AFPI Comment on Proposed Office of Personnel Management Regulation That Would Prevent the Reinstatement of Schedule F

James Sherk,  November 21, 2023

James Sherk
America First Policy Institute
1001 Pennsylvania Ave NW, Suite 510
Washington, D.C. 20004

November 15, 2023

Tim Curry
Deputy Associate Director
Accountability and Workforce Relations
U.S. Office of Personnel Management
1900 E Street, NW
Washington, DC 20415-1000

Re: RIN 3206-AO56, Comment of the America First Policy Institute in Opposition to the Proposed Rule Upholding Civil Service Protections and Merit System Principles, 88 Fed. Reg. 63862 (Sept. 18, 2023)

Dear Mr. Curry:

My name is James Sherk. I direct the Center for American Freedom at the America First Policy Institute (AFPI). AFPI is a 501(c)(3) non-profit, nonpartisan research institute that advances policies that put the American people first. Our guiding principles are liberty, free enterprise, national greatness, American military superiority, foreign policy engagement in the American interest, and the primacy of American workers, families, and communities in all we do.

America First Policy Institute (AFPI) submits the following comments in opposition to the Office of Personnel Management’s (OPM) proposed rule purporting to extend removal protections to career officials in positions of a confidential, policy-determining, policymaking, or policy-advocating character (hereafter “policy-influencing positions”). See Upholding Civil Service Protections and Merit System Principles, 88 Fed. Reg. 63862 (proposed Sept. 18, 2023), Docket ID OPM–2023–0013, RIN 3206–AO56.

The federal government’s hiring and removal processes suffer from widely recognized problems. Federal employees believe that agencies do not take appropriate actions to remove poor performers. Some career federal employees use their positions to advance their own policy preferences. The federal hiring process is seen as broken.

Civil service employment protections and hiring procedures cause these problems. They make it difficult to hire the best candidates and prohibitively difficult to dismiss employees for all but the worst offenses.

In 2020, President Trump addressed these problems with Executive Order 13957, “Creating Schedule F in the Excepted Service” (EO 13957). EO 13957 gave agencies greater flexibility in hiring, and eliminated appeals when dismissing, federal employees in senior policy-influencing positions. The order also retained protections against politically motivated or otherwise discriminatory dismissals. Had it remained in effect, EO 13957 would have given agencies greater discretion in hiring and firing senior career officials, while retaining their status as career employees selected based on merit. Trump Administration officials estimated the order would cover between two and three percent of the federal workforce. Similar policies have worked well in many states. However, President Biden rescinded EO 13957 shortly after taking office (Exec. Order 14003).

EO 13957 would have returned the federal civil service to its foundations. The Pendleton Act of 1883 created the modern civil service but deliberately did not give federal employees removal protections. The reformers who created the civil service feared that requiring “a virtual trial at law” to dismiss an employee would entrench incompetence and intransigence in the federal workforce. Not until the 1960s did the general federal workforce gain the ability to appeal dismissals. The experience of the past six decades has demonstrated the folly of that decision.

EO 13957 was grounded on firm legal authority. Title 5 specifically authorizes the President to exempt policy-influencing positions from civil service appeals. Statutory context makes clear this authority extends to both political appointees and career officials.

OPM now proposes regulations that criticize Schedule F and are intended to prevent a future administration from reinstating it. OPM’s criticisms are misplaced. EO 13957 recognized and protected the value of a career workforce that builds up institutional experience and expertise. If, as OPM contends, the order was intended to bring back the “spoils system” the President would not have prohibited patronage appointments. The order also reduced the incentive for political appointees to attempt to “burrow” into policy-influencing positions.

OPM’s proposed rule would instead make dismissing employees in senior policy-influencing positions for poor performance or intransigence considerably more difficult. This would “seal up” poor performers in the bureaucracy. It would also undermine the government’s democratic accountability by facilitating bureaucratic resistance.

OPM’s proposed rule would also discourage vetting prospective policies with career staff. The practical consequence of insulating career staff from accountability is political appointees cut them out of the loop to avoid leaks. If career officials feared leaking draft policies could end their careers, political appointees would have more freedom to seek their input.

OPM’s proposed rule is bad policy. OPM’s proposed revisions to part 752 are also patently unlawful. 5 U.S.C. § 7511(b) categorically exempts policy-influencing excepted service positions from Chapter 75’s adverse action procedures. OPM has no authority to extend civil service removal restrictions to employees in such positions. Nothing in title 5 says or implies civil service protections follow individual employees. OPM’s interpretation of chapter 75 would also violate Supreme Court precedent concerning permissible congressional restrictions on the removal of inferior officers.

OPM’s proposed rule would undermine the efficacy of the federal service. The Office should abandon this proposed rulemaking in its entirety.

Challenges Facing the Federal Workforce

The federal workforce faces longstanding and serious challenges. Agencies fail to address poor performers effectively. Misconduct—including policy resistance—occurs at unacceptably high levels. The federal hiring process is also widely recognized as broken. The federal workforce needs reform.

Agencies Do Not Address Poor Performers Effectively

The federal workforce has a widely recognized performance management problem. Merit System Principle 6 provides that “employees should be retained on the basis of the adequacy of their performance … employees should be separated who cannot or will not improve their performance to meet required standards” (5 U.S.C. § 2301(b)(6)). However, surveys show few federal employees believe their agencies do this.

Until 2022 the Federal Employee Viewpoint Survey (FEVS) asked employees if they believed that “in my work unit, steps are taken to deal with a poor performer who cannot or will not improve.” Over the past decade, agreement with this statement ranged from a low of 28 percent to a high of 42 percent (Office of Personnel Management, 2016, p. 39; Office of Personnel Management, 2021, p.15). Most federal employees have never agreed with this statement. As OPM is aware, this question was removed from the 2022 and subsequent FEVs at the request of federal unions, so data for 2022 is now unavailable.

Other surveys find similar results. The Merit Systems Protection Board’s (MSPB) Merit Principles Survey finds less than a quarter of federal employees believe their “organization addresses poor performers effectively” (Read, 2019, p. 3). A separate survey by the Government Business Council found that only 11 percent of federal employees say their agency fires poor performers who do not improve after counseling (Katz, n.d.).

Starting in 2019, a separate FEVS question asked employees how their work unit usually addresses poor performers. Only a fifth of respondents have reported that “there are no poor performers in my work unit.” Instead between 42 and 56 percent of employees report that poor performers usually “remain in the work unit and continue to underperform” (Office of Personnel Management, 2021, p.16; Office of Personnel Management, 2022a, p. 32).

Federal workforce data corroborates these survey results. Agencies removed just 3,900 of 1.6 million permanent tenured employees for performance and misconduct in FY 2022 (Office of Personnel Management, 2022b).[1] Agencies’ failure to remove poor performers makes them less effective. Business leaders report that poor performers demotivate and reduce the performance of productive employees (Hastings and Meyer, 2020, p. 8-10).

Misconduct and Policy Resistance Reduces Democratic Accountability

America was founded on the principle of government by consent of the governed. The people elect the President; the President appoints senior agency officials; those officials carry out the law with the assistance of their subordinates. The Constitution thereby gives the American people a role (albeit indirectly) in choosing the officials who govern them.

However, very few federal employees’ jobs depend on who wins the next presidential election. Only about 4,000 of the federal government’s 2.2 million employees are political appointees—less than 0.2 percent.[2] The President and his direct reports rely heavily on career staff. Career federal employees necessarily have significant discretion in implementing policy and enforcing the law—the principal ways the government exercises power over the American people.

For America’s government to be democratically accountable to the American people, career employees must impartially implement the law and the President’s policies. The civil service was predicated on the assumption that career employees would provide neutral expertise irrespective of their personal views. To their credit, many federal employees uphold this ideal.

Unfortunately, many do not. “As anyone who has ever held a senior position in the Executive Branch can attest, federal employees often regard themselves, not as subordinates duty-bound to carry out the President’s vision whether they personally agree with it or not, but as a free-standing interest group entitled to make demands on their superiors” (Feds for Medical Freedom v. Biden, 2023, at 391).

Evidence that a significant number of federal employees feel no obligation to provide neutral expertise is readily available. During the Trump Administration, bureaucratic resistance to President Trump’s policies became so widespread that it made national news. Within two weeks of Trump’s inauguration the Washington Post ran an article titled “Resistance from within: Federal workers push back against Trump.” The article documented career employee efforts to undermine the President’s agenda. It quoted one federal employee boasting how “[y]ou’re going to see the bureaucrats using time to their advantage” to block Trump policies (Eilperin, Rein, & Fisher, 2017). At the end of 2017, Bloomberg News ran a feature titled “Washington bureaucrats are quietly working to undermine Trump’s agenda” that similarly documented how “career staff have found ways to obstruct, slow down or simply ignore their new leader, the president” (Flavelle & Bain, 2017).

Trump Administration officials’ experiences corroborated these news reports. Senior administration officials reported many instances of career staff seeking to undermine policies they opposed. AFPI published a report documenting the types of bureaucratic resistance officials faced across the executive branch.[3] Examples of tactics career staff used to oppose Trump Administration policies include:

  • Withholding Information. Career employees in the Environmental Protection Agency (EPA) Office of General Counsel (OGC) routinely failed to keep political appointees informed about significant cases. OGC would have weekly staff meetings about agency litigation. EPA political appointees would subsequently double-check with Department of Justice (DOJ) lawyers and find out the career staff were not providing updates for critical cases. The career employees did not tell political appointees about significant cases EPA was involved in or the legal arguments EPA was making. Staff omissions were so frequent and significant that political appointees resorted to regularly checking PACER to see what was happening (Sherk, 2023, p. 8).

  • Refusing Ideologically Distasteful Work. In 2016, an Asian-American advocacy group asked the Department of Justice (DOJ) to investigate Yale and other Ivy League universities’ admissions practices. The advocates suspected racial discrimination against Asian-Americans in violation of the Civil Rights Act. The Educational Opportunities Section (EOS) within the DOJ Civil Rights Division exists to investigate such complaints. However, the complaint languished because EOS career staff did not support the case. Winning it would prohibit many race-based affirmative action programs.

    In 2018, political appointees directed the career deputy who ran the EOS to oversee the Yale investigation and move it along. The career deputy did so, despite philosophically disagreeing with the investigation. Two attorneys from outside the EOS were assigned to work on the case. The investigation took much longer than usual despite being a straightforward case. Finally, after two years of investigating, the DOJ uncovered strong evidence of racial discrimination against Caucasians and even stronger evidence of discrimination against Asian-Americans. Political appointees personally drafted and filed a racial discrimination complaint against Yale—a task that junior career lawyers would typically handle.

    DOJ then needed to assemble a team to pursue the Yale racial discrimination case. Political appointees asked EOS to provide eight lawyers to work on the case. The career staff refused outright, telling political appointees that none of them would work on it. DOJ instead had to assemble a team from outside EOS, primarily made up of employees borrowed from the DOJ Civil Division, the U.S. Attorney’s Office in Connecticut (where Yale is located), and the Civil Rights Division’s front office. DOJ had clear evidence of racial discrimination at Yale and a clear legal theory, but no EOS career lawyers would work on the case because it did not support their worldview (Sherk, 2023, pp. 9-10). The Supreme Court subsequently endorsed the theory underlying this lawsuit in Students for Fair Admissions v. Harvard (2023), holding that the Civil Rights Act prohibited Harvard University from discriminating against Asian-Americans based on their race.

  • Delays and Slow Walking. A Department of Labor enforcement agency has a subcomponent whose only job is to write regulatory and policy documents. The unit has about 10 to 15 career employees at any given time. In the fall of 2017, political appointees requested a status update on a draft proposed rule. The unit had been working on this rule since the start of the Trump Administration. It was a department priority and this unit’s primary responsibility during this period. Career staff reported the draft would not be complete until March 2018. Political appointees asked for the draft before the end of the year. Career staff said that pace was impossibly burdensome and would drive staff to quit. Political appointees subsequently calculated the staff’s proposed pace amounted to each career employee writing one line of text per day. The appointee estimated a competent private-sector lawyer could complete the draft in two to three weeks. Political appointees subsequently gave up on these career staff and wrote many policy documents themselves (Sherk, 2023, p. 11).

