Public comment: Improving performance, accountability and responsiveness in the civil service

Charles Ezell, Acting Director
Office of Personnel Management

Re: RIN 3206-AO80, Improving Performance, Accountability and Responsiveness in the Civil Service

Dear Acting Director Ezell,

On behalf of the Niskanen Center, I am pleased to share comments on the Office of Personnel Management’s (OPM) proposed rule regarding “Improving Performance, Accountability and Responsiveness of the Civil Service” posted on April 23, 2025. Enclosed with this letter I have also included a detailed explanation of several of our concerns and observations.

Underpinning our vision for the future is a belief that the federal government must have a workforce that is effective, accountable, and agile in order to deliver tangible impact to the public. For many years, we have sought to convince policymakers on both sides of the aisle that arbitrary downward pressure on federal headcounts,¹ overly restrictive rules on hiring² and firing,³ and the declining attractiveness of public service⁴ are key bottlenecks for other major policy priorities and merit significant reform.⁵ Indeed, we welcome new energy for these issues after neglect by leaders from both parties that allowed obvious flaws to linger uncorrected for decades after they were identified.⁶ The federal government has profound management challenges, and we are committed to working with all interested parties on responsible reform that empowers public servants to make thoughtful decisions and get things done without being bound by excessive bureaucracy.

The Niskanen Center is a nonprofit public policy organization that advocates for a government that provides social insurance and essential public goods, fosters market competition and innovation, invests in state capacity, and does not impede productive enterprise. We are committed to the principles of liberal democracy and an open society that encourages engagement, cooperation, discussion, and learning.

To that end, we agree with OPM that it is important to reform the performance management process in government to allow managers to create a culture of high performance on their teams. We also agree with President Trump’s nominee for Director of OPM when he said at his confirmation hearing that he would like to build a performance management system that accurately distinguishes between poor and high performers.⁷ As my colleague Jen Pahlka wrote in late 2024:

According to a Merit Systems Protection Board study, “three-quarters of supervisors of unacceptable performers reported attempting ten or more different approaches for addressing the performance problem of their most recent poor performer.” The same study concludes that even the “most effective” methods of resolving unacceptable performance available to federal managers are effective in “less than half of cases.” As with hiring challenges, this state of affairs affects productivity, retention, and motivation, as managers become consumed with HR processes while their low performers drag down the teams and their high performers become disillusioned.⁸

However, despite OPM’s professed desire to address this need in the preamble of this proposal, that is
not what this proposed regulation would accomplish if finalized and implemented. Accordingly, we must object to this proposed regulation.

The point of a system is what it does–or, as the 1826 Select Committee on reducing political patronage in the Executive branch put it“Names are nothing. The nature of a thing is in its substance.”⁹ In proposing this regulation, OPM is forgoing an opportunity to address the long-standing,¹⁰ well evidenced, and widely recognized¹¹ need to improve and streamline the performance management process¹² in favor of simply enabling arbitrary dismissals. This proposed regulation does nothing to improve differentiation between poor and high performers but instead seeks to create a fast-track process for the President to arbitrarily dismiss any employee, as if that would solve the problem. It will not.

If finalized, OPM’s proposal would make a series of relatively straightforward changes to 5 CFR: it creates a new schedule of excepted service employment, defines that employment as being limited to ‘career’ (which is to say, non-political) appointees performing ‘policy-influencing’ work based on the President’s discretion, and specifically excludes those employees from procedures granting excepted service employees the right to appeal their dismissal to the Merit Systems Protection Board (MSPB).

However, there are several unstated implications of these changes that significantly erode the current statutory protections for civil servants if it were to be finalized, directly contravening decades of statute:

• First, OPM does not further define ‘policy-influencing’ or related terms and leaves them so broad as to apply to nearly any federal employee. Current political appointees in Schedule C, for instance, perform a wide variety of ‘policy-influencing’ duties from clerical to management to strategic that mirror those performed by virtually all other employees. It is practically impossible to distinguish, then, from a federal employee that might be eligible for conversion and one that is not. OPM claims that it expects 50,000 positions to eventually be converted but there is nothing stopping this Administration or any future one from going much further. In fact, during the first Trump Administration, the same Director now in seat proposed moving nearly all of the Office of Management and Budget into a version of this schedule, including obviously non-policy roles like office managers, IT support staff, and administrative assistants.¹³
• Second, because the President has broad authority to determine what constitute ‘policy-influencing’, this regulation envisions a government where he could reclassify nearly any employee at his sole discretion–a move that would likely be unreviewable by any court or independent third party.¹⁴
• Third, following such a reclassification, the removal of administrative MSPB appeal rights means that any subsequent adverse action would also not be appealable to any court or other independent body based on the Supreme Court’s holding in United States v. Fausto.¹⁵ In general, employees need to start their appeal process for adverse actions at the MSPB and may elect to pursue their appeals further to the Federal Circuit Court of Appeals under 5 USC § 7703. However, because the Supreme Court held in Fausto that employees excluded from MSPB appeal rights could not seek further review in court, employees under the new Schedule Policy/Career would similarly lack the ability to appeal their adverse actions in any venue.

The opportunity this creates for abuse is obvious: if virtually everyone could be classified into Schedule Policy/Career, and if employees can’t appeal their reclassification into Schedule Policy/Career and then cannot appeal their subsequent dismissal, there is nothing stopping the President from ignoring the existing statutory removal process—and thereby Congress’ clear intent to permit appeals to the MSPB and to the Court of Appeals for the Federal Circuit—when he decides that it is inconvenient, where a
removal would violate the merit systems principles, or where the government could not produce evidence of poor performance. While we appreciate OPM’s stipulation that employees are not required to “personally or politically support the current President,”¹⁶ a rule without an enforcement mechanism is not much of a rule at all.

