Merit Hiring In A Merit System

A large national government can hire good people through merit selection.

The government did so for years, but it has proven difficult in recent times to select personnel based on their knowledge, skills, and abilities (KSA) as the law dictates.

Yet for the past 34 years, the U.S. civil service has been unable to distinguish consistently between strong and unqualified applicants for employment. As the Carter presidency was winding down, the U.S. Department of Justice and top lawyers at the OPM contrived with plaintiffs to end civil service IQ exam- inations because of concern about their possible impact on minorities.

The OPM had used the Professional and Administrative Career Examination (PACE) gen- eral intelligence exam to select college graduates for top agency employment, but Carter Administration officials—probably without the President’s informed con- currence—abolished the PACE through a legal consent court decree capitulating to demands by civil rights petitioners who contended that it was discriminatory. The judicial decree was to last only five years but still controls federal hiring and is applied to all KSA tests even today.

General ability tests like the PACE have been used successfully to assess the use- fulness and cost-effectiveness of broad intellectual qualities across many separate occupations. Courts have ruled that even without evidence of overt, intentional discrimination, such results might suggest discrimination.

This doctrine of disparate impact could be ended legislatively or at least narrowed through the regulatory process by a future Administration. In any event, the federal government has been denied the use of a rigorous entry examination for three decades, relying instead on self-evaluations that have forced managers to resort to subterfuge such as preselecting friends or associates that they believe are competent to obtain qual- ified employees.

In 2015, President Barack Obama’s OPM began to introduce an improved merit examination called USAHire, which it had been testing quietly since 2012 in a few agencies for a dozen job descriptions. The tests had multiple-choice questions with only one correct answer. Some questions even required essay replies: questions that would change regularly to depress cheating.

President Donald Trump’s OPM planned to implement such changes but was delayed because of legal concerns over possible disparate impact. Courts have agreed to review the consent decree if the Uniform Guidelines on Employee Selection Procedures setting the technical requirements for sound exams are reformed. A government that is unable to select employees based on KSA-like test qualifications cannot work, and the OPM must move forward on this very basic personnel management obligation.

The Centrality of Performance Appraisal. In the meantime, the OPM must manage the workforce it has. Before they can reward or discipline federal employees, managers must first identify who their top performers are and who is performing less than adequately. In fact, as Ludwig von Mises proved in his classic Bureaucracy,14 unlike the profit-and-loss evaluation tool used in the private sector, government performance measurement depends totally on a functioning appraisal system. If they cannot be identified in the first place within a functioning appraisal system, it is impossible to reward good performance or correct poor performance. The problem is that the collegial atmosphere of a bureaucracy in a multifaceted appraisal system that is open to appeals makes this a very challenging ideal to implement successfully.

The GAO reported more recently that overly high and widely spread performance ratings were again plaguing the government, with more than 99 percent of employees rated fully successful or above by their managers, a mere 0.3 percent rated as minimally successful, and 0.1 percent actually rated unacceptable.15 Why? It is human nature that no one appreciates being told that he or she is less than outstanding in every way. Informing subordinates in a closely knit bureaucracy that they are not performing well is difficult. Rating compatriots is even consid- ered rude and unprofessional. Moreover, managers can be and often are accused of racial or sexual discrimination for a poor rating, and this discourages honesty.

In 2018, President Trump issued Executive Order 1383916 requiring agencies to reduce the time for employees to improve performance before corrective action could be taken; to initiate disciplinary actions against poorly performing employees more expeditiously; to reiterate that agencies are obligated to make employees improve; to reduce the time for employees to respond to allegations of poor performance; to mandate that agencies remind supervisors of expiring employee probationary periods; to prohibit agencies from entering into settlement agreements that modify an employee’s personnel record; and to reevaluate proce- dures for agencies to discipline supervisors who retaliate against whistleblowers. Unfortunately, the order was overturned by the Biden Administration,17 so it will need to be reintroduced in 2025.

The fact remains that meaningfully evaluating employees’ performance is a critical part of a manager’s job. In the Reagan appraisal process, managers were evaluated on how they themselves rated their subordinates. This is critical to

responsibility and improved management. It is essential that political executives build policy goals directly into employee appraisals both for mission success and for employees to know what is expected. Indistinguishable from their coworkers on paper, hard-working federal employees often go unrewarded for their efforts and are often the system’s greatest critics. Federal workers who are performing inadequately get neither the benefit of an honest appraisal nor clear guidance on how to improve. Political executives should take an active role in supervising per- formance appraisals of career staff, not unduly delegate this responsibility to senior career managers, and be willing to reward and support good performers.

Merit Pay

Performance appraisal means little to daily operations if it is not tied directly to real consequences for success as well as failure. According to a survey of major U.S. private companies—which, unlike the federal government, also have a profit-and-loss evaluation—90 percent use a system of merit pay for performance based on some type of appraisal system. Despite early efforts to institute merit pay throughout the federal government, however, compensation is still based primarily on seniority rather than merit.