  • Unacceptable Work Product. All politically sensitive regulations at the Education Department had to be written by political appointees. Career employees assigned to produce drafts of these regulations would come back with “completely unusable” drafts that either diverged significantly from Department priorities or would never withstand judicial review. So political appointees had to do it themselves. For example, the Education Department’s Title IX rule (providing due process when students are accused of sexual misconduct) was drafted almost entirely by political appointees (Nondiscrimination on the Basis of Sex, 2020). Involving career employees in drafting the regulations served only to preview the arguments that opponents of the rule would eventually make in the courts and the public sphere once the rule was published. Political appointees at many other agencies reported similar experiences (Sherk, 2023, p. 12).

  • Leaking. The Interior Department wanted to revise Obama Administration regulations on preventing well blowouts in offshore drilling. While everyone agreed regulations were necessary, industry experts reported that the Obama Administration regulations were more costly and cumbersome than necessary to achieve the safety objectives (Oil and Gas and Sulfur Operations, 2018). Political appointees and career staff disagreed on how to best implement new safety standards. The career employees responded by refusing to do the work necessary to draft the new rule. They would also send internal e-mails deliberately mischaracterizing what political appointees had requested. After much effort, political appointees were ultimately able to get the rule issued. Career employees promptly leaked to the press their opposition and the e-mails mischaracterizing the process. Multiple articles came out about how political appointees had strong-armed career staff (Mann, 2020). The career staff used leaks to make it politically costly to overrule their policy preferences (Sherk, 2023, p. 14).

  • Insubordination. Peter Ohr, a National Labor Relations Board (NLRB) regional director, refused to implement directives on conducting union elections. During the COVID-19 pandemic the NLRB set protocols to determine under what circumstances regions should conduct in-person or mail-in elections (2020). The NLRB typically prefers in-person elections, which have higher turnout and are more representative of employee views. During the pandemic the NLRB expanded the use of mail-in elections, which have lower turnout. Mail-in elections are also seen as easier for unions to win because—unlike employers—unions can campaign at workers’ homes (Morris, 2020). Ohr disregarded these protocols. He scheduled only mail-in elections, even when protocols called for higher turnout, in-person elections (Sherk, 2023, p. 14). A Freedom of Information Act request subsequently uncovered an e-mail exchange in which another regional director celebrated how Ohr’s “brave resistance” prevented President Trump’s appointees from doing “more damage” (Nelson, 2021).

Some career employee resistance has been public and on the record. In 2020 the Federal Labor Relations Authority (FLRA) issued a decision decertifying the National Association of Immigration Judges (U.S. Department of Justice, Executive Office for Immigration Review, 2020). Jessica Bartlett, the FLRA regional director for Washington, D.C., engaged in “willful non-compliance” with that decision for more than 18 months (U.S. Department of Justice, Executive Office for Immigration Review, 2022a, at 626). She defied FLRA regulations and refused to act on that ruling or on a January 2022 ruling reiterating the original directive. Not until the FLRA explicitly ordered her to decertify the union within seven days did she comply (U.S. Department of Justice, Executive Office for Immigration Review, 2022b).

Political scientists have long documented that some career staff pursue their policy preferences over and against those of elected officials (Johnson & Libecap, 1994, pp. 156-171). In one widely cited case study, researchers examined the stringency of clean air enforcement during the Reagan Administration. EPA Administrator Anne Gorsuch—the mother of future Supreme Court Justice Neil Gorsuch—attempted to implement President Reagan’s deregulatory directives. EPA career staff nonetheless increased enforcement stringency. The case study concluded that “the influence of elected institutions is limited when an agency has substantial bureaucratic resources and a zeal for their use” (Wood, 1988).

More recent scholarship has systematically examined how “misalignment” between career officials’ policy views and politically appointed leadership affects agency performance. Researchers find that such misalignment produces cost overruns and delays, which they attribute to “a general ‘morale effect,’ whereby misaligned bureaucrats are less motivated to pursue the organizational mission” (Spenkuch et al., 2023).

The civil service is premised upon the idea that career employees will faithfully provide neutral expertise. In reality, many career federal employees use their positions to advance their personal policy preferences. This directly undermines the federal government’s democratic accountability.

Concerns with democratic accountability go beyond policy setting to effective provision of service. Many government activities have no ideological or political valence; Americans simply want them performed well. Democrats and Republicans alike agree that veterans should get high-quality healthcare. Misconduct or poor performance by government employees can undermine service delivery. For example, senior career officials at the Department of Veterans Affairs (VA) created secret waiting lists to hide the long delays facing veterans seeking care. Hundreds of veterans died at just one VA facility while on these secret wait lists (Boyle, 2016).

For the government to be accountable to the American people federal employees must be meaningfully accountable to the elected President—both in policymaking and in executing the law.

The Competitive Hiring Process is Broken

The federal workforce faces a third widely recognized problem: the competitive hiring process is broken. There is widespread consensus that the federal hiring process needs reform. It takes agencies an average of about 100 days—more than three months—to fill vacant positions in the competitive service (Wagner, 2020). GAO reports this “lengthy hiring time” is driven by “rigid and complex” competitive service hiring procedures (Government Accountability Office, 2019a, p. 28).

Private employers do not have to use these procedures and can hire qualified applicants much more quickly. Consequently, GAO reports that agencies “struggle to compete for talented workers—which is one reason why federal human capital management is an issue on our High Risk list” (Government Accountability Office, 2019b). As the Partnership for Public Service explains:

[T]he federal recruiting and hiring process is in drastic need of repair. The federal government has long struggled to attract the talent it needs … Too often, the applicant experience is miserable, plagued by confusing job announcements, a USAJobs platform that is difficult to use, and a cumbersome hiring process that can take months to complete …
[I]n a sign that some agencies struggle to recruit top talent, more than half of all competitive examining certificates—the list of applicants presented to hiring managers for consideration—saw no selection made. The government must think creatively about how it can attract the next generation of talent and be willing to try new approaches” (2020, pp. 1,7).

Congress, the President, and agencies have reacted to the broken hiring process by working around it. OPM argues that federal positions “are, by default, placed in the competitive service” and the competitive service is “the norm rather than the exception” (88 Fed. Reg. 63874). However, OPM’s data shows that 47% of new federal hires in FY 2022 were excepted service appointments. Through the first six months of FY 2022 a narrow majority of new federal hires have joined the excepted service (Office of Personnel Management, 2022b).[4] The competitive service is the norm on paper only. In practice, it is now the exception as much as the rule.

Moreover, agencies now widely use “Direct Hire Authority” (DHA) to fill competitive service positions without going through the competitive hiring process. An MSPB study recently found that DHA hiring rose from 4% of all competitive appointments in the early 2000s to 27% in FY 2018 (2021, p. 3). Accounting for both excepted service and competitive service DHA hires, only about a third of newly hired federal employees are appointed through competitive hiring procedures.[5]

“This rapid growth [in direct hiring] is a clear indicator of frustration with the results of current competitive service hiring procedures” (Merit Systems Protection Board, 2021, p. 23). In the words of the Partnership for Public Service, “a broken federal hiring process has made it challenging for agencies to recruit and hire the vital talent needed to ensure our government works as effectively and efficiently as it could” (Partnership for Public Service, n.d.). Agencies work around the broken federal hiring process when they can. When they cannot, they must either leave positions vacant or hire sub-optimal candidates.

OPM’s proposed rule argues that EO 13957 “failed to address the fact that the competitive hiring process permits agencies to assess all competencies that are related to successful performance of the job” (88 Fed. Reg. 63869). OPM’s argument ignores the widely recognized flaws in the competitive hiring process, which have driven Congress, the President, and agencies to frequently bypass it. “Rigid and complex” procedures and “lengthy hiring time[s]” make it difficult for agencies to recruit top talent. Talent shortfalls are problematic for any federal position, but they are especially problematic in policy-influencing positions that can affect the performance of an entire agency.

Additionally, many leadership qualities are intangible. Agencies cannot easily quantify applicants’ temperament, acumen, or sound judgment—but these are essential attributes for employees in senior policy influencing positions. So not only do competitive hiring procedures make it difficult for agencies to hire top talent, they sharply constrain agencies’ discretion in evaluating inherently intangible qualities that are nonetheless essential for successful performance in the role.

Civil Service Procedures Cause These Challenges

Civil service procedures are one of the major reasons the federal government faces hiring, performance management, and misconduct problems. GAO, the MSPB, and the Partnership for Public Service recognize that competitive hiring procedures prevent agencies from hiring the best talent. Civil service dismissal procedures are also lengthy, complex, and uncertain. They make removals very difficult, entrenching poor performance and intransigence in the federal workforce.

The Government Accountability Office (GAO) estimates it takes agencies six months to a year (or more) to dismiss a poor performer using Chapter 43 procedures (2015). Afterwards employees typically have multiple options for administrative appeals, such as appealing to the MSPB, filing a union grievance, or bringing an Equal Employment Opportunity (EEO) complaint. Appeals often take even longer to complete. The average arbitration award in a dismissal grievance, for example, comes 17 months after the employee’s initial separation from federal service (Sherk, 2022, p. 7). After administrative appeals conclude, employees can typically appeal to federal court. Dismissing a federal employee is a complex process that often takes several years.

Civil service protections have the practical effect of making the removal of a federal employee for all but the worst behavior prohibitively difficult. A longtime federal management consultant (and former civil servant) wrote a book documenting his experiences trying to improve agency operations. He explains that:

[A]pologists for the system will always say, ad nauseam, that the problem isn’t the rules and procedures; it’s the supervisors and managers who won’t do the “hard work” required to document problems and comply with all those rules and procedures. But that tired truism only serves to confirm what’s painfully obvious to supervisors and managers – that in many if not most cases it’s simply too hard, not to mention risky and time-consuming, to run all the many wickets required to take meaningful action. The fact that any performance- or discipline-based actions are pursued at all, despite the formidable hurdles involved, is a tribute to the grit and determination of the managers involved.

Case in point: in my consulting career I worked on a lot of big projects with grand(iose) objectives, but among the more satisfying things I ever did was to help an individual manager run those wickets, and successfully remove two non-performing employees. This particular manager was new to government, having been recruited based on his extensive private sector experience to help start up a small, highly specialized organization. He simply could not countenance the presence of highly-paid, non-performing employees – despite advice to the contrary from others more experienced in the ways of Fedworld – and was determined to get them off the taxpayers’ dime. He didn’t know it was too hard, or that as a federal manager you shouldn’t waste time on Quixotic exercises like removing incompetent employees. He just did it, at great cost to his own productivity and morale during the many months it took to conclude the actions. While the end result was very positive for the agency, the time, energy, and resources required to reach that outcome detracted materially from the mission, and took a heavy toll on everyone involved in the process (Mills, 2010, pp. 30-31).

These problems are systematic; surveys show federal supervisors widely lack confidence in their ability to remove subordinates who merit dismissal. MSPB’s Merit Principles Survey asked federal supervisors if “a subordinate employee engaged in serious misconduct, are you confident that you would be able to remove that employee?” Just two-fifths of federal managers reported confidence that they could. MSPB also asked supervisors if “a subordinate employee was deficient in a critical performance element after completion of a PIP [Performance Improvement Period], are you confident that you would be able to remove that employee?” Just one-quarter of supervisors were confident that they could (Merit Systems Protection Board, 2019, pp. 6, 15). Supervisors find it very challenging to remove employees for poor performance or misconduct (including policy resistance).

Chapters 43 and 75 have proven to be longstanding and entrenched barriers to effectively addressing performance and conduct issues. During the Clinton Administration an MSPB study found that nearly four-fifths of federal supervisors have managed subordinates with performance problems, but less than a quarter attempted to remove or demote them. The study concluded that “many supervisors believe it is simply not worth the effort to attempt to remove federal employees who cannot or will not perform adequately” (Merit Systems Protection Board, 1995, pp. 2, 5).

Civil service procedures consequently drive the performance and misconduct problems discussed in section I. The reality is that they give federal employees “a de facto form of life tenure, akin to that of Article III judges … What’s more, federal employees know it—and they take full-throated advantage of it” (Feds for Medical Freedom v. Biden, 2023, at 391). Civil service procedures empower intransigent employees to pursue their personal agendas with little practical fear of dismissal, while shielding poor performers from accountability.