Historically, describing the need for presidential power to dismiss anyone as being about “performance” or “fitness” has long been a strategy employed to expand political control over the federal workforce for partisan ends. Thomas Jefferson, for example, described the need to drum up charges of “malversation or inherent disqualification” to fire Federalists¹⁷ in order to make room for Republican appointments and thus avoid popular outcry about partisanship in government.¹⁸ He significantly remade the civil service to favor Republicans using this mechanism.¹⁹ The 1820 Tenure in Office Act,²⁰ which limited many federal
jobs to four-year terms, was originally presented as a mechanism to remove poor performers without
embarrassing them.²¹ It formed the core statutory basis for the Jacksonian spoils system eight years later²² where “offices, instead of being considered as public trusts, to be conferred on the deserving, were regarded as the spoils of victory, to be bestowed as rewards for partisan services, without respect to merit.”²³ Richard Nixon’s team famously felt that it was untenable to keep referring to their management strategy as “‘politicizing the Executive Branch,’ and instead call[ed] it something like strengthening the Government responsiveness.”²⁴ That scheme ended in a disgraced president and clamour for reform. What Jefferson, Jackson, Nixon, and many other presidents in between had in common is that they sought to cloak their efforts to increase partisan control over the federal government in the language of performance management and accountability without actually building mechanisms that improved their organizations—just as OPM has sought to do here.

Congress, recognizing that such dismissals create serious risks to the stability and state capacity of the federal government, has acted over the last several decades to limit the President’s authority to dismiss career federal employees. Starting with the Lloyd-La Follette Act in 1912, through the Veterans Preference Act of 1944, the Civil Service Reform Act of 1978, and the sharply-titled Civil Service Due Process Amendments of 1990, Congress has progressively granted federal employees the right to appeal their dismissals to independent third parties and the courts, both on procedural and on substantive grounds, to prevent arbitrary dismissals for partisan ends. OPM cannot choose to ignore these by creating workarounds whenever they choose. Additionally, if this regulation were to go into effect, it would have a profound negative impact on the candor and intellectual integrity of the federal workforce and the policy processes they support. As the Grant Civil Service Commission recognized in their 1874 report to the President:

The practice of making appointments and removals in controlling reference to personal and partisan influence and spoils, directly and powerfully tended, in every grade of life, to discourage and overawe honest and manly thought and speech….the courage and fidelity that might, under a better system, have disclosed and removed great abuses were overawed and silenced.²⁵

The report goes on to recount how fearful civil servants were to go on the record about the goings on of their agencies for fear of reprisal, citing a Congressman who declared in 1868 that “[n]othing but the assurance of secrecy could procure us evidence of how the people were being plundered.”²⁶

These observations that were intuitive to reformers in the 19th century have been validated by researchers over a hundred years later. For instance, following a landmark study by Harvard Business School professor Amy Edmondson in 1999, the business literature has developed a robust evidence base showing that “team psychological safety” or the ability of employees to speak up without fear of retribution directly contributes to improved team and firm performance.²⁷ A recent meta-analysis of studies on the impact of fear on work performance found that “fear is negatively and significantly related to job performance” among other undesirable organizational and firm outcomes.²⁸ Evidence from across disciplines²⁹ suggests that decision-making processes improve³⁰ in quality when a variety of viewpoints are heard and considered,³¹ including political ones.³² Employee attitudes and engagement have also been long linked to organizational performance outcomes across organizations, industries, and locations.³³ This effect is also found in the federal government, as researchers have sought to connect employee engagement–measured by the Federal Employee Viewpoint Survey–with empowerment, performance, and other key ingredients to effective management.³⁴

The Trump Administration itself has acknowledged this effect in a way when attempting to impose requirements for more ideological diversity at America’s universities.³⁵ But, while it’s not clear that the Federal government can or should have any constitutional say in the intellectual life of academia,³⁶ it should seek to create a robust discourse inside its own agencies and among its own employees. Hanging the threat of arbitrary dismissal over the heads of federal employees—as this proposed regulation does—would have precisely the kind of chilling effect on internal discourse that healthy organizations must avoid. This regulation will degrade the quality of the policymaking process (and thereby the resulting policy and governing outcomes) by making career federal employees afraid to speak up when they have opinions that could be construed as running counter to the direction of a given Administration. Employees will know this and temper their views accordingly.³⁷ This effect is already playing out in real time at the Social Security Administration, for example, as it implemented a fraud detection scheme that resulted in significantly degraded service and failed to catch much fraud–a result that could have been foreseen had career staff felt comfortable speaking up. Instead, public reporting suggests that career staff stayed silent because “they were afraid to lose their jobs.”³⁸

This is the fundamental flaw with partisan influence over employment in the federal government: while a business that stifles internal disagreement may make poor choices and go out of business as a consequence,³⁹ no comparable feedback loop exists to disincentivize the imposition of groupthink or the breakdown in communications inside the federal government. In fact, in the case of some agencies, implementation of a poorly considered or underbaked decision can be lethal, as it was in the case of NASA’s flawed management of the Challenger mission.⁴⁰ More recently, the chaotic rollout of the Administration’s January 2025 freeze on all federal financial assistance illustrates the need for careful decision making: the Administration claims that it didn’t intend to shut down Medicaid payments to states but the initial memo instructed agencies to do so.⁴¹ Without robust internal discourse, these apparently inadvertent impacts weren’t surfaced until after the fact. Yuval Levin, who served in the Bush Administration, described how a well-functioning policy process would have prevented this recently:

“You would have had a meeting at the Office of Management and Budget where you bring in the chiefs of staff or senior political appointees from the various cabinet departments and lay out for them: What’s in this memo? What is it going to mean for you? Are we shutting down the Medicaid payment portal? And somebody at OMB. would have said: Well, no, this doesn’t affect payments to individuals. And the guy would have said: Well, those payments actually go to hospitals. Am I supposed to shut them down? It would have been a conversation. And they would have said: No, we’re not touching Medicaid.”⁴²

This description is accurate and would be familiar to anyone who has worked at OMB in other presidential administrations of both parties. Reclassifying wide swaths of the federal workforce to employment at the pleasure of a mercurial President—including and especially OMB⁴³—would make this type of improvisational policymaking process the norm rather than an embarrassing exception.

Evidence from around the world⁴⁴ and at home⁴⁵ points towards merit principles and nonpartisan program management as ingredients to effective government.⁴⁶ This is the reason that Congress originally took action to professionalize and protect the civil service in the 19th century and, as the government grew more complex, to strengthen those protections throughout the 20th century. We should look back at their experience with a government of “brawling politicians, broken-down hacks, and imbecile persons”⁴⁷ as a cautionary tale, not as an ideal to be emulated.