Merit pay for executives and managers was part of the Carter reforms and was implemented early in the Reagan presidency. Beginning in the summer of 1982, the Reagan OPM entered 18 months of negotiations with House and Senate staff on extending merit pay to the entire workforce. Long and detailed talks between the OPM and both Democrats and Republicans in Congress ensued, and a final agreement was reached in 1983 that supposedly ensured the passage of legislation creating a new Performance Management and Recognition System (PMRS) for all, (not just management) GS-13 through GS-15 employees.

Meanwhile, the OPM issued regulations to expand the role of performance related to pay throughout the entire workforce, but congressional allies of the employee unions, led by Representative Steny Hoyer (D) of government employee– rich Maryland, stoutly resisted this extension of pay-for-performance and, with strong union support, used the congressional appropriations process to block OPM administrative pay reforms. Bonuses for SES career employees survived, but per- formance appraisals became so high and widely distributed that there was little relationship between performance and remuneration.

Ever since the original merit pay system for federal managers (GM-13 through GM-15 grade levels, just below the SES) was allowed to expire in September 1993, little to nothing has been done either to reinstate the federal merit pay program for managers or to distribute performance rating evaluations for the SES, much less to extend the program to the remainder of the workforce. A reform-friendly President and Congress might just provide the opportunity to create a more comprehensive performance plan; in the meantime, however, political executives should use exist- ing pay and especially fiscal awards strategically to reward good performance to the degree allowed by law.

Making the Appeals Process Work. The nonmilitary government dismissal rate is well below 1 percent, and no private-sector industry employee enjoys the job security that a federal employee enjoys. Both safety and justice demand that managers learn to act strategically to hire good and fire poor performers legally. The initial paperwork required to separate poor or abusive performers (when they are infrequently identified) is not overwhelming, and managers might be motivated to act if it were not for the appeals and enforcement processes. Formal appeal in the private sector is mostly a rather simple two-step process, but government unions and associations have been able to convince politicians to support a multiple and extensive appeals and enforcement process.

As noted, there are multiple administrative appeals bodies. The FLRA, OSC, and EEOC have relatively narrow jurisdictions. Claims that an employee’s removal or disciplinary actions violate the terms of a collective bargaining agreement between an agency and a union are handled by the FLRA, employees who claim their removal was the result of discrimination can appeal to the EEOC, and employ- ees who believe their firing was retribution for being a whistleblower can go to the OSC. While the MSPB specializes in abuses of direct merit system issues, it can and does hear and review almost any of the matters heard by the other agencies. Cases involving race, gender, religion, age, pregnancy, disability, or national origin can be appealed to the EEOC or the MSPB—and in some cases to both—and to the OSC. This gives employees multiple opportunities to prove their cases, and while the EEOC, MSPB, FLRA, and OSC may all apply essentially the same burden of proof, the odds of success may be substantially different in each forum. In fact, forum shopping among them for a friendlier venue is a common practice, but fre- quent filers face no consequences for frivolous complaints. As a result, meritorious cases are frequently delayed, denying relief and justice to truly aggrieved individuals. The MSPB can and does handle all such matters, but it faces a backlog of an estimated 3,000 cases of people who were potentially wrongfully terminated or disciplined as far back as 2013. From 2017–2022 the MSPB lacked the quorum required to decide appeals. On the other hand, as of January 2023, the EEOC had a backlog of 42,000 cases.

While federal employees win appeals relatively infrequently—MSPB adminis- trative judges have upheld agency decisions as much as 80 percent of the time—the real problem is the time and paperwork involved in the elaborate process that managers must undergo during appeals. This keeps even the best managers from bringing cases in all but the most egregious cases of poor performance or mis- conduct. As a result, the MSPB, EEOC, FLRA, and OSC likely see very few cases compared to the number of occurrences, and nonperformers continue to be paid and often are placed in nonwork positions. Having a choice of appeals is especially unique to the government. If lower-pri- ority issues were addressed in-house, serious adverse actions would be less subject

to delay. With the proper limitation of labor union actions, the FLRA should have limited reason for appeals. The EEOC’s federal employee section should be transferred to the MSPB, and many of the OCS’s investigatory functions should be returned to the OPM. The MSPB could then become the main reviewer of adverse actions, greatly simplifying the burdensome appeal process.

Making Civil Service Benefits Economically and Administratively Rational. In recent years, the combined wages and benefits of the executive branch civilian workforce totaled $300 billion according to official data. But even that amount does not properly account for billions in unfunded liability for retirement and other government reporting distortions. Official data also report employment as approximately 2 million, but this ignores approximately 20 million contractors who, while not eligible for government pay and benefits, do receive them indirectly through contracting (even if they are less generous). Official data also claim that national government employees are paid less than private-sector employees are paid for similar work, but several more neutral sources demonstrate that pub- lic-sector workers make more on average than their private-sector counterparts. All of this extravagance deserves close scrutiny.