In addition to sheltering poor performers, removal protections directly make federal employees less productive. Economists consistently find that giving employees removal protections reduces their productivity (Ichino and Riphahn, 2005; Martins, 2009; Riphahn, 2004; Scoppa, 2010; Scoppa and Vuri, 2014). Unsurprisingly, some employees do not work as hard when firing them is difficult. Economists similarly find that organizations where supervisors can fire employees are substantially more productive than organizations where they lack this authority. The mere possibility of firing substantially increases employee productivity (Corgnet, Hernán-González, and Rassenti, 2015). Making it harder to remove poor performers encourages poor performance.

These problems are especially serious in policy-influencing positions. Poor performance or misconduct by a line federal employee primarily affects the performance of only their duties. An ineffective wage and hour investigator, for example, might miss overtime violations at a particular establishment. But poor performance or misconduct by a policy-influencing employee can affect how the entire agency performs its mission. Guidance drafted by an ineffective regulation writer, for example, could impair the Environmental Protection Agency’s effectiveness in maintaining clean air and water for the public, the Food and Drug Administration’s ability to ensure safe and effective drugs, or the Transportation Department’s ability to maintain the traveling public’s safety. Incompetence or ineffectiveness in these roles can have life or death consequences for the public.

OPM asserts that agencies can address problematic performance or conduct by using the civil service removal procedures set forth in chapters 43 and 75 of title 5, United States Code (88 Fed. Reg. 63862, 63868). OPM does not address the evidence showing that these procedures are a major reason these problems exist in the first place. OPM’s argument also ignores the fact these procedures have proven ineffective in addressing these challenges—as demonstrated by their persistence.

The Schedule F Solution

President Trump addressed these longstanding challenges with Executive Order 13957 on Creating Schedule F in the Excepted Service. The order created a new “Schedule F” in the excepted service for career officials in policy-influencing positions. Under 5 U.S.C. § 7511(b)(2), such positions are exempt from chapter 75 removal protections. EO 13957 directed agencies to identify policy-influencing career positions and ask OPM to put them into Schedule F.

Schedule F was designed for career officials, not political appointees. Section 6 of the order prohibited agencies from committing Prohibited Personnel Practices (PPP) against employees in, or applicants for, Schedule F positions. Agencies could not, for example, hire or fire employees in Schedule F based on their party registration or campaign contributions. The order made it clear that the White House Office of Presidential Personnel—which vets political appointments across the government—would have no role in filling Schedule F positions.

EO 13957 effectively made policy-influencing career positions at-will, without removal protections, while maintaining their status as non-political positions filled based on merit and lasting beyond a single presidential administration. Trump Administration officials estimated that if OPM granted every request, Schedule F would cover a maximum of 50,000 positions—about 2 percent of the federal workforce (Swan, 2022).

Schedule F would have effectively addressed the workforce problems discussed in section I, at least with regard to senior policy-influencing positions. Under Schedule F, the removal protections that make dismissing problematic employees prohibitively difficult would no longer apply. And, because Schedule F was in the excepted service, agencies could quickly fill these positions without the “rigid and complex” competitive service hiring procedures that impair federal recruitment. The order would have effectively given agencies direct hire authority for policy-influencing positions. EO 13957 was designed to make senior civil service leadership more responsive to agency priorities.

Taking Successful State Policies National

Supreme Court Justice Louis Brandeis wrote that “a single courageous state may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country” (New State Ice Co. v. Liebmann, 1932). Much of OPM’s case for preventing the return of Schedule F rests on the incorrect supposition that this reform was novel, risky, and would yield harmful results. The opposite is true. The federal civil service used to operate largely at-will, and many state workforces still do. Schedule F would have simply adopted reforms at the federal level that have proven effective in the states.

Several states have experienced similar challenges with removal protections and civil service hiring procedures. They have responded by making some or all state employees at-will. State reforms include:

  • Arizona enacted legislation in 2012 making most state government employees at-will;

  • Florida removed employment protections for state managers and supervisors in 2001;

  • Georgia placed state employees hired after July 1, 1996, in a new civil service system without employment protections, and that system now covers almost all Georgia state government employees;

  • Indiana increased the number of at-will employees in state government in 2011, while reducing the importance of seniority for those retaining employment protections;

  • Mississippi has periodically temporarily exempted several state agencies from civil service protections;

  • Missouri enacted legislation in 2018 making the vast majority of state government employees functionally at-will;

  • Texas abolished its centralized civil service system in 1985; and

  • Utah has exempted about a third of state employees from employment protections and also passed legislation in 2022 making all newly hired managers and supervisors at-will.

These reforms have been successful. Evaluations generally show positive results, while fears of a return to patronage failed to materialize (Coggburn, 2007; Condrey & Battaglio, 2007; Cournoyer, 2012; Gossett, 2003; French & Goodman, 2011; Kim & Kellough, 2014). For example, one study asked state HR directors in Georgia, Florida, Texas, and Mississippi about their experiences with at-will employment. By a 60% to 18% margin, they reported that at-will employment “helps ensure employees are responsive to the goals and priorities of agency administrators.” Majorities reported that at-will employment “provides essential managerial flexibility,” “makes the HR function more efficient,” and is “an essential piece of modern government management.” Conversely, little patronage hiring or improper termination was reported (Coggburn et al., 2010, Tables 1 and 2). Many states operate effective, professional state governments with at-will employment. EO 13957 would have taken these successful reforms national, at least with respect to policy-influencing positions.

OPM contends that “there is little evidence to support the notion that … allowing the firing of career civil servants without appropriate process … will improve the government’s performance” (88 Fed. Reg. 63881).[6] OPM also argues that civil service employment protections “are a small price to pay to deliver to the American people a merit-based civil service rather than a system based on political patronage” (88 Fed. Reg. 63880).

These arguments ignore the evidence from the states that at-will employment is both consistent with a merit system and can improve government performance—indeed “is an essential piece of modern government management.” Many states (including two of America’s three largest) operate effective and professional state governments with at-will employment for senior career officials. The federal government can and should do the same.

Returning to the Founding Principles of the Merit Service

Schedule F would have returned a small portion of the federal workforce to the principles under which the merit service long operated and that prevailed during the civil service’s golden era. The 19th-century reformers who created America’s civil service believed that tenure and job protections were inimical to merit. The Pendleton Act consequently deliberately made minimal changes to the dismissal process. While it prohibited removing employees because they made—or failed to make—political contributions, the Pendleton Act did not otherwise interfere with the President’s general authority to remove employees (Frug, 1976, p. 955).

Civil service reformers wanted to eliminate patronage by regulating hiring, while leaving the government free to remove problematic employees. One historian of the civil service summarized the view of the reformers as “if the front door were properly tended, the back door would take care of itself” (Van Riper, 1958, p. 102). The Pendleton Act implemented this vision, and the civil service operated on this principle for the next eight decades. Not until President John F. Kennedy signed Executive Order 10988 in 1962 did the general federal workforce gain the right to appeal removals (88 Fed Reg. 63864).[7]

OPM points to the Lloyd-La Follette Act of 1912 as giving federal employees a property right to their jobs (88 Fed. Reg. 63866). While the Supreme Court interpreted the Act as having that effect in Arnett v. Kennedy (1974), the Act and its predecessor executive orders were not understood (or applied) to have this effect before that decision.

The Lloyd-La Follette Act originated from a Civil Service Commission request for a presidential order requiring agencies to explain their reasons for removing employees as a safeguard against politically motivated removals. In 1897 President William McKinley issued an executive order providing that civil servants could be removed only “for just cause, upon written charges … of which the accused … shall have an opportunity to make defense” (Frug, 1976, p. 956). The Civil Service Commission subsequently became concerned that McKinley’s order could be interpreted to require a trial to determine if “just cause” existed. The Civil Service Commission feared that “to require [a trial] would not only involve enormous labor, but would give a permanence of tenure in the public service quite inconsistent with the efficiency of that service.”

Consequently, and upon the Civil Service Commission’s recommendation, President Theodore Roosevelt issued a follow-up executive order in 1902 clarifying that “just cause” means any cause that promotes the efficiency of the service and that trials or examination of evidence were unnecessary to remove an employee (Frug, 1976, p. 957). In February 1912, President William Howard Taft issued Executive Order 1471, reaffirming the McKinley and Roosevelt orders. The Civil Service Commission explained that Executive Order 1471 did not give federal employees any right to their office or a hearing before they could be discharged:

The rules are not framed on a theory of life tenure, fixed permanence, nor vested right in office. It is recognized that subordination and discipline are essential, and that therefore dismissal for just cause shall be not unduly hampered. The rules have at all times left the power of removal as free as possible, providing restraints only to ensure its proper exercise ... the public service is governed by the same theory as private service, in which tenure of place depends upon good behavior and efficiency … The only restriction that has been imposed is that employees should not be removed for political or religious reasons or upon secret charges …

Appointing officers, therefore, are entirely free to make removals for any reasons relating to the interests of good administration, and they are made the final judges of the sufficiency of the reasons. No examination of witnesses or any trial or hearing is required ... The rule is merely intended to prevent removals upon secret charges and to stop political pressure for removals .... No tenure of office is created except that based upon efficiency and good behavior (U.S. Civil Service Commission, 1913, p.21-22).

Congress codified Executive Order 1471 into law verbatim as the Lloyd-La Follette Act of 1912. The law required agencies to provide employees with a notice and an opportunity to respond before removal, while expressly providing that “no examination of witnesses nor any trial or hearing shall be required.”

Lloyd-La Follette statutorily codified the existing civil service policy that prohibited removals for narrowly defined purposes (i.e., political activities or religious belief), while otherwise giving agencies free rein to determine when employees should be removed. Pre-Arnett, lower courts “consistently rejected” the notion that Lloyd-La Follette and its predecessor executive orders gave federal employees a right to contest dismissals (Frug, 1976, pp. 958, 970).

From the passage of the Pendleton Act of 1883 until the 1960s federal employees had neither tenure, a judicially cognizable property interest in their jobs, nor any right to appeal removals. Employment at the discretion of the agency, coupled with prohibitions on discriminatory removals, was the founding and longstanding vision for the merit service.

In retrospect, this period was the civil service’s golden era. In the 1950s and early 1960s more than 70% of Americans reported they trusted the government to do what is right “just about always” or “most of the time.” By the end of the 1960s that figure had fallen to 62 percent, and it fell below 30% by the end of the 1970s. Today just 16% of Americans trust the government to mostly do what is right (Pew Research Center, 2023). Americans broadly trusted and respected the civil service when agencies had broad discretion to remove employees. Public trust in government plummeted after that changed—and has never recovered.

OPM argues that its proposed rule guards against a return to Schedule F “needlessly politicizing our nation’s nonpartisan career civil service.” The Office also argues that reinforcing removal protections is necessary to prevent harmful turnover among career professionals (88 Fed. Reg. 63880). These arguments ignore the fact that EO 13957 simply returned a small portion of the civil service to its historical baseline—the baseline that prevailed during its golden era. Schedule F would have made up to 50,000 policy-influencing employees effectively at-will, while retaining prohibitions against discriminatory removals. This was how the government operated between 1883 and 1962 under the Pendleton and Lloyd-La Follette Acts. The historical evidence refutes OPMs contention that returning 2% of the civil service to this policy would impair government operations.

Civil Service Laws Authorize Schedule F

Schedule F was firmly grounded on presidential authority under Title 5 and the Constitution. Statutory context makes clear that the terms “positions of a confidential, policy-determining, policy-making, or policy-advocating character” include both career and political appointments. The law also allows the President to except positions from the competitive service to facilitate policy control of the bureaucracy. OPM leadership may now disagree on policy grounds, but the President had clear legal authority to issue EO 13957.

The Policy Influencing Exceptions Include Career Positions

The Civil Service Reform Act of 1978 (CSRA) created 5 U.S.C. § 7511. That section prohibits employees in excepted service “positions of a confidential, policy-determining, policy-making, or policy-advocating character” from appealing adverse actions. OPM has historically made limited use of this authority; OPM regulations place only “positions of a confidential or policy-determining nature” in Schedule C, a schedule historically only used for political appointments (5 C.F.R. § 6.2). EO 13597 recognized that many career positions also involve making, determining, or advocating for policy. The order created Schedule F in the excepted service for such career positions. Under § 7511(b)(2) employees transferred to Schedule F could no longer appeal adverse actions to the MSPB.