There is a better way. If the Administration is serious about addressing challenges in the federal government, there are steps it could take in collaboration with Congress to reform the current process and streamline removals for poor performance or conduct. For example, in the case of employee misconduct, it could articulate a set of standards for all federal employees that draws inspiration from the “10 Deadly Sins” that Congress established for employees of the Internal Revenue Service in 1998,⁴⁸ mandating termination of employees found to have violated them. The process for appealing such terminations could then be expedited and involve only a finding of fact by an independent third party. Congress could also streamline the process for internal agency appeals, unifying the process for both bargaining unit and non-bargaining unit employees and providing strict procedural limits on the length of time it takes to reach a final decision. It could unify the appeals process between the MSPB and the Equal Employment Opportunity Commission into one single adjudicative body specific to the federal workforce, eliminating the possibility of duplicative appeals. It could vest the authority for dismissals in the senior career civil servants at each agency that are tasked with tending to the organizational health of their long-term homes. These are reforms that could be debated, in the open, in Congress.

With regard to performance management, downward pressure on the budgets of agency human resources offices and lack of meaningful financial performance incentives are issues the Administration and Congress could positively impact. Appropriators could invest in the intellectual capacity of agencies to create and experiment with performance appraisal systems that actually differentiate low-performers from high-performers, learning from the private sector’s considerable experience in creating the same. It could adopt some of the pay-for-performance proposals that prior administrations⁴⁹ had advocated for, allowing managers to incentivize their best employees financially to sustain performance over time. OPM could also reverse the closure of programs like the Federal Executive Institute⁵⁰ that were designed to give managers the tools to expertly direct work, coach employees, and get high performance out of their teams. Private companies know this approach has significant return on investment and accordingly spend astronomical sums of money training their managers⁵¹ while the federal workforce is forced to go without. Management is a muscle, and we need to stop starving it of calories if we truly want a high-performing government.

The American people deserve a dynamic government, a robust policymaking process, and a non-partisan
civil service. Implementation of this regulation, if finalized, would run counter to those ideals. We
respectfully suggest that OPM retract this rule and begin working on a bipartisan basis with members of
Congress to come up with a more thoughtful, lasting solution to the performance management process
that improves state capacity rather than degrades it. We would welcome the chance to work with OPM
and Congress on such a project.

Sincerely,
Gabe Menchaca
Senior Policy Analyst, State Capacity Initiative
Niskanen Center

OPM’s Proposed Regulation Creates a Fast-Track Dismissal Process with No Real Limits

At its core, OPM’s proposed regulation makes a series of relatively straightforward changes to 5 CFR: it creates a new schedule of excepted service employment, defines that employment as being limited to ‘career’ (which is to say, non-political) appointees performing ‘policy-influencing’ work based on the President’s discretion, and specifically excludes those employees from procedures granting excepted service employees the right to appeal their dismissal to the Merit Systems Protection Board (MSPB). Notably, without the right to appeal to the MSPB, it is also likely that such employees would not be able to appeal their dismissal to a federal court except in cases where their dismissal was potentially illegal under other statutes (e.g., unlawful discrimination would remain appealable to the Equal Employment Opportunity Commission).

This ability to seek judicial review in the Federal Circuit was established by Congress as a core reform of the 1978 Civil Service Reform Act (CSRA).⁵² Prior to the CSRA, employees challenged their removals or the actions of the Civil Service Commission in a variety of venues and the process was thought to be too confusing and involve too many potential players. Initially, under the CSRA, employees in the competitive service (but not the excepted service) had the right to appeal to the MSPB and then to the Court of Appeals for the Federal Circuit. At the beginning, this right was not granted to employees in the excepted service. In 1989, the Supreme Court held in United States v. Fausto⁵³ that because Congress had deliberately made this exclusion, employees in the excepted service could not appeal their dismissal to the MSPB or seek judicial review. While this was remedied in 1992 with an expansion of appeal rights to nearly all excepted service employees, Congressional exclusions still remain for “confidential, policy-determining, policy-making or policy-advocating character.”⁵⁴

As a consequence of these changes, OPM estimates that ~50,000 employees may eventually occupy this schedule⁵⁵ but the text of the regulation could easily be construed to touch nearly any employee. For example, Schedule C appointments (i.e., those that are traditionally defined as having a “confidential or policy determining character”) that are typically held by low-level political appointees includes an extremely wide variety of duties that are held by a large number of career federal employees as evidence from a cursory flip through the Plum Book:⁵⁶ scheduling, preparing travel, writing communications, drafting documents, budget formulation, directing law enforcement activities, personnel management, etc. Indeed, because the President has broad authority⁵⁷ to create exceptions in the competitive service, it seems likely that the only limit on which employees are covered by “Schedule Policy/Career” is the President’s determination that they qualify. It’s not clear whether an employee has the ability to appeal such a reclassification or whether they retain their prior appeal rights and it will likely be for the courts to determine as OPM presses forward with implementation.⁵⁸

This possibility is not based on simple speculation. During the first Trump Administration, the Office of Management and Budget was one of the only agencies to complete its review of positions for reclassification into Schedule F, the precursor to Schedule Policy/Career and the results of that analysis became public in 2024 as a result of a Freedom of Information Act (FOIA) request. At the time, OMB–led by the same Director as it is today–proposed reclassifying not just OMB’s directly policy-related staff but also mission support roles like administrative assistants, human capital staff, records management professionals, and office managers.⁵⁹ There is no reason for observers to believe that implementation of a nearly identical policy in Schedule Policy/Career would look significantly different five years later. The confluence of these factors makes the opportunity for abuse obvious: in an effort to remove a given employee or group of employees for any reason (to include partisan ones), the President could simply elect to reclassify them into “Schedule Policy/Career” and then immediately terminate them. Since the employees do not have an established right to appeal their termination to a federal court (as a consequence of not having the right to appeal to the MSPB) there is no independent check on this power and it could, in theory, be wielded against virtually any federal employee, at any time, for any reason.

Congress Clearly Meant to Permit Employee Appeals of Proposed Removals

It is implausible that Congress intended to grant the President the authority to completely avoid such procedures while at the same time establishing them in great detail over the last century.

