frankly, cronyism. There is still an open question of what the optimal degree of civil service protections is—should it be 99.5% of the workforce, or more, or less? How the proposed changes will affect government performance will hinge on many unknowns—such as how many positions and which positions will be converted to Schedule F.

To be clear, though, the debate over technicalities of Schedule F might miss the forest for the trees. What is happening more broadly right now is an attempt to systematically shrink and delegitimize the civil service. Schedule F is just a small part of this agenda.

How could the proposed changes affect public trust in government institutions and their ability to respond effectively to public needs?

The situation is evolving quickly, and it is hard to say how the proposed policy changes will play out. What is clear, however, is that the current efforts to reform the federal workforce have been accompanied by attempts to slander civil servants as lazy, unaccountable, rogue elements within the government. This is likely to erode public trust in government institutions across the political spectrum. Those who are receptive to the negative messaging will distrust the state even more. Those who traditionally viewed government workers as mission-driven, impartial civil servants will worry about the emergence of a partisan bureaucracy. 

One of the Trump administration’s stated goals is to dismantle the “deep state,” generally understood to mean a clandestine network of unelected government officials that has its own agenda outside of the elected administration. Have you found any evidence that such an organization exists?

I have certainly found no evidence for a conspiratorial “deep state” like you describe. What we did find is the presence of political misalignment frictions: civil servants who might not agree with their superior tend to do a worse job. That alone does not mean that there is an outright conspiracy, or that we should fire every individual who disagrees. Civil servants tend to be experienced experts in their subject matter, and it is very possible that the benefit of their expertise by far outweighs the cost of political misalignment.

How might the initiatives undertaken by unelected leaders of DOGE influence these dynamics?

The fact that the majority of civil servants are Democratic is often seen as evidence for a deep state. Attempts by DOGE to remove civil service protections and abandon DEI initiatives are, in turn, viewed as ways to drive out these rogue actors and achieve greater political alignment across the entire federal government. I am skeptical that these attempts will achieve their desired goal. Rather than reflecting nefarious attempts by Democrats to infiltrate the federal bureaucracy, the evidence I have seen is pointing to another story: Democrats are simply more pro-government and prefer to work for the state! Rather than driving out civil servants, DOGE would benefit from finding ways to make public service more attractive for Republicans. I doubt that demonizing civil servants is conducive to this. 

How do the current workforce restructuring efforts compare with previous administrative changes you’ve studied?

Civil service reforms tend to take time. The introduction of the Pendleton Act 1883 initially only covered around 10% of the federal workforce. It took two decades for the civil service protections to cover slightly less than half of the federal workforce. The pace of the current restructuring effort, in contrast, is much faster and more disruptive. The federal government, however, is much too important to just “move fast and break things.” 

Many are comparing parts of the current restructuring effort—the dismantling of DEI—to Woodrow Wilson’s mandate to segregate the federal government by race. This is an episode I have also studied. While a direct comparison is a stretch, there are interesting similarities in the justification of the policy. Wilson’s argument for segregation was that it would improve bureaucratic efficiency, removing “frictions” between white and black workers. We found that the segregation policy was, in fact, distortionary, leading highly skilled Black workers to be demoted to menial positions. This misallocation of talent increased the already-high earnings gap between Black and white workers by 20%, eroding much of the gains Black workers had made since Reconstruction. Current proposals to remove DEI seem to be likewise based on the premise of improving merit-based selection and performance—diversity is seen as an impediment to the functioning of the bureaucracy. Whether the proposed replacement of rule-based competitive hiring with discretionary appointments is more merit-based, however, is debatable.