OPM now contends that EO 13957 was based on a faulty legal premise and that the phrase “positions of a confidential, policy-determining, policy-making, or policy-advocating character” is just “a shorthand way” of describing political appointments. OPM argues EO 13957 “improperly appl[ied]” these terms to career positions “contrary to congressional intent” (88 Fed. Reg. 63872-63873). OPM proposes regulatory amendments to clarify that these statutory terms refer exclusively to non-career political appointments and do not encompass career employees. While OPM may wish that the CSRA contained those limitations, it does not. The text of the CSRA makes clear the President’s authority to issue EO 13957.

A basic canon of statutory construction is that words should be understood to have their ordinary, everyday meaning unless the context indicates otherwise. Nothing in the words “confidential, policy-determining, policy-making, or policy-advocating” hints at covering only political appointments, or references the duration of an employee’s tenure. Instead, the CSRA makes clear these terms cover both career and non-career positions.

The CSRA also created the Senior Executive Service (SES). The CSRA explicitly distinguishes between SES career and non-career (i.e., political) appointments. It also specifically limits SES civil service appeals to “a career appointee” (5 U.S.C. § 7541). But chapter 75 includes no similar distinction between career and non-career status for either excepted service employees or excepted service adverse action appeals.

Another basic canon of statutory construction is that if “Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion” (INS v. Cardoza-Fonseca, 1987). Congress expressly distinguished career and non-career status for SES adverse action appeals but omitted that language for the excepted service. This strongly implies the terms “confidential, policy-determining, policy-making, or policy-advocating” should not be read as a “shorthand” reference to political appointments. Congress used different language in the same chapter of the same law to distinguish career and political appointees.

Moreover, the CSRA expressly applies the terms “policy-determining” and “policy-making” to career positions. The CSRA describes SES members as exercising “policy-making” and “policy-determining” functions (5 U.S.C. § 3132(a)(2)(E)). The CSRA also caps non-career SES appointments at 10 percent of the SES (5 U.S.C. § 3134(b)). The very law that authorized excepting “policy-determining” and “policy-making” positions from chapter 75 used these terms to describe positions that must be held primarily by career officials.

The “presumption of consistent usage” is another canon of statutory construction; courts presume a word or phrase bears the same meaning through a statute (Brown v. Gardner, 1994). If the terms “policy-making” and “policy-determining” encompass career positions in § 3132, then they also encompass career positions in § 7511.

A 1994 amendment to title 5 also presupposed that career incumbents can lose statutory protections if their positions are declared policy-influencing. The CSRA initially excluded policy-influencing positions in the excepted service from PPP restrictions. Public Law 103-424 amended 5 U.S.C. § 2302(a)(2)(B) to limit this exclusion to positions declared policy-influencing “prior to the [relevant] personnel action.” The amendment prevented a position reclassification from retroactively sanctioning PPPs.

Courts presume that amendments have real and substantial effect; they avoid statutory interpretations that render subsequent amendments superfluous (Babbitt v. Sweet Home Chapter, Communities for a Great Oregon, 1995). Under OPM’s interpretation the policy-influencing terms are merely shorthand for political appointments with no relevance to career incumbents. In this view, the government cannot deprive career incumbents of statutory protections by reclassifying their positions as policy-influencing. If that were correct the 1994 amendment would have been entirely redundant. OPM’s interpretation thus violates yet another canon of statutory construction, the presumption of purposive amendment.

OPM’s arguments that the policy-influencing terms encompass only political appointees are weak. The Office cites the Brownlow report (88 Fed. Reg. 63872), which concluded that “the positions which are actually policy-determining, however, are relatively few in number. They consist, in the main, of the heads of executive departments, under secretaries and assistant secretaries … and a limited number of other key positions” (Presidents Committee on Administrative Management, 1937, p. 8). The Brownlow report may provide insight into the definition of a “policy-determining” employee. But the CSRA also excepted “policy-making” and “policy-advocating” employees from adverse action appeals. Far more employees are involved in “making” and “advocating” for policy than ultimately “determining” it. The Brownlow report accords with how EO 13957 interpreted chapter 75. Moreover, § 3132(a)(2) postdates the Brownlow report by four decades and gives a much broader scope to the terms “policy-determining” and “policy-making.”

OPM cites a pair of MSPB decisions to argue that the policy-influencing terms are shorthand for political positions (88 Fed. Reg. 63873). MSPB decisions have little relevance here. Chapter 75 gives the President, OPM, and agency heads responsibility for determining that positions are policy-influencing. MSPB has no role, deferring instead to their determination. MSPB case law does not and cannot determine the scope of these exemptions.

OPM also cites legislative history and committee reports to argue that Congress understood the policy-influencing language to refer to political appointments (88 Fed. Reg. 63872). However, this legislative history merely confirms that the policy-influencing exception covers Schedule C political appointees, as it surely does. But the legislative history does not state that the policy-influencing terms cover only political appointments and exclude career employees. Further, the CSRA expressly applied them to career positions. Courts “do not resort to legislative history to cloud a statutory text that is clear” (Ratzlaf v. United States, 1994).

The CSRA makes clear that the § 7511(b)(2) exception includes career positions. OPM can apply this exception sparingly; most administrations have done so. But the Trump Administration had—and a future administration will have—full legal authority to extend the policy-influencing exception to career positions.

The President Can Except Positions to Ensure Policy Control

The CSRA also allows the President to except positions from the competitive service for the purpose of removing civil service protections. OPM argues that title 5 “does not purport to confer authority on the President to except positions from the provisions of chapter 75” (88 Fed. Reg. 63868). However, 5 U.S.C. § 3302 authorizes the President to make necessary exceptions from the competitive service when “conditions of good administration warrant.” Congress did not limit this authority to cases where competitive examination is impractical. Combatting performance problems or policy resistance also constitutes “conditions of good administration.” Further, § 7511(b)(2) expressly authorizes the President to either directly, or indirectly through OPM, determine that positions are policy influencing and therefore excluded from adverse action appeals. In combination, these authorities authorize excluding existing positions from chapter 75.

The fact that prior presidents have restrained themselves in their dealings with subordinates does not imply they lacked this authority. The Supreme Court has already concluded that “policymaking positions in government may be excepted from the competitive service to ensure presidential control, see 5 U.S.C. §§ 2302(a)(2)(B), 3302, 7511(b)(2)” (Free Enterprise Fund v. Public Company Accounting Oversight Board, 2010). Opponents of Schedule F may wish the President did not have this authority. Nonetheless, he does.

OPM implies that the Constitution prevents removal restrictions from being eliminated. Federal courts have consistently rejected this argument. OPM accurately explains that the Supreme Court has held that civil service protections give government employees a property interest in their job. Consequently, the government cannot constitutionally dismiss tenured employees without due process (88 Fed. Reg. 63865-63866). However, the government can remove civil service protections. Doing so extinguishes the underlying property interest they create. Federal courts have repeatedly rejected challenges to state civil service reforms. They have uniformly held the legislative or administrative proceedings that removed tenure protections provided all the process that was due.[8] The courts have also held that federal agencies can declare positions policy-influencing and thereby eliminate civil service protections (Stanley v. Dep’t of Justice, 2005; Stanley v. Gonzales, 2007). Both the CSRA and the Constitution permitted Schedule F.

OPM’s Schedule F Criticism Misplaced

OPM’s proposed rule extensively criticizes EO 13957 and Schedule F. These criticisms are based on misconceptions. Schedule F rejected the spoils system, recognized the importance of career expertise, and reinforced the merit system. It also discouraged political appointees from burrowing in. Contrary to OPM’s assertions, EO 13957 would have had little effect on recruitment and retention—particularly in technical positions to which it did not apply. OPM’s opposition to Schedule F is misplaced.

Schedule F Rejected the Spoils System

OPM’s rulemaking is premised on the notion that EO 13957 was attempting to resurrect the spoils system. This is false. The order rejected patronage and recognized the importance of career expertise. Schedule F was instead designed to make senior civil service leadership more responsive to agency priorities.

OPM’s central rationale for this rulemaking—which is designed to prevent the rapid reissuance of EO 13957—is the need to prevent a return to the spoils system. OPM argues that “this rulemaking is needed to preserve the integrity of the Federal career workforce as an independent entity free of political influence” (88 Fed. Reg. 63878). OPM maintains the rule will reduce the risk of a reissued Schedule F “needlessly politicizing our nation’s nonpartisan career civil service” (88 Fed. Reg. 63881). This central rationale is wholly mistaken. EO 13957 rejected the spoils system, and the spoils system is an anachronism in any event.

Under the patronage system, the party in power rewards its activists and donors with government jobs. If the party loses power, many—if not all—of those employees get replaced by the other party’s activists. Justice Powell’s dissenting opinion in Elrod v. Burns (1976)—in which a majority of the Court held the patronage system unconstitutional—explained the rationales for this system: it encouraged participation in the democratic process and stable political parties. These rationales depend on the ability of the party in power to reward campaign supporters with government jobs.

As explained in section III, EO 13957 forbade agencies from filling Schedule F positions based on political affiliation or campaign contributions. It also excluded the White House Office of Presidential Personnel from involvement with Schedule F. So the order expressly barred the type of patronage appointments that drive the spoils system. OPM’s proposed rule completely ignores these facts. OPM’s argument that EO 13957 would somehow resurrect the spoils system while it prohibited patronage appointments is self-contradictory.

OPM also ignores the fact that, if EO 13957 was intended to fill the bureaucracy with political loyalists, President Trump chose an extremely odd way of doing it. He could have directly converted career positions to political positions, dismissed career incumbents through a reduction in force, and filled the roles with political appointees. Presidents of both parties have done this before (Exec. Order 12300, Exec. Order 12940). Legal challenges to such actions failed (Brunton v. United States, 1981). If President Trump wanted more patronage appointments, it would have been considerably easier to accomplish directly under existing authorities than to create a new excepted service schedule.

EO 13957 did not do this because it was motivated by quite different concerns. As discussed in sections I and II, removal restrictions insulate career officials from accountability to the President while the competitive hiring process makes recruiting top talent difficult. Schedule F addressed these problems. By giving agencies de facto direct hire authority for policy-influencing positions, while eliminating removal restrictions, the order gave them tools to hire talented senior officials and hold them accountable. This would make senior career officials more responsive to agency leadership, while retaining their status as career employees whose positions last beyond a single administration.

OPM’s concerns about a return to the patronage system also ignore the evidence that the federal government ended patronage because it had become obsolete. The federal government grew rapidly after the Civil War. Between 1871 and 1881 the federal workforce doubled, increasing from 51,000 to 100,000 civilian employees (Johnson & Libecap, 1994, p. 17). This expanded scale made monitoring and managing patronage employees much harder for Congress and the President. Elected officials had to spend much of their time vetting and arranging patronage appointments, while appointees were subject to less rigorous scrutiny and often provided poor services that angered the voting public. Patronage also focused federal employees attention on the parochial concerns of local party machines instead of the national concerns of the President and Congress. By the early 1880s it had become clear to both Congress and the President that the spoils system did not benefit them. Congress passed—and the President signed—the Pendleton Act because patronage no longer served their interests (Johnson & Libecap, 1994, pp. 12-47).

The issues that made the spoils system untenable in the 1880s have grown in the intervening century and a half. The federal workforce is over twenty times the size it was in the early 1880s, while far more federal positions require specialized subject matter expertise. Converting even a fraction of the bureaucracy to political appointments would be administratively infeasible and counterproductive. No President will bring back the spoils system because doing so would make it more difficult to implement the President’s agenda.

Recruiting qualified personnel for existing political positions has proven challenging enough for Presidents of both parties. The Partnership for Public Service tracks 811 key positions requiring presidential appointment with Senate consent. By the end of their first year in office Presidents George W. Bush, Barack Obama, Donald Trump, and Joe Biden had not even nominated candidates for between 134 and 256 of these positions (Partnership for Public Service, 2022). As of the date of this comment—nearly three years into his administration—President Biden still has not nominated candidates for 78 such positions (Partnership for Public Service, 2023). The supposed threat of a President converting even the two percent of the federal workforce covered by Schedule F—50,000 positions—to political appointments is farcical. No administration could fill that many positions with qualified personnel or would want responsibility for doing so.[9] Administration’s struggle to fill their current slate of political appointments.

External observers also recognize this reality. Scholars at the University of Arizona and Montana State University explain that:

[Federal unions argue] that, left to their own devices, politicians will readily reinstate the spoils system and that only the efforts of public-minded citizens and the courts prevent them from doing so. Although such a characterization is useful to federal unions, it ignores the interests of federal politicians that we have emphasized in this volume. There is no returning to patronage. Since 1883, the President and the Congress have had important reasons for limiting patronage, and those reasons remain today (Johnson & Libecap, 1994, p. 171).