Beginning with the Lloyd-La Follette Act in 1912,⁶⁰ Congress imposed certain procedural requirements on the removal of career federal employees, including some degree of due process: employees had to be notified in writing of the reasons for their removal and given the chance to respond before they could be removed. Initially, they explicitly did not require a formal hearing or trial for such appeals. Lloyd-La Follette also included the limitation that employees could only be removed from the permanent civil service “for such cause as will promote the efficiency of said service.”⁶¹ Congress then subsequently established the right of employees to appeal removals proposed under this statute in 1944, when the Veterans Preference Act⁶² (VPA) provided preference-eligible veterans with the right to appeal their dismissal to the Civil Service Commission (CSC). The VPA echoed the requirements in Lloyd-La Follette requiring advance written notice of a proposed removal and the opportunity to respond. Going further, however, the statute specifically authorized preference-eligible veterans (e.g., those who had been honorably discharged from active duty and had a “service-connected disability”) to appeal their proposed dismissal to the independent Civil Service Commission.

President Kennedy subsequently granted all employees the right to an appeal via internal agency process in Executive Order in 1962⁶³ and President Nixon explicitly granted all employees in the competitive service precisely the same Civil Service Commission appeal rights as preference-eligible veterans via Executive Order in 1969.⁶⁴ These grants of authority by the Executive branch were subsequently codified in the Civil Service Reform Act of 1978, which split the Civil Service Commission into the MSPB, OPM, and Federal Labor Relations Authority, vesting the legacy-CSC responsibility to hear employee appeals at the MSPB for career employees in the competitive service.⁶⁵ Congress subsequently expanded this appeal process to cover most excepted service employees in 1990,⁶⁶ responding to the Supreme Court’s ruling a year earlier in United States v. Fausto⁶⁷ that the existing text of the CSRA did not explicitly entitle them to such review. At the time, the Committee explained its reasoning, writing that the court had “cut off an alternative method of judicial review for excepted service employees, saying that Congress, in passing the Civil Service Reform Act of 1978, had intended to deprive excepted service employees, other than those who were veterans preference eligible, of the right to challenge adverse actions. This bill explicitly provides those rights.”⁶⁸

In reversing OPM’s prior justification for last year’s rulemaking, OPM contends that “Congressional intent is determined by text of the law Congress passes.”⁶⁹ If this is the case, it seems clear that, as Congress has passed multiple laws over the last hundred years clearly requiring some degree of process and independent review of dismissals–first by the CSC and then by the MSPB and federal courts–that Congress did not intend to allow the President to establish a fast-track mechanism to dismiss career federal employees at will without any form of appeal. We do not need to rely on Committee reports to divine this intent, as Congress has clearly enshrined these appeal rights in law and subsequently expanded them. Implementation of the proposed regulation in their current form, then, contravenes this intent that has been established through successive legislative actions by allowing the Administration to ignore them at will.

If the Administration wishes to make a portion or the entirety of the federal workforce “at will” the appropriate avenue for making that change is legislation. Congress, of course, is empowered to make changes to the present statutory protections for federal employees and OPM is free to propose changes to whatever end the President prefers–but the Executive branch cannot unilaterally ignore Congressional intent by declaration or regulation.

OPM’s Proposed Regulation Will Have a Chilling Effect on the Federal Workforce & Policymaking

Federal agencies’ many Congressionally-authorized missions and mandates are operationalized through policy documents. Rules, regulations, solicitations, notices of funding opportunity, guidance documents, etc. are how agencies turn the Congress & the President’s direction into impact for the general public. Most of these policy documents are the result of internal agency deliberation, productive engagement between career and political staff, and (very often) some kind of interagency policymaking and review process. These processes are designed to both ensure consistency across the executive branch and to probe weak points in policy before it is issued. For example, the Office of Management and Budget coordinates interagency review on all legislative materials under Circular A-19⁷⁰ to ensure that the executive branch speaks with one voice to Congress and that all relevant viewpoints are considered on a given document before it is transmitted to Congress.⁷¹ This proposed regulation, for instance, is itself an output of some kind of policy making process at OPM and (perhaps) the interagency human capital policy community. OMB’s Office of Information and Regulatory Affairs (OIRA) would have weighed in somewhere in the process, as would the Office of Performance and Personnel Management. It may have undergone some degree of revision as a result of feedback received from reviewers before OPM issued it.

Historically, this intra and interagency policy process is thought to improve the quality of governance: by having a wide range of experts provide critical feedback on proposed policy matters prior to implementation, such a process helps ensure that there are not unknown, unintended consequences or at least that political leadership is aware of the potential downside risks or tradeoffs prior to issuance. Such processes are long-standing and widespread: OMB and its predecessor, the Bureau of the Budget, have coordinated some version of a review process for legislative materials (proposals, testimony, reports, technical drafting assistance, etc.) since the Administration of Franklin Delano Roosevelt nearly 90 years ago to avoid disconnects between agencies and the President and among agencies.⁷² The National Security Council has also, for instance, historically tried to maintain a robust policy process to present the President with fully vetted options for military or foreign policy action designed to advance a given Administration’s policy goals after some degree of internal and interagency discourse.⁷³ In these processes, candor and analytical rigour are critically important: reviewers must have both a diversity of viewpoints and the freedom to express those viewpoints candidly without fear of retribution or retaliation. It’s telling, for example, that some critics of the national security decision-making process charge that the system focuses too much on consensus and thereby encourages groupthink rather than allowing a variety of viewpoints to be presented to the President for eventual decision making.⁷⁴

OPM’s proposed regulation seems tailor-made to stifle this kind of discussion by hanging the threat of dismissal over the heads of employees–especially those participating in high-level policymaking discussions–by making their employment entirely contingent on the President’s will. It’s easy to imagine, for example, a particularly mercurial executive immediately dismissing an expert that expresses a dissenting opinion for being “insufficiently loyal.” Employees will simply temper their professional opinions on contentious topics for fear of triggering this type of reprisal or experts with strongly held views will never sign up to be public servants in the first place. Documents will reach the President without the benefit of strong steelmanning by career experts who have institutional experience and could otherwise point out flaws that might not be obvious to short-term political appointees. The American people will suffer from the implementation of hastily-considered or poorly-thought-through policies.

Research Shows the Employees Perform Better When They’re Not Afraid of Being Fired

This view is supported by the academic and business literature that have long found that employee voice is a critical input to good decision making and organizational performance.⁷⁵ In 1999, for example, Harvard Business School’s Amy Edmondson released a foundational study on “team psychological safety” that found a clear link between “a shared belief held by members of a team that the team is safe for interpersonal risk taking” and the team’s ability to learn and perform.⁷⁶ Revisiting this concept in a 2023 meta-analysis, Edmondson and Derrick Brandsby found that these concepts continued to appear consistently throughout academic research on organizational psychology and management:

Most of this work implicitly or explicitly recognizes the collaborative nature of work as a factor in motivating the study of psychological safety. As was true a decade ago, work today is profoundly interdependent, and effective collaboration is necessary for achieving results. Furthermore, collaboration depends on the timely and candid sharing of information, opinions, questions, and concerns. Increasingly, diverse expertise across disciplinary boundaries must be combined effectively to accomplish organizational goals. Spanning expertise boundaries is challenging when people are reluctant to speak up, due to the inherent risk of being seen as ignorant, incompetent, or intrusive. This is true for product design, patient care, strategy formulation, scientific research, and rescue operations; in each of these contexts, moreover, organizational behavior research has identified psychological safety as a factor in predicting success (Edmondson 1999, 2019).⁷⁷

These findings have been replicated across other industries, locations, teams. For example, a 2019 review of the non-medical literature on the impact of diverse team composition, for example, found that the literature generally showed a positive relationship between presence of a diversity of viewpoints on teams and financial, operational, and innovation performance in organizations from 1999 onward.⁷⁸ A 2013 study of corporate boards found a similar effect when it looked just at the political diversity of directors, “suggest[ing] that differences in viewpoints among corporate board members are an important mechanism that improves monitoring effectiveness leading to better firm performance.”⁷⁹

In general, these findings are also consistent with other studies that center on the role that employee engagement–that is the “involvement and enthusiasm of employees in their work and workplace”⁸⁰—plays in driving organizational performance. A 2002 meta-analysis, for example, found “that employee satisfaction and engagement are related to meaningful business outcomes at a magnitude that is important to many organizations and that these correlations generalize across companies.”⁸¹ This effect continues to be replicated year-after-year in surveys by firms like Gallup that show “[t]he relationship between engagement and performance at the business/work unit level is substantial and highly generalizable across organizations.”⁸² This effect has been replicated in the federal government with researchers correlating scores on OPM’s Federal Employee Viewpoint Survey (FEVS) with measures of organizational performance, employee empowerment, and other factors that have been shown to drive positive outcomes.⁸³

Research has also looked at the alternative negative framing of this phenomenon, the relationship between employee “fear” and performance outcomes, largely finding the same phenomenon at work. For example, a 2024 a meta-analysis by business school faculty at two Maryland universities⁸⁴ tested seven hypothesis on the topic across 54 other studies, including 1) that there “is a negative relationship between fear and job performance,” 2) that there “is a negative relationship between fear and [organizational citizenship behavior]” or the tendency of employee to go above and beyond their job duties to contribute positively to their organization, and 3) that there “is a positive relationship between fear and [counterproductive work behavior]” or the actions that harm or undermine organizations, among others. In analyzing each of these hypotheses, the researchers found statistically significant support for each–in other words, the literature in general showed that employees that are subjected to cultures of fear generally perform worse, fail to go the extra mile, and are more likely to undermine the organization they work for.

The Research Also Shows that A Non-Partisan Civil Service Produces Better Governing Outcomes

In addition to the general body of evidence discussed above and primarily derived from the business and management disciplines, political scientists have also investigated the role that partisanship plays specifically in generating better governing outcomes (e.g., program performance.) A recent analysis of 96 relevant studies from around the world found, for example, significant support for the hypothesis that adherence to non-partisan, meritocratic values improved program performance across many different domains of study.⁸⁵ This corroborates other comparative academic findings, for example, that highly professionalized bureaucracies–especially ones that maintain adherence to merit-based hiring practices–lead to lower levels of corruption across 52 countries (including not only the United States but many peer nations).⁸⁶

Even more specifically to the US federal government, research has also shown that program outcomes are improved through reduced partisan control over program management. In 2012, Vanderbilt Researchers Nick Gallo and David Lewis investigated over 1,000 federal programs to “compare the performance of federal programs administered by appointees from the campaign or party against programs run by other appointees or career professionals.” Using data from OMB’s much-maligned but innovative Program Assessment Rating Tool (PART) from the George W. Bush administration, Gallo and Lewis generally found that “programs run by political appointees perform worse on average than programs run by career managers.” While they hypothesize that this could be due to a number of factors–short tenure, for example, or a general lack of interest among political leaders in how they were rated–they do suggest that there is a degree of trading off that some Presidents do between management competence and political responsiveness. Notably, they also gesture at a potential reform that OPM should consider exploring in the future as it seeks to raise the bar on agency performance:

“Reducing appointees might help agencies find the right balance between appointees and careerists that will improve performance.”⁸⁷

Put simply, the research is clear: highly empowered, highly-engaged employees, with diverse opinions and backgrounds, not subjected to cultures of intentionally-created fear, contribute to better decision-making and better performance outcomes. In proposing this regulation, OPM is ignoring this well-established body of management and business literature in favor of a policy that seems purpose-designed to instill fear among federal employees.