A former longtime civil servant and federal management consultant put it somewhat more pithily. He explained that:

The sad irony in all this is that the deep dysfunction of today’s federal employment culture stems from pursuit of an objective that is no longer relevant. All of the myriad rights, entitlements, safeguards, and due process mechanisms that have proliferated over the years, and which in turn have led to the malaise and torpor we see today, are rooted in the 19th century notion that the civil service must be shielded and protected: from political influence, manipulation, or corruption. But the very real scandals and threats that motivated the original architects of our civil service tradition have long since receded into historical memory, replaced by a new and vastly different set of challenges …few even modestly well-informed people seriously believe that today’s rank-and-file federal jobs represent a treasure trove of potential patronage plums, ripe for the picking by rapacious politicos save only for the bulwark of the federal civil service system.

Compared to 1883, a federal job today holds nothing like the allure it once did, at least to most people ... the really big bucks aren’t in the political appointments game. If your goal is to tap the public treasury, you don’t want to be an assistant secretary; you want to be a division manager of a defense contractor (ideally, with a Congressperson or two in your corner) …

The notion of a pristine, protected federal civil service is an anachronism, a 19th century solution to problems from an era long past, fundamentally ill-suited to the realities of contemporary American government ... To borrow an acronym from the lexicon of Fedworld, our legacy civil service has indeed been OBE: overtaken by events (Mills, 2010, pp. 56, 58, 60, 62).

The spoils system is an anachronism. Even if the Supreme Court had not declared it unconstitutional in Elrod v. Burns, neither EO 13957 nor anything else would threaten its return to the federal workforce. OPM offers no reason to believe otherwise. OPM’s central rationale for this rulemaking is combatting a problem that became irrelevant over a century ago. Regulations are no more necessary to protect Americans from the spoils system than from unsafe horse-and-buggy designs.

Removal Restrictions Unnecessary to Protect Merit

Even if the threat of the patronage system had not “long since receded into historical memory,” OPM’s proposed regulations would have little justification. OPM argues that removal restrictions “are a small price to pay” for “a merit-based civil service rather than a system based on political patronage” (88 Fed. Reg. 63880). This argument is a non-sequitur. As discussed in section III.B, the merit system operated for eight decades with federal employees generally unable to appeal dismissals; the Lloyd-La Follette Act expressly provided that no trial or hearing would be required to effectuate removals. Many state governments currently operate at-will. Nonpartisan, merit-based civil services can, do, and did operate effectively at-will. Schedule F’s elimination of those restrictions is fully consistent with an effective merit service.

Nonetheless, OPM’s confusion on these points is understandable. The Office explains that a request from federal unions prompted this rulemaking (88 Fed. Reg. 63877). Federal unions have long used the specter of the spoils system to oppose civil service reforms:

In countering civil service reform efforts, federal unions can resort to the use of popular myths about patronage. That is, any effort to reduce the privileges held by rank-and-file employees and to strengthen political control of the bureaucracy can be cast as a return to patronage. In this way, the bureaucracy is attempting to control or to mold the flow of information in the debate over civil service reform. By raising the “devil” of patronage, the discussion is diverted from an analysis of the private benefits received by career employees under the current system to a debate over the risks of dismantling civil service protections and reinstating patronage …

[F]ederal employee unions have been skillful in capitalizing on public fears of patronage by labeling reforms of the civil service as returns to the spoils system The fear of patronage is a convenient ploy to which federal employee unions can turn in responding to criticisms about the performance of the bureaucracy (Johnson & Libecap, 1994, pp. 171, 181-182).

Federal employees enjoy job protections that are virtually unheard of in the private sector. Federal employees and their unions would be hard pressed to defend this benefit on its own terms. So they have historically instead attacked efforts to increase federal employee accountability as a return to patronage (Johnson & Libecap, 1994, pp. 181-182). Union claims that EO 13957 would reinstitute the spoils system—despite the order prohibiting patronage—are simply another example of this phenomenon.

The argument that government employees need removal restrictions to protect the merit system is a “popular myth.” Such historically inaccurate propaganda provides no basis for a rulemaking. OPM should withdraw this rule in its entirety.

Recruitment and Retention Concerns Meritless

OPM’s recruitment and retention concerns are also meritless. OPM contends that this proposed rule would aid agency recruitment and “supports the retention of Federal career professionals who provide [] continuity of institutional knowledge and subject matter expertise.” OPM also cites “a vast body of research” that shows the government benefits from career federal employees who enjoy stable public jobs (88 Fed. Reg. 63880-63881).

This research is neither in dispute nor relevant to this rulemaking. EO 13957 itself recognized and appreciated the value of career expertise in the federal workforce. It prohibited patronage and stipulated that Schedule F positions would last beyond a presidential term. It did so precisely because career officials have valuable experience and institutional knowledge. The order simply sought to better leverage this expertise to improve agency functioning and advance the President’s agenda.

Contrary to OPM’s concerns, Schedule F employees would keep their jobs so long as they performed well and faithfully advanced the President’s agenda. They would continue to enjoy the benefits of long-term stable jobs and to train and mentor newer and less experienced federal employees. OPM’s argument would only make sense—and this body of research would only be relevant—if Schedule F employees were converted to political appointees who lost their jobs with each presidential transition. The order expressly did not do that. The Office justifies its rule by arguing against an executive order that never existed. That misconception does not justify preemptively preventing the President from holding senior policy-influencing officials more accountable.

OPM’s recruitment concerns are similarly meritless. Schedule F would have virtually no applicability to technical positions such as IT and cybersecurity that OPM cites as ongoing recruitment challenges. Those positions —like well more than 90% of federal jobs—are not policy-influencing. Recruitment challenges in positions wholly unaffected by EO 13957 provide no justification for this rule. Nor does OPM offer any evidence that making policy-influencing career positions at-will—as opposed to converting them to political appointments—would create recruitment challenges. As discussed in section III.A, many states, such as Texas, Georgia, and Florida, operate at-will career workforces. By a 19 percentage point margin, HR professionals in these states conclude that at-will employment “makes the HR function more efficient” (Coggburn et al., 2010, p. 196). OPM’s recruitment concerns have not materialized in states with at-will workforces. OPM presents no reason to believe they would materialize in the federal government either.

Schedule F Discouraged Burrowing In

OPM bizarrely argues that EO 13957 would help political appointees “burrow” into permanent civil service positions (88 Fed. Reg. 63869). This objection makes little sense; Schedule F would have eliminated the value of burrowing in—or at least, burrowing into a policy-influencing position. Civil service protections are what make burrowing in valuable; without them, each new administration could simply dismiss the previous administration’s embedded partisans.[10] Retention would only occur if the new President liked their performance. Under Schedule F, political appointees burrowed into policy-influencing positions would essentially serve on permanent probation. This would protect the integrity of the civil service.

Hiring for Schedule F would also remain apolitical. PPPs—including selection based on politics—would remain prohibited. Instead, hiring would use the same procedures currently used to fill other career excepted service schedules. Burrowing in would have been no greater threat in Schedule F than in Schedule A.

Schedule F Reinforced the Merit System

OPM argues that Schedule F “undermined the foundations of the civil service and merit system principles” (88 Fed. Reg. 63869). This criticism is misplaced. Almost two-thirds of federal employees are now hired outside of competitive service procedures. EO 13957 retained the same merit-based hiring procedures currently used in the excepted service and for DHA positions in the competitive service. Agencies could not hire or fire Schedule F employees for non-merit factors.

Schedule F instead reinforced the merit system. As discussed in section I.A, federal employees widely believe their agencies do not follow Merit System Principle 6. Their modal response to the question of what usually happens to poor performers is that they “remain in the work unit and continue to underperform” (Office of Personnel Management, 2021, p.16; Office of Personnel Management, 2022a, p. 32). Removal restrictions largely prevent agencies from implementing merit principles—as the founders of the civil service feared they would. EO 13957 would give agencies the effective ability to remove poor performers in senior career positions—exactly what Merit System Principle 6 calls for.

OPM’s Proposed Rule Would Hurt the Federal Workforce

Based on these misconceptions OPM has proposed regulations designed to prevent a future administration from bringing Schedule F back. OPM’s rule would regulatorily define policy-influencing positions to cover only political appointees, allow employees whose positions are reclassified as policy-influencing to retain civil service appeals, and allow MSPB review of determinations that positions are policy influencing. The latter reform is transparently designed to allow procedural challenges to determinations that positions are policy-influencing. The rule is designed to prevent a future administration from reissuing Schedule F. OPM’s proposed rule would hurt the federal workforce in several ways. It would encourage poor performance and bureaucratic resistance, discourage vetting policy proposals with career staff, and encourage burrowing in. OPM should abandon this proposed rule in its entirety.

OPM’s Proposed Rule Would Encourage Bureaucratic Resistance and Poor Performance

OPM’s proposed rule would protect poor performers and support bureaucratic resistance. The rule would force agencies to use chapter 43 and 75 procedures to remove employees, regardless of whether they have policy-related duties. As discussed in section II, this would make removing policy-influencing employees prohibitively difficult—giving career employees more latitude to advance their personal policy preferences while protecting poor performers. OPM’s proposed rule almost entirely ignores these problems.

The rulemaking presumes that federal employees impartially carry out presidential and congressional policies without regard to their personal views. Virtually all scholars who study the civil service recognize this is not the case. A voluminous academic literature examines the “principal-agent” problem in public administration—the fact that Congress and the President (the principals) must rely on career federal employees (the agents) to implement their policies, but the agents may have different goals and are difficult to monitor or control. As a result agents can cause policy implementation to differ—sometimes dramatically—from what principals intend (McCubbins et al., 1987; Potter, 2017a; Wood, 1988). Scholars find it “very clear that bureaucrats are not neutral parties in the policymaking process. Rather, they have their own set of interests that they actively work to protect” (Potter, 2017b).

Scholars disagree over whether bureaucratic resistance is good or bad. Some argue that it serves as a moderating function that prevents extreme policy swings (Feinstein & Wood, 2022) and creates a beneficial “internal” separation of powers within the executive branch (Katyal, 2006). Others argue it undercuts democratic accountability by allowing unelected bureaucrats to undermine elected officials’ policies (Johnson & Libecap, 1994, pp. 156-171). But scholars virtually universally accept the fact that federal employees have their own policy views and often seek to advance them (Nou, 2019). Scholars describe the federal workforce as exhibiting “politicized competence” that empowers career bureaucrats to “bend policy to their liking” (Gailmard & Patty, 2007, p. 875). Indeed, after the 2016 election many scholars publicly rejoiced that insulated civil servants would frustrate President Trump’s agenda. For example, one UCLA law professor explained that:

[A]gencies aren’t presidential fiefdoms. They are, themselves, embedded in a secondary scheme of administrative separation of powers, where power is divided—triangulated—among presidential appointees, politically insulated civil servants, and the public writ large authorized to participate in most administrative matters ... this will be a fantastically rude awakening for President-elect Trump, CEO of Trump Organization whose most famous pre-presidential slogan was “you’re fired.”

Consider the civil service. Our professional, politically insulated civil service is ostensibly well positioned to limit presidential overreaching. Presidential appointees are insufficiently numerous and often insufficiently sophisticated to unilaterally make policy. Instead, they necessarily rely on career officials who possess the requisite expertise and bureaucratic know-how to get things done …

The civil service thus can (and should) be seen as a weighty institutional rival to presidential appointees, well situated legally and culturally to resist or stealthily reshape policies they find unreasoned or unlawful. Because of their political insulation and long-term job stability, the civil servants are anything but true subordinates—certainly nothing like Trump’s apprentices. Instead, as scholars such as Neal Katyal, Gillian Metzger, Eric Posner, and Paul Verkuil also recognize, they can be formidable counterweights, with a deep reservoir of technocratic skills and administrative ingenuity to influence, if not dictate, substantive outcomes (Michaels, 2016).

The proposed rule utterly ignores these facts. It does not examine or discuss whether federal employees may have their own goals and motivations or how they behave when their goals differ from the President’s. It disregards the vast literature finding that federal employees do not offer “neutral expertise” but actively seek to align agency policy with their views. OPM spends just two sentences claiming that agencies can use chapter 75 to discipline an employee who “refuses to implement lawful direction from leadership” (88 Fed. Reg. 63862). As discussed in section I.B, chapter 75 procedures have proven insufficient to address such misconduct.