1. Gabe Menchaca, “From Gore to DOGE: The Bipartisan History of Failed Workforce Reform,” Niskanen Center, April 24, 2025, https://www.niskanencenter.org/from-gore-to-doge-the-bipartisan-history-of-failed-workforce-reform. ↩︎
2. Cassandra Madison and Gabe Menchaca, “Beyond the Checkbox: Unlocking Skills-Based Hiring under the Chance to Compete Act,” Niskanen Center, April 23, 2025, https://www.niskanencenter.org/beyond-the-checkbox-unlocking-skills-based-hiring-under-the-cca. ↩︎
3. Jennifer Pahlka and Andrew Greenway, “The How We Need Now: A State Capacity Agenda for 2025,” Niskanen Center,
   December 20, 2024, https://www.niskanencenter.org/the-how-we-need-now-a-capacity-agenda-for-2025. ↩︎
4. Brink Lindsey, “State Capacity: What Is It, How We Lost It, and How to Get It Back,” Niskanen Center, November 18, 2021, https://www.niskanencenter.org/state-capacity-what-is-it-how-we-lost-it-and-how-to-get-it-back. ↩︎
5. Jennifer Pahlka, Recoding America: Why Government Is Failing in the Digital Age and How We Can Do Better (New York: Metropolitan Books, 2023). ↩︎
6. National Commission on the Public Service, Leadership for America: Rebuilding the Public Service (Washington, DC, 1989), https://www.google.com/books/edition/Leadership_for_America_Task_force_report/2MAZoUqVRz4C. ↩︎
7. U.S. Senate, Committee on Homeland Security and Governmental Affairs, Nomination of Scott Kupor to Be Director, Office of Personnel Management, 118th Cong., 1st sess., April 3, 2025, https://www.hsgac.senate.gov/hearings/nominations-9/. ↩︎
8. Pahlka and Greenway, “The How We Need Now,” December 20, 2024. ↩︎
9. U.S. Senate, Report of the Select Committee to Inquire into the Expediency of Reducing the Patronage of the Executive
   Government, 19th Cong., 1st sess., S. Doc. 19-22 (May 4, 1826), https://www.govinfo.gov/app/details/SERIALSET-00126_00_00-004-0022-0000/summary. ↩︎
10. U.S. Government Accountability Office, Federal Employee Redress: A System in Need of Reform, GAO/T-GGD-96-110 (Washington, DC, April 23, 1996), https://www.gao.gov/assets/t-ggd-96-110.pdf ↩︎
11. 11 U.S. Government Accountability Office, Issues Related to Poor Performers in the Federal Workplace, GAO-05-812R
    (Washington, DC, June 29, 2005), https://www.gao.gov/assets/gao-05-812r.pdf. ↩︎
12. U.S. Government Accountability Office, Improved Supervision and Better Use of Probationary Periods Are Needed to Address Substandard Employee Performance, GAO-15-191 (Washington, DC, February 2015), https://www.gao.gov/assets/gao-15-191.pdf ↩︎
13. Jory Heckman, “Schedule F Plans Show ‘Far Higher’ Impact on Federal Workforce than First Anticipated, NTEU Warns,” Federal News Network, February 27, 2024, https://federalnewsnetwork.com/workforce/2024/02/schedule-f-plans-show-far-higher-impact-on-federal-workforce-than-anticipated-nteu-warns/. ↩︎
14. In Roth v. Brownell, 215 F.2d 500 (D.C. Cir. 1954), the D.C. Circuit held that an attorney who had been removed by Executive Order from the “classified” service–the precursor to the “competitive service”–retained his prior competitive status for the purposes of removal procedures when he was terminated several years later. Given the temporal distance from this ruling and the significant revisions to civil service rules in the intervening 70 years, it seems likely that the courts will once again be called upon to sort out whether this remains the case. ↩︎
15. United States v. Fausto, 484 U.S. 439 (1989). ↩︎
16. Federal Register 90, no. 79 (April 23, 2025): 17,182. ↩︎
17. Thomas Jefferson to James Monroe, March 7, 1801, in The Papers of Thomas Jefferson, vol. 33, ed. Barbara B. Oberg (Princeton, NJ: Princeton University Press, 2006), 208–9, https://founders.archives.gov/documents/Jefferson/01-33-02-0166. ↩︎
18. Carl Russell Fish, The Civil Service and the Patronage (New York: Longmans, Green, 1905), 29-51. ↩︎
19. Ibid. ↩︎
20. Act of May 15, 1820, ch. 102, 3 Stat. 582 (“Tenure in Office Act”). ↩︎
21. Fish, Civil Service and the Patronage, 67. Quoting “Fragment of a manuscript letter from the Bureau of Rolls, Washington,
    loaned to the author by Dr. U. B. Phillips.” ↩︎
22. Congressional Research Service, “Removals in Jacksonian America through the Nineteenth Century,” in Constitution
    Annotated, https://constitution.congress.gov/browse/essay/artII-S2-C2-3-15-3/ALDE_00013109/#ALDF_00018934. ↩︎
23. U.S. Senate, Report of the Select Committee Appointed to Inquire into the Extent of the Executive Patronage, 23d Cong., 2d
    sess., S. Rep. 108 (February 9, 1835), 3, https://www.govinfo.gov/app/details/SERIALSET-00268_00_00-071-0108-0000. ↩︎
24. U.S. Senate, Select Committee on Presidential Campaign Activities, The Senate Watergate Report: The Final Report, 93d Cong., 2d sess. (Washington, DC: Government Printing Office, 1974), chap. 3. ↩︎
25. 25 U.S. Civil Service Commission, Report of the Civil Service Commission to the President (Washington, DC, April 15, 1874), 11-12, https://babel.hathitrust.org/cgi/pt?id=coo1.ark:/13960/t3ws9884v&seq=21 (accessed May 20, 2025). ↩︎
26. Ibid. ↩︎
27. Amy C. Edmondson and Derrick P. Bransby, “Psychological Safety Comes of Age: Observed Themes in an Established
    Literature,” Annual Review of Organizational Psychology and Organizational Behavior 10 (2023): 55-78,
    https://doi.org/10.1146/annurev-orgpsych-120920-055217. ↩︎
28. Sasha Pustovit, Chao Miao, and Shanshan Qian, “Fear and Work Performance: A Meta-Analysis and Future Research
    Directions,” Human Resource Management Review 34, no. 3 (2024): 101018, https://doi.org/10.1016/j.hrmr.2024.101018. ↩︎
29. L. E. Gomez and Patrick Bernet, “Diversity Improves Performance and Outcomes,” Journal of the National Medical Association 111, no. 4 (2019): 383-92, https://doi.org/10.1016/j.jnma.2019.01.006. ↩︎
30. Ethan R. Burris, “The Risks and Rewards of Speaking Up: Managerial Responses to Employee Voice,” Academy of Management Journal 55, no. 4 (2012): 851-75. ↩︎
31. Susan E. Jackson, Karen E. May, and Kristina Whitney, “Understanding the Dynamics of Diversity in Decision-Making Teams,” in Team Effectiveness and Decision Making in Organizations, ed. Richard A. Guzzo and Eduardo Salas (San Francisco: Jossey-Bass, 1995), 204-61. ↩︎
32. Incheol Kim, Christos Pantzalis, and Jung Chul Park, “Corporate Boards’ Political Ideology Diversity and Firm Performance,” Journal of Empirical Finance 21 (2013): 223-40, https://doi.org/10.1016/j.jempfin.2013.02.002. ↩︎
33. James K. Harter, Frank L. Schmidt, and Theodore L. Hayes, “Business-Unit-Level Relationship between Employee Satisfaction, Employee Engagement, and Business Outcomes: A Meta-Analysis,” Journal of Applied Psychology 87, no. 2 (2002): 268-79, https://doi.org/10.1037/0021-9010.87.2.268. ↩︎
34. Sergio Fernandez, James L. Perry, Meredith A. Newman, and Bryan J. Collins, “Assessing the Past and Promise of the Federal Employee Viewpoint Survey for Public Management Research: A Research Synthesis,” Public Administration Review 75, no. 3 (2015): 382-94, https://doi.org/10.1111/puar.12368. ↩︎
35. The White House, “Fact Sheet: President Donald J. Trump Reforms Accreditation to Strengthen Higher Education,” press release, April 23, 2025, https://www.whitehouse.gov/fact-sheets/2025/04/fact-sheet-president-donald-j-trump-reforms-accreditation-to-strengthen-hi gher-education (accessed May 20, 2025).. ↩︎
36. Natalie Schwartz, “Harvard University Won’t Yield to Trump Administration’s Demands,” Higher Ed Dive, April 14, 2025, https://www.highereddive.com/news/harvard-university-rejects-trump-demands/745331. ↩︎
37. In his classic piece on corporate communication, the Cognitive Style of Powerpoint, Edward Tufte discusses the degree to which “information architectures mimic the hierarchical structure of the bureaucracy producing those architectures.” He goes on to discuss the role that poor organizational communication–in particular, low resolution powerpoint slides–played at NASA in the 1986 Challenger Disaster. Edward R. Tufte, The Cognitive Style of PowerPoint: Pitching Out Corrupts Within (Cheshire, CT: Graphics Press, 2003). ↩︎
38. Hannah Natanson, Lisa Rein, and Meryl Kornfield, “How DOGE’s Grand Plan to Remake Social Security Is Backfiring,” Washington Post, May 16, 2025, https://www.washingtonpost.com/politics/2025/05/16/doge-social-security-musk-trump-cuts. ↩︎
39. Paul C. Nutt, Why Decisions Fail: Avoiding the Blunders and Traps That Lead to Debacles (San Francisco: Berrett-Koehler, 2002). ↩︎
40. E. Frank Harrison, “Challenger: The Anatomy of a Flawed Decision,” Technology in Society 15, no. 2 (1993): 161-83, https://doi.org/10.1016/0160-791X(93)90001-5. ↩︎
41. Rebecca Pifer, “Lawmakers Report States Shut Out of Medicaid Payment Portals Following Trump Funding Freeze,” Healthcare Dive, January 28, 2025, https://www.healthcaredive.com/news/medicaid-payment-shutdown-trump-funding-freeze/738532. ↩︎
42. Ezra Klein, “The Breaking of the Constitutional Order,” New York Times, February 5, 2025, https://www.nytimes.com/2025/02/05/opinion/ezra-klein-podcast-yuval-levin.html. ↩︎
43. Heckman, “Schedule F Plans Show ‘Far Higher’ Impact,” 2024. ↩︎
44. Carl Dahlström, Victor Lapuente, and Jan Teorell, “The Merit of Meritocratization: Politics, Bureaucracy, and the Institutional Deterrents of Corruption,” Political Research Quarterly 65, no. 3 (2012): 656-68, https://doi.org/10.1177/1065912911408109. ↩︎
45. Nick Gallo and David E. Lewis, “The Consequences of Presidential Patronage for Federal Agency Performance,” Journal of Public Administration Research and Theory 22, no. 2 (2012): 219-43, https://doi.org/10.1093/jopart/mur010. ↩︎
46. Eloy Oliveira, Carlos Pereira, Marcus André Melo, and Victor Lapuente, “What Does the Evidence Tell Us about Merit Principles and Government Performance?,” Public Administration 102, no. 2 (2024): 668-90, https://doi.org/10.1111/padm.12945. ↩︎
47. U.S. House of Representatives, Report of the Joint Select Committee on Retrenchment, 40th Cong., 2d sess., H.R. Rep. 47 (May 25, 1868), https://archive.org/details/civilservicerepo00unse/page/n11/mode/2up. ↩︎
48. Internal Revenue Service Restructuring and Reform Act of 1998, Pub. L. 105-206, 112 Stat. 685 (July 22, 1998). ↩︎
49. U.S. Office of Personnel Management, A Fresh Start for Federal Pay: The Case for Modernization (Washington, DC, April 2002), https://books.google.com/books/about/A_White_Paper.html?id=W-fe4J4Z4Z0C. ↩︎
50. Anastasia Obis, “White House Moves to Eliminate Federal Executive Institute,” Federal News Network, February 11, 2025, https://federalnewsnetwork.com/management/2025/02/white-house-moves-to-eliminate-federal-executive-institute. ↩︎
51. Deloitte, “With $1.4 B Investment, Deloitte Looks to Reshape the Future of Learning and Development,” press release, December 14, 2022, https://www2.deloitte.com/us/en/pages/about-deloitte/articles/press-releases/with-1-4b-investment-deloitte-looks-to-reshap e-the-future-of-learning-and-development.html. ↩︎
52. Civil Service Reform Act of 1978, Pub. L. 95-454, 92 Stat. 1111 (October 13, 1978), codified at 5 U.S.C. § 7703. ↩︎
53. United States v. Fausto, 484 U.S. 439 (1989). ↩︎
54. 5 U.S.C. § 7511. ↩︎
55. Federal Register 90, no. 79 (April 23, 2025): 17,182. ↩︎
56. U.S. Office of Personnel Management and U.S. Senate, United States Government Policy and Supporting Positions (Plum Book), 2024 (Washington, DC: Government Publishing Office, November 12, 2024), https://www.govinfo.gov/app/details/GPO-PLUMBOOK-2024. ↩︎
57. 5 U.S.C. § 3302. ↩︎