Moreover, scholars have documented that career employees have many “levers of resistance” to advance their own agendas (Nou, 2016). Outright refusal to implement policy is the most extreme and easiest to prove in dismissal proceedings. Subtler tactics are harder to prove, but equally effective—if not more so—in stymieing the President’s agenda. These tactics include slow-walking, withholding information, leaking, constraining agency discretion under the Administrative Procedure Act, and enlisting internal and external allies (Nou, 2016; Sherk, 2023).

Early in the Trump Administration the Washington Post published a veteran federal employee’s guide to “useful tools” for “subtly subvert[ing] stupid orders” without outright revolting. This 42-year veteran of the civil service echoed the tactics scholars have documented. He advised federal employees to:

  • “Only provide minimal information requested”
  • “Fail to find information”
  • “Miss deadlines while ‘doing your best’ (after all, we were all overworked). That might get you a poor review next time, maybe, but it won’t get you canned.”
  • “Keep two sets of data (requires some care!)”
  • “Communicate with friendly Congress people … through your personal email.”
  • “Cultivate trusted media sources … sure beats going through a PAO (public affairs officer)” (Davidson, 2017).

Federal employees widely used these tactics to frustrate Trump Administration policies (Sherk, 2023). Political appointees may know full well that career staff are slow walking draft regulations but still find it challenging to prove the employees’ performance met the legal standard for dismissal. This is especially true if the relevant policy staff do not report directly to a political appointee but have like-minded supervisors who oppose their dismissal. In such circumstances it is virtually impossible for political appointees to effectuate disciplinary actions over supervisors’ objections. Unsurprisingly, researchers find bureaucrats routinely slow-walk regulations they oppose and expedite regulations they support (Potter, 2017a).

Political appointees may similarly be quite sure they know who leaked a draft document but be unable to prove it to the level necessary to sustain a dismissal. As the voluminous literature finding that career officials resist policies they oppose demonstrates, chapter 43 and 75 procedures are insufficient to combat these “levers of resistance.” Civil service protections make it very difficult for agencies to implement policies that senior career officials oppose—even if the voters strongly support them. OPM’s proposed rule would reinforce the job protections that enable policy resistance.

This would particularly impair conservative presidential administrations. The nominally non-partisan federal workforce has moved well to the ideological left. For example, scholars have examined career federal employees’ donations to political candidates over election cycles (Feinstein & Wood, 2022). In the 1990s the average federal employee’s campaign donations went to somewhat more liberal candidates than Americans as a whole, but not greatly so. As the below figure shows, however, in the 2000’s federal employees’ campaign donations shifted decisively to the left.

The above figure shows the evolution of federal career employee campaign donations for a subset of prominent agencies. [11] The vertical axis shows how many standard deviations federal employees stood from the U.S. ideological mean. Positive numbers reflect giving to more conservative candidates, negative numbers to more liberal candidates. Starting with the 2003-2004 election cycle, career employee donations begin moving leftwards. By the 2013-2014 election cycle (the last for which data is available) the donations from the average civil servant in the median federal agency were 1.19 standard deviation to the left of the average donor.[12] This means the typical career civil servant gave to more liberal candidates than 88 percent of all Americans.

Notably, the average career employee gave to more liberal candidates that cycle than President Barack Obama’s agency heads. In 2013 and 2014 career employees gave to more liberal candidates than agency heads did at the Departments of Agriculture, Commerce, Defense, Education, Interior, Health and Human Services, Homeland Security, Housing and Urban Development, Labor, Transportation, and State. They also did so in the Agency for International Development, the Environmental Protection Agency, the National Archives and Record Administration, and the Small Business Administration. President Obama’s political appointees were not as liberal as the career employees they supervised.

Federal employees are, on average, considerably more liberal than America as a whole. In most cabinet agencies they are even more liberal than Democratic political appointees. Policy friction between career staff and political appointees is thus particularly salient during conservative presidencies (Spenkuch et al., 2023). OPM’s rule would make it easier for predominantly liberal career employees to undercut a conservative President’s policies.

This rule would also protect wholly nonideological poor performance and misconduct. Existing civil service procedures make it very hard to hold career officials accountable for their conduct. For example, the MSBP reinstated many VA officials dismissed for misconduct (Graves v. Department of Veterans Affairs, 2016; Rubens v. Department of Veterans Affairs, 2016; Weiss v. Department of Veterans Affairs; 2023). Schedule F would allow agencies to make many senior positions at-will. This would make senior leadership more responsive to agency priorities—including nonideological priorities like effective service delivery. State governments have found that at-will employment has exactly this effect (Coggburn et al., 2010, Table 1). OPM’s proposal would prevent senior career officials from effectively being held accountable to the President—and, through him, to the American people.

Discourage Vetting Policies with Career Staff

OPM’s proposed rule would also discourage agency leadership from vetting policy proposals with career staff. OPM argues that career civil servants have institutional experience and subject matter expertise that political appointees may lack and that “[t]heir ability to offer objective analyses and views, without fear of reprisal or loss of employment, contribute to the reasoned consideration of policy options and thus the successful functioning of incoming administrations and our democracy” (88 Fed. Reg. 63862-63863). This argument is partly correct and partly deeply mistaken.

OPM is correct that career employees possess valuable institutional knowledge and perspectives. Political appointees value career staff feedback—even, and in some cases especially, opposition to their policy proposals. This feedback helps political appointees identify blind spots and other policy shortcomings. For example, during the Trump Administration political appointees in the Department of Labor (DOL) considered career staff feedback “Red Teaming.” Very liberal DOL career experts extensively critiqued every substantive draft agency regulation. This enabled political appointees to preemptively address legal or policy shortcomings before the rules went public. DOL leadership believed that—contrary to career employees’ apparent intentions—Red Teaming meaningfully improved agency policymaking (Sherk, 2023, p. 20).

However, OPM’s proposal misses how agency leaders react to career staff resistance. When career employees go beyond critiquing policies to actively undermining them, political appointees often respond by sidelining the career staff. This largely or entirely excludes them—and their input—from the policymaking process. Scholars have lamented this phenomenon in the Trump Administration (Nou, 2019, p. 367). However, it is a rational response that is often necessary to circumvent career staff opposition.

OPM career staff have firsthand experience with this phenomenon. In 2018, the Trump Administration developed a triad of executive orders reforming the civil service. The drafts sought to streamline the federal dismissal process, promote efficient Labor-Management relations, and rein in expensive subsidies to federal unions. These drafts ultimately became Executive Orders 13836, 13837, and 13839. The Trump Administration estimated that, when fully implemented, the orders would save taxpayers at least $100 million annually (Rein, 2018).

The author of this comment served on the White House Domestic Policy Council in the Trump Administration. The author has firsthand knowledge that, late in the drafting process, the drafts were circulated to OPM career staff for their feedback. A career OPM employee promptly leaked the draft orders to federal unions (that employee is likely involved in preparing this proposed rule and has firsthand knowledge of these facts). Federal unions promptly began lobbying the White House to scuttle the orders.

The White House had planned to issue the orders on Tuesday, May 29th or Wednesday, May 30th, after the 2018 Memorial Day long weekend. Instead, seeing as the orders were already drawing interest group opposition, White House senior staff moved the signing up to late afternoon on Friday, May 25th. This drove criticism that the White House was attempting to bury coverage of the orders. The opposite was true. The White House wanted widespread coverage; staff believed the orders showed President Trump keeping his promises. Subsequent polling showed federal employees themselves overwhelmingly supported the reforms (Wagner, 2018). Nonetheless, an OPM career employee did not—and tried to undermine them.

The Trump Administration learned from this experience: OPM career staff were entirely cut out of the development of Schedule F. The White House realized sharing policy proposals with OPM career staff was tantamount to sending them to federal unions and other reform opponents. OPM career staff likely could have provided valuable input during EO 13957’s development. But those benefits did not outweigh the cost of giving hostile special interests advance notice.

If policy-influencing career staff did not enjoy civil service protections, political appointees could simply dismiss employees they knew or strongly suspected leaked deliberative policy documents. Similarly, they could straightforwardly remove employees who slow-walked policies, produced unusable work product, or were otherwise intransigent. This would discourage policy resistance. It would also enable political appointees to trust the remaining career staff with confidential policy proposals, gaining the benefits of their expertise and critiques.

OPM’s proposal would instead reinforce removal restrictions. It would push administrations whose policy preferences differ from the bureaucracy’s to sideline career staff instead of sharing information with them. OPM argues that including career staff in “reasoned consideration of policy options” promotes “the successful functioning of incoming administrations and our democracy.” If OPM believes this, it should not adopt policies that will predictably discourage sharing policy proposals with career staff.

Proposed Rule Encourages Burrowing-In

OPM’s proposed rule would also encourage political appointees to burrow into policy-influencing positions. The ability of a future administration to quickly reinstate Schedule F makes burrowing pointless. Political appointees would know that, regardless of whether they formally completed their probationary period, they would serve at-will under the next administration. Attempts to advance their policy agenda would only lead to their dismissal. Schedule F prevents political appointees from ensconcing themselves in the bureaucracy.

By eliminating this possibility, OPM’s proposed rule makes burrowing in quite valuable. The academic literature discussed in sections I.B and VI.A shows tenured employees significantly influence agency policy. OPM’s proposed rule gives political appointees strong incentives to burrow into agencies so they can continue advancing their preferred policies.

OPM procedures are supposed to prevent improper conversions. However, as the quadrennial controversy over burrowing-in demonstrates, these safeguards are imperfect. Every four years the media, outside observers, and members of Congress raise concerns about appointees improperly burrowed in (Rein & Gearan, 2021; Ollstein & Cassella, 2021; Samuels, 2021; Lee, 2007). The potential reinstatement of EO 13957 reduces incentives to burrow in in the first place. Taking Schedule F’s reissuance off the table would encourage political appointees to seek improper conversions.

Proposed Rule Legally Flawed

In addition to being bad policy, OPM’s proposed rule is legally flawed. The CSRA expressly precludes OPM’s proposed part 752 revisions that give some policy-influencing employees civil service protections. OPM also adopts an implausible construction of the CSRA that—if accepted by the courts—would render civil service protections unconstitutional for many inferior officers. OPM was wise to include a severability clause, as the Office lacks authority to adopt some of these proposals.

Part 752 Revisions Unlawful

Part of OPM’s proposed rule is patently unlawful. OPM proposes modifying its part 752 regulations to give civil service protections to employees transferred into the excepted service or involuntarily moved between excepted service schedules. As OPM explains, this would mean that, under the proposed rule, career employees transferred into a re-issued Schedule F would keep their adverse action protections and MSPB appeals. OPM argues that its proposed changes:

reflect the impact of statutory requirements—namely, that once an employee meets certain conditions, the individual gains certain statutory procedural rights and civil service protections which cannot be taken away from the individual by simply moving the employee’s position into the excepted service, or within the excepted service, as long as the employee continues to occupy the same or similar position (88 Fed. Reg. 63871).

OPM tellingly does not cite any statutory provisions to this effect; none exist. Instead, 5 U.S.C. § 7511(b)(2) categorically exempts policy-influencing positions in the excepted service from chapter 75’s coverage. This directive contains no qualifiers; Congress said nothing about civil service protections following an employee unless they change positions. If an excepted service employee occupies a position appropriately determined to be policy-influencing, then chapter 75 does not apply to them.

OPM’s proposal to switch civil service procedures to attach to a particular person, instead of a particular job classification, would be a major and arbitrary shift. That is not how Congress drafted title 5.

Unsurprisingly, OPM fails to cite any cases holding that employees retain removal restrictions after their positions are held to be policy-influencing. OPM instead cites two cases that deal with an entirely different issue. In McCormick v. Department of the Air Force (2002) the Federal Circuit examined when an individual becomes a non-probationary “employee” under § 7511(a). The MSPB examined the same question in Greene v. Def. Intel. Agency (2005). Neither case considered the § 7511(b) policy-influencing exception or even contains the word “policy.”