58. In Roth v. Brownell, 215 F.2d 500 (D.C. Cir. 1954), the D.C. Circuit held that an attorney who had been removed by Executive Order from the “classified” service–the precursor to the “competitive service”–retained his prior competitive status for the purposes of removal procedures when he was terminated several years later. Given the temporal distance from this ruling and the significant revisions to civil service rules in the intervening 70 years, it seems likely that the courts will once again be called upon to sort out whether this remains the case. ↩︎
59. Jory Heckman, “Schedule F Plans Show ‘Far Higher’ Impact on Federal Workforce than First Anticipated, NTEU Warns,” Federal News Network, February 27, 2024, https://federalnewsnetwork.com/workforce/2024/02/schedule-f-plans-show-far-higher-impact-on-federal-workforce-than-antic ipated-nteu-warns/. ↩︎
60. Lloyd-La Follette Act, ch. 389, § 6, 37 Stat. 555 (August 24, 1912). ↩︎
61. Ibid. ↩︎
62. Veterans’ Preference Act of 1944, ch. 287, 58 Stat. 387 (June 27, 1944). ↩︎
63. Exec. Order No. 10,987, “Employee-Management Cooperation in the Federal Service,” 27 Fed. Reg. 550 (January 17, 1962). ↩︎
64. Exec. Order No. 11,491, “Labor-Management Relations in the Federal Service,” 34 Fed. Reg. 17,605 (October 29, 1969). ↩︎
65. Civil Service Reform Act of 1978. ↩︎
66. Civil Service Due Process Amendments of 1990, Pub. L. 101-376, 104 Stat. 461 (August 19, 1990). ↩︎
67. United States v. Fausto, 484 U.S. 439 (1989). ↩︎
68. U.S. House of Representatives, Civil Service Due Process Amendments, H.R. Rep. No. 101-328 (1990), reprinted in 1990 U.S.C.C.A.N. 695, 698. ↩︎
69. Federal Register 90, no. 79 (April 23, 2025): 17,182. ↩︎
70. U.S. Office of Management and Budget, Circular No. A-19 (Revised): “Legislative Coordination and Clearance” (Washington, DC: Executive Office of the President, September 20, 1979), https://whitehouse.gov/wp-content/uploads/2017/11/Circular-019.pdf. ↩︎
71. Richard L. Revesz, “The Evolution of Regulatory Review,” Administrative Law Review 77, no. 1 (Winter 2025): 131-225,
    https://administrativelawreview.org/wp-content/uploads/sites/2/2025/03/ALR77.1_Revesz.pdf. ↩︎
72. Richard E. Neustadt, “Presidency and Legislation: The Growth of Central Clearance,” The American Political Science Review 48, no. 3 (1954): 641–71, https://doi.org/10.2307/1951044. ↩︎
73. John W. Rollins, The National Security Council: Background and Issues for Congress, CRS Report R44828 (Washington, DC: Congressional Research Service, October 19, 2022), https://crsreports.congress.gov/product/pdf/R/R44828 ↩︎
74. Jim Thompson, “The National Security Council’s Decision-Making Process: When Consensus Becomes a Constraint,” FAS Blog (Federation of American Scientists), January 22, 2025, https://fas.org/publication/when-consensus-becomes-a-constraint/ ↩︎
75. Ethan R. Burris, “The Risks and Rewards of Speaking Up: Managerial Responses to Employee Voice,” Academy of Management Journal 55, no. 4 (August 2012): 851-75, https://doi.org/10.5465/amj.2010.0562. ↩︎
76. Amy C. Edmondson, “Psychological Safety and Learning Behavior in Work Teams,” Administrative Science Quarterly 44, no. 2 (June 1999): 350-83, https://doi.org/10.2307/2666999. ↩︎
77. Amy C. Edmondson and Derrick P. Bransby, “Psychological Safety Comes of Age: Observed Themes in an Established Literature,” Annual Review of Organizational Psychology and ↩︎
78. L. E. Gomez and Patrick Bernet, “Diversity Improves Performance and Outcomes,” Journal of the National Medical Association 111, no. 4 (August 2019): 383-92, https://doi.org/10.1016/j.jnma.2019.01.006. ↩︎
79. Incheol Kim, Christos Pantzalis, and Jung Chul Park, “Corporate Boards’ Political Ideology Diversity and Firm Performance,” Journal of Empirical Finance 21 (2013): 223-40, https://doi.org/10.1016/j.jempfin.2013.02.002. ↩︎
80. Gallup, “How to Improve Employee Engagement in the Workplace,” Gallup Workplace (blog), https://www.gallup.com/workplace/285674/improve-employee-engagement-workplace.asp. ↩︎
81. James K. Harter, Frank L. Schmidt, and Theodore L. Hayes, “Business-Unit-Level Relationship between Employee Satisfaction, Employee Engagement, and Business Outcomes: A Meta-Analysis,” Journal of Applied Psychology 87, no. 2 (April 2002): 268-79, https://doi.org/10.1037/0021-9010.87.2.268. ↩︎
82. James K. Harter et al., Gallup Q12® Meta-Analysis: The Relationship between Engagement at Work and Organizational Outcomes, 11th ed. (Washington, DC: Gallup, 2024), https://www.gallup.com/workplace/321725/gallup-q12-meta-analysis-report.aspx (accessed May 20, 2025). ↩︎
83. Sergio Fernandez, James L. Perry, Meredith A. Newman, and Bryan J. Collins, “Assessing the Past and Promise of the Federal Employee Viewpoint Survey for Public Management Research: A Research Synthesis,” Public Administration Review 75, no. 3 (May 2015): 382-94, https://doi.org/10.1111/puar.12368. ↩︎
84. Sasha Pustovit, Chao Miao, and Shanshan Qian, “Fear and Work Performance: A Meta-Analysis and Future Research Directions,” Human Resource Management Review 34, no. 3 (September 2024): Article 101018, https://doi.org/10.1016/j.hrmr.2024.101018. ↩︎
85. Eloy Oliveira, Carlos Pereira, Marcus André Melo, and Victor Lapuente, “What Does the Evidence Tell Us about Merit Principles and Government Performance?,” Public Administration 102, no. 2 (2024): 668-90, https://doi.org/10.1111/padm.12945. ↩︎
86. Carl Dahlström, Victor Lapuente, and Jan Teorell, “The Merit of Meritocratization: Politics, Bureaucracy, and the Institutional Deterrents of Corruption,” Political Research Quarterly 65, no. 3 (September 2012): 656-68, https://doi.org/10.1177/1065912911408109. ↩︎
87. Nick Gallo and David E. Lewis, “The Consequences of Presidential Patronage for Federal Agency Performance,” Journal of
    Public Administration Research and Theory 22, no. 2 (April 2012): 219-43, https://doi.org/10.1093/jopart/mur010. ↩︎