Instead, the only cases to examine this question conclude agencies can eliminate restrictions on removing policy-influencing employees who remain in their positions. In Stanley v. Dep’t of Justice (2005) and Stanley v. Gonzalez (2007) the 9th and Federal circuits heard challenges to the removal of two bankruptcy trustees. Chapter 75 covered the employees when they were initially appointed. Attorney General Janet Reno subsequently declared bankruptcy trustees policy-influencing positions and several years later Attorney General John Ashcroft fired them. The courts upheld these dismissals because the trustees now occupied policy-influencing positions; they no longer had MSPB appeal rights.

OPM argues 5 U.S.C. § 7511(c) authorizes its part 752 regulations (88 Fed. Reg. 63870). That section permits OPM to extend chapter 75 to cover positions “excepted from the competitive service by regulation of the Office which [are] not otherwise covered by this subchapter.” However, policy-influencing positions can hardly be considered “not otherwise covered”— § 7511(b)(2) expressly excludes them from MSPB appeals. And even if policy-influencing positions did fall within § 7511(c) authority, that section applies only to positions OPM regulatorily excepts from the competitive service. It does not apply to positions OPM excepts without rulemaking or positions the President excepts. This section provides no support for a blanket extension of chapter 75. At most it could authorize extending adverse action appeals to positions OPM regulatorily excludes from the competitive service.[13] OPM, however, purports to extend chapter 75 to cover all employees who once possessed civil service protections and remain in their positions. This authority exists nowhere in statute.

OPM cannot extend by regulation adverse action appeals to employees denied them by statute. As the Supreme Court explained in Chevron v. Natural Resources Defense Council (1984):

First, always, is the question whether Congress has directly spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.

This is a Chevron Step One question. Congress has spoken directly and unambiguously on chapter 75’s scope. It does not apply to policy-influencing positions in the excepted service. Whether incumbents in the same position previously enjoyed removal protections is immaterial. OPM proposed a severability clause for good reason; the Office knows its proposed rule exceeds statutory authority.

OPM’s Interpretation Renders Many Applications of Chapter 75 Unconstitutional

OPM interprets chapter 75 to prohibit lifting restrictions on removing tenured employees, even if they hold policy-influencing positions (88 Fed. Reg. 63877). In OPM’s view career incumbents retain adverse action protections if their position is subsequently declared policy influencing; the policy-influencing terms refer only to non-career, political positions. As discussed in section IV, OPM’s interpretation fails as a matter of statutory construction. Moreover, OPM does not appear to have considered the implications of its interpretation: accepting this construction would render many inferior officers’ civil service protections unconstitutional.

The federal workforce consists of “Officers of the United States” (who are subject to the U.S. Constitution’s Appointments Clause) and subordinate employees (who are not). Officers are subdivided into “principal officers” authorized to take final actions binding the executive branch and “inferior officers” without such authority. The Supreme Court has held that the Constitution gives the President (and his agency heads) broad authority to dismiss constitutional officers at will, with two exceptions.

The Court has upheld removal protections for principal officers leading “multimember expert agencies that do not wield substantial executive power” and for inferior officers “with limited duties and no policymaking or administrative authority. These exceptions represent “the outermost constitutional limits of permissible congressional restrictions on the President's removal power" (Seila Law v. Consumer Finance Protection Bureau, 2020). The Supreme Court also held in Free Enterprise Fund v. Public Company Accounting Oversight Board (2010) that these removal protections cannot be combined. If an independent agency is led by principal officers with valid removal protections, subordinate inferior officers cannot possess removal protections.

As Justice Breyer’s dissenting opinion in Free Enterprise Fund noted, many senior career officials are inferior officers. Indeed, he explained that “efforts to define [the term] invariably conclude that the term’s sweep is unusually broad.” The Department of Justice holds that any continuing position that wields delegated sovereign authority is a constitutional office (2007). Consequently, federal employees delegated authority to perform statutory functions are generally officers subject to the appointments clause.[14]

Justice Breyer worried that prohibiting multilevel removal protections would jeopardize inferior officers’ civil service protections in independent agencies. Chief Justice Roberts’s majority opinion rejected those concerns because:

[s]enior or policymaking positions in government may be excepted from the competitive service to ensure Presidential control, see 5 U.S.C. §§ 2302(a)(2)(B), 3302, 7511(b)(2) … Nothing in our opinion, therefore, should be read to cast doubt on the use of what is colloquially known as the civil service system within independent agencies (Free Enterprise Fund v. Public Company Accounting Oversight Board, 2010).

Restrictions on removing federal officers are constitutionally unproblematic precisely because the President can waive them. Notably, the Supreme Court cited the exact statutory authorities President Trump invoked to issue EO 13957.

However, OPM states that it “does not interpret chapter 75 as allowing the President … to waive” civil service protections (88 Fed. Reg. 63877). Accepting that construction of the CSRA would give every inferior officer in every tenure-protected independent agency binding multilevel removal protections. They and their agency heads could only be removed for cause and the President could not dispense with either level of protection. Free Enterprise Fund held such multilevel removal protections unconstitutional.

OPM’s construction would also give inferior officers with substantive policymaking or administrative authority binding removal protections—namely, chapter 75 adverse action appeals. Seila Law explains this exceeds “the outermost limits” of congressional authority. OPM’s interpretation would constitutionally invalidate civil service protections for important swathes of the federal bureaucracy. Indeed, OPM’s interpretation could eliminate more civil service protections than EO 13957 would have.

If OPM’s interpretation were correct (and it is not) then many applications of chapter 75 are categorically unconstitutional under current precedents. OPM has adopted an inverted canon of constitutional avoidance. Courts typically strain ambiguous statutory language to avoid unconstitutional results. OPM is contorting unambiguous statutory text to reach an unconstitutional result.

OPM can narrowly apply chapter 75’s policy-influencing exceptions if it chooses. But OPM cannot change the fact those terms encompass career appointees or prevent a future administration from applying them more broadly. “The President can always choose to restrain himself in his dealings with subordinates. He cannot, however, choose to bind his successors by diminishing their powers, nor can he escape responsibility for his choices by pretending that they are not his own” (Free Enterprise Fund v. Public Company Accounting Oversight Board, 2010).


The federal workforce suffers from widely recognized problems with poor performance and misconduct (including policy-resistance), as well as a broken hiring process. EO 13957 and Schedule F enabled the President to effectively address these problems in policy-influencing positions. This proposed rule would ensconce these problems in the federal workforce and exceeds OPM’s statutory authority. It also rejects the founding vision for the merit service.

George William Curtis was the president of the National Civil Service Reform League. He helped draft, and played a major role in passing, the Pendleton Act. Curtis explained that:

[I]t is better to take the risk of occasional injustice from passion and prejudice, which no law or regulation can control, than to seal up incompetency, negligence, insubordination, insolence, and every other mischief in the service, by requiring a virtual trial at law before an unfit or incapable clerk can be removed (Frug, 1976, p. 955).

OPM should heed the wisdom of one of the principal founders of the merit service. The Office should reject this proposed rule in its entirety.


James Sherk
Director, Center for American Freedom
The America First Policy Institute

Works Cited

Arnett v. Kennedy, 416 U.S. 134 (1974).

Babbitt v. Sweet Home Chapter, Communities for a Great Oregon, 515 U.S. 687 (1995).

Brown v. Gardner, 513 U.S. 115 (1994).

Brunton v. United States, 518 F. Supp 233 (S. D. Ohio 1981).

Boyle, D. (2016, Oct. 5). More than 200 veterans died while waiting for treatment at just one VA hospital, damning report reveals. The Daily Mail.

Chevron U.S.A., Inc. v. Natural Resources Defense Council, 467 U.S. 837 (1984).

Coggburn, J. (2006). At-Will Employment in Government: Insights from the State of Texas. Review of Public Personnel Administration, 26(2), 158-177.

Coggburn, J., Battaglio, R. P., Bowman, J.S., Condrey, S. E., Goodman, D., & West, J. (2010). State Government Human Resource Professionals’ Commitment to Employment At Will. The American Review of Public Administration, 40, 189-208.

Condrey, S. & Battaglio, R. P. (2007). A return to Spoils? Revisiting Radical Civil Service Reform in the United States. Public Administration Review, 67(3), 425 - 436.

Corgnet, B., Hernán-González, R., and Rassenti, S. (2015). Firing threats: Incentive effects and impression management. Games and Economic Behavior, 91. 97 – 113.

Cournoyer, C. (2012, June 8). Civil Service Reform: Lessons from Georgia and Indiana. Governing.

Davidson, J. (2017, Feb. 1). Many feds don’t like Trump’s program, but they’re not revolting. The Washington Post.

Department of Justice. (2007). Officers of the United States within the Meaning of the Appointments Clause. Office of Legal Counsel.

Dibble v. Quinn, 793 F. 3d 803 (7th Cir. 2015).

Eilperin, J., Rein, L., & Fisher, M. (2017, Jan. 31). Resistance from within: Federal workers push back against Trump. The Washington Post.

Elrod v. Burns, 427 U.S. 347 (1976).

Executive Order 12300, 46 Fed. Reg. 18683 (March 26, 1981).

Executive Order 12940, 59 Fed. Reg. 61519 (Nov. 30, 1994).

Executive Order 13957, 85 Fed. Reg. 67631 (Oct. 26, 2020).

Executive Order 13836, 83 Fed. Reg. 25329 (June 1, 2018).

Executive Order 13837, 83 Fed. Reg. 25335 (June 1, 2018).

Executive Order 13839, 83 Fed. Reg. 25343 (June 1, 2018).

Executive Order 14003, 86 Fed. Reg. 7231 (January 27, 2021)

Feds for Medical Freedom v. Biden, 63 F. 4th 366 (5th Cir. 2023).

Feinstein, B. D. & Wood, A. K. (2022). Divided agencies. Southern California Law Review, 95(4). 731-784.

Flavelle, C. & Bain, B. (2017, Dec. 18). Washington Bureaucrats are Quietly Working to Undermine Trump’s Agenda. Bloomberg News.

Free Enterprise Fund v. Public Company Accounting Oversight Board, 561 U.S. 477 (2010).

French, E. & Goodman, D. (2011). Assessing the temporary use of at-will employment for reorganization and workforce reduction in Mississippi state government. Review of Public Personnel Administration, 31(3), 270-290.

Frug, G. E. (1976). Does the Constitution Prevent the Discharge of Civil Service Employees? University of Pennsylvania Law Review, 124, 942-1012.

Gailmard, S. & Patty, J. W. (2007). Slackers and zealots: Civil service, policy discretion, and bureaucratic expertise. American Journal of Political Science, 51. 873-889.

Gattis v. Gravett, 806 F.2d 778 (8th Cir.1986).

Graves v. Department of Veterans Affairs. (2016). Merit Systems Protection Board, Central Regional Office. Docket No. CH-0707-16-0180-J-1.

Greene v. Defense Intelligence Agency, 100 M.S.P.R. 447 (2005).

Government Accountability Office. (2015). Improved supervision and better use of probationary periods are needed to address substandard employee performance. (GAO Report No. GAO-15-191).

Government Accountability Office. (2019a). Federal workforce: Key talent management strategies for agencies to better missions. (GAO Report No. GAO-19-181).

Government Accountability Office. (2019b). Human Capital: Improving federal recruiting and hiring efforts. (GAO Report No. GAO-19-696).

Hastings, R. and Meyer, E. (2020). No rules rules: Netflix and the culture of reinvention. New York: Penguin Press.

Ichino, A. & Riphahn, R. T. (2005). The Effect of Employment Protection on Worker Effort: Absenteeism during and after Probation. Journal of the European Economic Association, 3(1). 120-143.

INS v. Cardoza-Fonseca, 480 U.S. 421 (1987).

Johnson, R. N. & Libecap, G. D. (1994). The federal civil service system and the problem of bureaucracy. University of Chicago Press.

Katyal, N. K. (2006). Internal separation of powers: Checking today’s most dangerous branch from within. Yale Law Journal, 115(9). 2314-2349.

Katz, E. (n.d.). Firing line. Government Executive.

Katz, E. (2022, July 22). If Trump is reelected, his aides are planning to purge the civil service. Government Executive.

Kim, J. & Kellough, J. E. (2014). At-will employment in the states: Examining the perceptions of agency personnel directors. Review of Public Personnel Administration, 34(3), 218-236.

Lee, C. (2007, Oct. 5). Political appointees ‘burrowing in.’ The Washington Post.

Mann, T. (2020, Feb. 26). When safety rules on oil drilling were changed, some staff objected. Those notes were cut. The Wall Street Journal.

Martins, P. S. (2009) Dismissals for cause: The difference that just eight paragraphs can make. Journal of Labor Economics, 27(2). 257-279.

McCubbins, M. D., Noll, R. G., & Weingast, B. R. (1987). Administrative procedures as instruments of political control. Journal of Law, Economics, and Organization, 3(2). 243-277.

McCormick v. Department of the Air Force, 307 F. 3d 1339 (Fed. Cir. 2002).

McMurtray v. Holladay, 11 F.3d 499 (5th Cir.1993).

Merit Systems Protection Board. (1995). Removing poor performers in the federal service. Office of Policy and Evaluation.

Merit Systems Protection Board. (2019). Remedying unacceptable employee performance in the federal civil service. Office of Policy and Evaluation.

Merit Systems Protection Board. (2021). Direct-hire authority under 5 U.S.C. § 3304: Usage and outcomes. Research Brief. Office of Policy and Evaluation.

Michaels, J. (2016, Nov. 11). Administrative checks and balances in the Trump Administration. Just Security.

Mills, F. (2010). Civil disservice: Federal employment culture and the challenge of genuine reform. iUniverse.

Morris, S. (2020, Sept. 22). Mail-in Ballot Considerations for NLRB Union Elections. Law360.

National Labor Relations Board. (2020). NLRB Establishes Standards for Mail- and Manual-Ballot Representation Elections during the COVID-19 Pandemic.

Nelson, M. (2021). Records Shed Light on Biden’s Day-One Firing of Trump’s NLRB general counsel. Freedom Foundation.

Nondiscrimination on the Basis of Sex in Education Programs or Activities Receiving Federal Financial Assistance, 85 Fed. Reg. 30026 (May 19, 2020).

Nou, J. (2016). Bureaucratic Resistance from Below. Yale Journal on Regulation, Notice and Comment.

Nou, J. (2019). Civil Servant Disobedience. University of Chicago Law School, Public Law and Legal Theory Working Papers.

Office of Personnel Management. (2016). Federal Employee Viewpoint Survey: Governmentwide Management Report.

Office of Personnel Management. (2021). Federal employee viewpoint survey: government-wide management report.

Office of Personnel Management. (2022a). Federal Employee Viewpoint Survey: Governmentwide Management Report.

Office of Personnel Management. (2022b). Federal Workforce Data.

Oil and Gas and Sulfur Operations in the Outer Continental Shelf-Blowout Preventer Systems and Well Control Revisions, 83 F.R. 22128 (May 11, 2018).

Ollstein, M. A. & Cassella, M. (2021, Jan. 19). The ‘deep state’ of loyalists Trump is leaving behind for Biden. Politico.

Partnership for Public Service. (n.d.). Federal hiring.

Partnership for Public Service. (2020). A Time for Talent: Improving Federal Recruiting and Hiring.

Pew Research Center. (2023, Sept. 19). Public Trust in Government: 1958-2023.

Potter, R. A. (2017a). Slow-Rolling, Fast-Tracking, and the Pace of Bureaucratic Decisions in Rulemaking. Journal of Politics, 79(3). 841-855.

Potter, R. A. (2017b, June 13). The strategic calculus of bureaucratic delay. Midwest Political Science Association.

President’s Committee on Administrative Management. (1937). Report of the President’s Committee.

Ratzlaf v. United States, 510 U.S. 135 (1994).

Rea v. Matteucci, 121 F.3d 483 (9th Cir.1997).

Read, J. (2019, Sept.). Get ready, it’s performance appraisal time. Issues of Merit, 24(3).

Rein, L. (2018, May 25). Trump takes aim at federal bureaucracy with new executive orders rolling back civil service protections. The Washington Post.

Rein, L. & Gearan, A. (2021, Jan. 24). Biden is firing some top Trump holdovers, but in some cases, his hands may be tied. The Washington Post.

Riphahn, R. T. (2004). Employment protection and effort among German employees. Economic Letters, 85(3). 353-357.

Rubens v. Department of Veterans Affairs. (2016). Merit Systems Protection Board, Northeastern Regional Office. Docket No. PH-0707-16-0151-J-1.

Samuels, B. (2021, Mar. 9). Dozens of Trump appointees ‘burrow’ into Biden government. The Hill.

Scoppa, V. (2010). Shirking and employment protection legislation: Evidence from a natural experiment. Economic Letters, 107(2). 276-280.

Scoppa, V. & Vuri, D. (2014). Absenteeism, unemployment and employment protection legislation: evidence from Italy. IZA Journal of Labor Economics, 3(3). 1-25.

Seila Law v. Consumer Finance Protection Bureau, 140 S. Ct. 2183 (2020).

Sherk, J. (2022, Sept. 14). Union Arbitrators overturn Most Federal Employee Dismissals. America First Policy Institute.

Sherk, J. (2023, Sept. 10). Tales From the Swamp: How Federal Bureaucrats Resisted President Trump. America First Policy Institute.

Spenkuch, J. L., Teso, E., & Xu, G. (2023). Ideology and performance in public organizations. Econometrica. 91(4).

Stanley v. Department of Justice, 423 F. 3d 1271 (Fed. Cir. 2005).

Stanley v. Gonzales, 476 F. 3d 653 (9th Cir. 2007).

Students for Fair Admissions, Inc. v. President and Fellows of Harvard College, 600 U.S. ___ (2023).

Swan, J. (2022, July 22). A radical plan for Trump’s second term. Axios.

U.S. Civil Service Commission. (1913). Twenty-Ninth Annual Report of the United States Civil Service Commission: For the Fiscal Year Ended June 30th, 1912.

U.S. Department of Justice, Executive Office for Immigration Review and National Association of Immigration Judges, 71 FLRA 1046 (2020).

U.S. Department of Justice, Executive Office for Immigration Review and National Association of Immigration Judges, 72 FLRA 622 (2022a).

U.S. Department of Justice, Executive Office for Immigration Review and National Association of Immigration Judges, 72 FLRA 733 (2022b).

U.S. Government Policy and Supporting Positions (Plum Book). (2020). Senate Committee on Homeland Security and Governmental Affairs and House Committee on Government Reform.

Van Riper, P. P. (1958). History of the United States Civil Service. Evanston, IL: Row, Peterson, & Co.

Wagner, E. (2018, June 18). Survey: Half of Feds Support White House Attempts to Ease Firing Process. Government Executive.

Wagner, E. (2020, Feb. 26). OPM Announces Adjustments to Annual Time-to-Hire Metrics. Government Executive.

Weiss v. Department of Veterans Affairs, Merit Systems Protection Board, Docket No. NY-0707-16-0149-C-1.

Wood, D. B. (1988). Principals, Bureaucrats, and Responsiveness in Clean Air Enforcements. The American Political Science Review, 82(1). 213-234.

[1] FY 2022 is the most recent year for which OPM has released full-year data on FedScope. In FY 2022 the probationary period was two years for competitive service employees in the Department of Defense (which accounts for over one-third of the non-Postal federal civilian workforce). The probationary period is two years for excepted service employees. FedScope data cubes, maintained by the Office of Personnel Management, show that agencies removed 3,900 permanent full-time non-seasonal employees with at least two years of service for performance or misconduct in FY 2022. This represents about one-quarter of one-percent of the 1.6 million permanent non-seasonal full-time federal employees with at least two years of service employed in the executive branch that year.

[2] During the 2021 presidential transition there were 3,762 executive branch positions available for presidential appointees, non-career members of the Senior Executive Service, and Schedule C political appointees (U.S. Government Policy and Supporting Positions, 2020, p. 212). That year the executive branch employed 2.2 million civilian employees, excluding employees of the U.S. Postal Service (Office of Personnel Management, 2022b).

[3] AFPI is appending a copy of the Sherk (2023) report and all examples of bureaucratic misconduct contained therein to this comment for inclusion in the administrative record for this rulemaking.

[4] OPM FedScope data cubes show agencies made 120,070 new hires into excepted service appointments in FY 2022 and 54,281 through the first half of FY 2023. They made 132,627 new hires into competitive service appointments in FY 2022 and 54,079 through the first half of FY 2023. These figures exclude new hires into the Senior Executive Service.

[5] MSPB reports that 73% of competitive service hires were non-DHA appointments in FY 2018. FedScope data cubes show that excepted service appointments were a bare majority of all new hires in the first six months of FY 2023. If DHA appointments have not fallen since FY 2018, then non-DHA competitive service appointments accounted for 36.5% of all federal appointments in the first half of FY 2023. If DHA appointments have continued to grow the proportion of appointments utilizing competitive hiring procedures will be proportionately lower.

[6] AFPI notes that the sources OPM cites in footnote 186 of the NPRM to justify this proposition exclusively compare the performance of at-will political appointees to tenured career employees. However, the relevant comparison for purposes of analyzing Schedule F is how at-will career officials perform relative to tenured career officials. Career state HR directors with experience with at-will employment widely report it is an essential modern management tool—a tool that this proposed rule would deny federal agencies.

[7] As OPM notes, Congress gave veterans—and only veterans—the ability to appeal removals to the Civil Service Commission in 1944 and passed legislation making the Commission’s decisions binding in 1948.

[8] For example, in Gattis v. Gravett (8th Cir. 1986) the 8th Circuit rejected a challenge by two police majors to their dismissal after Arkansas passed legislation removing majors from the state civil service protections. The court held that the legislative process provided all the process that was due and their former property interest in their jobs had been “extinguished.” See also Pittman v. Chicago Board of Education (7th Cir. 1995) (Illinois law that stripped principles of tenure was not an unconstitutional “taking” of their property and satisfied due process); Rea v. Matteucci, (9th Cir. 1997) (no due process violation where statute reclassified employee from permanent to non-permanent status); McMurtray v. Holladay, (5th Cir. 1993) (no due process violation where statute provided that personnel actions of state agency would be exempt from procedures of merit review system)

[9] Katz (2022) claims that those involved in the efforts to reissue Schedule F have “identified 50,000 current employees that could be dismissed under the new authority” and are “creating lists of names to supplant existing civil servants.” The author of this comment was the “former Trump Administration official” who was the source for that article. The author expressly told Mr. Katz that this characterization was false. The Trump White House arrived at the 50,000 position figure using estimates based on the occupational distribution of the federal workforce. Individual federal positions were not separately identified, much less employees by name. Nor is the author aware of any efforts to create lists of candidates to replace career incumbents. Flatly inaccurate reporting is no basis for a federal rulemaking.

[10] Dismissing burrowed-in political appointees would not be a prohibited personnel practice, and thus not prohibited under section 6 of EO 13957. 5 U.S.C. § 2302 prohibits discriminating against federal employees based on their political affiliation or campaign contributions. However, burrowed-in political appointees would be dismissed based on their history of holding a non-career appointment and transitioning immediately to a career position, not their political affiliation as such.

[11] Feinstein and Wood provided their data to AFPI. The figures cited in this subsection come from analysis of this data; they are not publicly reported in their report. We include as an attachment, and incorporate into the administrative record for this rulemaking, Feinstein and Wood’s (2022) data on CF scores (essentially standard deviations from the mean) for career federal employee and agency head campaign donations between the 1995-1996 and 2013-2014 election cycles.

[12] Feinstein and Wood (2022) have data on civil servant campaign donations from 23 agencies in the 2013-2014 election cycle. Across these agencies the average civil servant CF score ranged from a high of -0.66 (Department of Justice) to a low of -1.33 (Environmental Protection Agency). The median agency was the Department of State, with an average civil servant score of -1.19. These figures put EPA career employees in the most liberal 8 percent of campaign donors that cycle, while Department of Justice employees were in the 25 percent most liberal campaign donors.

[13] Part 752 revisions limited to positions regulatorily excepted by OPM would not prevent a future administration from bringing back Schedule F. It would simply require the President to place positions in Schedule F rather than OPM doing so. This would impose some administrative burdens on the Executive Office of the President but would not substantively impair the President’s ability to make the reforms.

[14] The principal exception is that purely advisory duties do not create offices; developing a report required by Congress does not make a federal employee an officer.

Join The

By providing your information, you become a member of America First Policy Institute and consent to receive emails. By checking the opt in box, you consent to receive recurring SMS/MMS messages. Message and data rates may apply. Message frequency varies. Text STOP to opt-out or HELP for help. SMS opt in will not be sold, rented, or shared. You can view our Privacy Policy and Mobile Terms of Service here.