Administrative Values and Public Personnel Management

The Motivations for Reform

The push for civil service reform comes from multiple actors operating from a variety of motives, but because desired reforms typically require changes in civil service laws, political leaders, including elected legislators and executive branch officials, are almost always directly involved. In the U.S. states, for example, governors often lead reform efforts, as was the case with Zell Miller in Georgia, Jeb Bush in Florida, Haley Barbour in Mississippi, Mark Sanford in South Carolina, and Scott Walker in Wisconsin. Very often, political leaders advocating for civil service reform operate from an ideological point of view grounded on conservative principles regarding government and public policy. These leaders typically favor market-based solutions to social problems and are likely to believe that prosperity is best promoted within a society “as a result of freedom of individual initiatives and not as a result of collective action” (de Montricher, 1998, p.122). Accordingly, government regulation and constraints should be reduced whenever possible to produce cost savings and efficiency. With respect to the civil service, reforms designed to decentralize authority for personnel policy, and to thereby reduce the delays that excessive centralized oversight can produce, are seen as keys to increasing efficiency. In addition, reforms have also focused on the introduction of performance-based incentives such as pay for performance (Brewer, 2000, 2001; Kellough & Selden, 1997; Milkovich & Wigdor, 1991; Organisation for Economic Co-operation and Development, 1999; Perry, Engbers, & Jun, 2009) and an expansion of the concept of employment at will (Battaglio, 2010; Battaglio & Condrey, 2009; Brewer & Walker, 2013; Coggburn, 2006; Coggburn et al., 2010; Condrey & Battaglio, 2007; Kim & Kellough, 2014).

While the ideological motivation for reform may be powerful, there usually is a pragmatic political interest at work as well. Of course, the idea that political office holders may be interested in the bureaucracy is not news. In the United States, political patronage was essentially unchallenged as the basis for appointment to civil service positions for the first 94 years of the Republic. Similarly, in the years following the Soviet Union’s collapse, newly independent countries relied extensively on patronage to fill bureaucratic positions. Patronage builds accountability to political authority, and to risk stating the obvious, political leaders in elected positions usually desire a public bureaucracy that is responsive to their policy preferences. Bureaucratic responsiveness to politicians’ agendas helps to ensure that policies and programs consistent with those agendas are implemented faithfully. Responsiveness to political leadership is enhanced as control of the bureaucracy is strengthened. But traditional civil service merit systems, by design, limit political control of employees. To the extent that politicians generally would like to have greater political control over government employees, constraints on political influence imposed by traditional civil service rules must be removed or reduced. As a result, there is a political interest in civil service reforms such as those intended to establish individual performance standards and incentives, reduce oversight by bi-partisan civil service commissions, and reduce employee rights in the workplace (Carroll, 1995). The objective is illustrated well by B. G. Peters and Pierre (2004) in the title of their edited volume on the topic that reads: Politicization of the Civil Service in Comparative Perspective: The Quest for Control.

In an empirical investigation of this subject, McGrath (2013) found that state-level politicians in the United States were willing to act on their preferences for “radical” civil service reform (characterized by the decentralization of the personnel management system) only when their political party held a majority of the seats in the state legislature. He found further that their probability of success in moving a reform forward depended on the size of their majority (McGrath, 2013); that is, reforms were more likely to be enacted when the size of the majority was larger. When the majority was small or thin, members of the majority party were more cautious as they realized that a switch in majority control could mean that their opponents could overturn any reform passed or might even benefit from the reform if they became the majority. Importantly, McGrath (2013) found also that the propensity for reform was tempered by the strength of public employee unions and by the proportion of seats in the legislature held by Democrats. The period of time covered by McGrath’s study was from 1996 to 2005. In short, reform during this period was more likely to occur in states where Republicans were in control politically, where their majority in the state legislature was significant, and where public employee unions were weak. The fact that reform was more probable when Republicans were in control is not surprising as the agenda fits well with a conservative political ideology, as noted above, and conservatives may be interested in reining in public bureaucracies that are perceived as sympathetic to liberal Democratic objectives.

Interestingly, while ideological and political motivations for civil service reform are important bases for action, reform packages are usually not sold to the public using those arguments. More commonly, politicians speak of the need to “modernize” the civil service and make it more efficient. They complain of lethargic public organizations that need revitalizing. Remarks by Governor Zell Miller on reform in the state of Georgia in 1996 are instructive:

Folks, the truth of the matter is that a solution in 1943 [the year the merit system was established in Georgia] is a problem in 1996. The problem is governmental paralysis, because, despite its name our present merit system is not about merit. It offers no reward to good workers. It only provides cover for bad workers. (State of Georgia, 1996, pp. 6-7)

Similarly, the Director of Governor Miller’s Commission on Privatization (and former State Labor Commissioner), Joe Tanner was quoted in the Atlanta Journal and Constitution as saying that the state’s civil service system was “rigid” and “inflexible.” He referred to the system as “antiquated,” but he went further to reveal what seems to be an ideological preference:

We don’t believe that public employees should have any protection that the private employees don’t have . . . Why should government be any different? (“Editorial: Merit in Personnel Reform,” 1995)

One might question how the age of a system should necessarily be an indicator of its quality. Whether a civil service system is “modern” or reflects the most recent fads in organization design should not alone determine its value. A more appropriate test is whether the principles underlying the system are sound and whether it promotes effective and equitable performance of core personnel management functions such as hiring, promotion, compensation, and discipline. Governor Miller’s claim that the Georgia civil service system paralyzed state government in 1996 may have been a rhetorical flourish necessary, at least in part, to forcefully make a point. But there is no doubt that the dramatic reforms in Georgia (decentralization, pay-for-performance, and at-will employment) fit well with his political ideology and his desire to strengthen political control of the state bureaucracy (Nigro & Kellough, 2000).

There are, of course, many non-ideological or non-political reasons for reform. Practical problems often arise and require adjustments to the technology of public personnel management. For example, there may be a proliferation of job classification grades or separate pay systems. Pay grades are intended to reflect meaningfully distinct levels of work deserving of different levels of compensation. As the number of pay grades increases, however, the overlap in pay between adjacent grades increases, given a reasonable range of salaries within each grade and the fact that the system cannot pay salaries significantly more than what the labor market allows. The overlap between grades erodes the distinctiveness and meaning of each separate grade. Notably, prior to the reforms in Georgia, there were approximately 58 separate pay grades in the state’s compensation system—an excessive number by any standard. That number was reduced by approximately one half in the reform advocated by Governor Miller. Similarly, large numbers of different pay systems can unduly restrict management’s ability to reassign or transfer employees. In the 1980s, the Reagan Administration confronted this problem and successfully reduced the number of federal payroll systems from 132 to 53 (Pollitt & Bouckaert, 2011). It is also true that job descriptions may become outdated, that employee training may be ineffectively designed or delivered, and that hiring processes may be characterized by lengthy delays that result in many of the best applicants taking jobs elsewhere or withdrawing from the process. These kinds of problems are real, and reform measures should be implemented to address them.

Nevertheless, in discussions of reform, it is useful to remember that administrative arrangements are never value neutral (Waldo, 1948). Indeed, most reforms are proposed precisely because they advocate or promote particular values, but the value trade-offs implicit in the process are often overlooked: When administrative reforms advance one set of values, they often simultaneously neglect or suppress alternative values. The preference for one value over another is often obscured by the headline objectives of reforms—which have lately centered on reducing costs and increasing the efficiency of the civil service. In addition, reforms are typically presented with a sense of urgency, and reform proponents usually portray the status quo as wholly unsatisfactory. Nonetheless, efforts to alter any set of administrative structures or processes may at times solve some problems while creating others.

Civil Service Reform Reconsidered

Calls for personnel reform, whether referred to as reengineering, reinventing, or revitalization, have stressed market-driven initiatives and the enhancement of managerial authority. As we have noted, the reforms are often a reaction to traditional merit-based civil service systems, which are designed to remove partisanship, favoritism, and patronage from the public service. Traditional systems rest on core principles including selection on the basis of open and competitive examinations, political neutrality, and relative security of tenure. These principles are enforced through the promulgation of rules designed specifically to limit the discretion of managers, supervisors, and leaders of public agencies. The establishment and enforcement of such rules necessarily requires the presence of a central authority. In the United States, the original organization given that authority at the federal level was the U.S. Civil Service Commission (CSC), which was replaced by the U.S. Office of Personnel Management (OPM) following implementation of the Civil Service Reform Act of 1978. ¹ Central personnel agencies establish civil service regulations and review agency practices to enforce compliance. The unambiguous purpose is to restrict the actions of political executives and line managers.

But civil service system rules and restrictions can be excessive. The structures imposed are bureaucratic, slow, and cumbersome. The limitations on managerial discretion and the additional layers of personnel system red tape can result in time delays and inefficiencies in essential public personnel administration processes involving employee recruitment, selection, management, and retention. Political leaders who seek reform are interested in finding ways to correct what they see as the excesses of traditional centralized personnel systems while “letting managers manage” and enhancing political control of the bureaucracy. More importantly, these reforms often change “who gets what, when and how” in government in line with the reformers’ ideological preferences (Lasswell, 1936).

Many of the reforms proposed as part of the new public management/reinventing government movement borrowed ideas from the concept of “reengineering” that had previously been popular in the private sector (Champy & Hammer, 1993; Obolensky, 1994; T. J. Peters & Waterman, 1984). The underlying premise was that organizational efficiency and effectiveness can be significantly improved by streamlining organizational procedures, empowering managers to make more decisions at their levels in the hierarchy, and focusing on results rather than process. The manifestations of reform took shape in a number of ways, but as we have argued, much attention was paid to the implementation of pay for performance programs, the decentralization of personnel authority, and the elimination or reduction of merit system protections for public employees. These reforms are examined in more detail below.

Pay for Performance in the Public Sector

The logic of pay for performance is compelling. Typically, it simply requires that pay increases be associated with measured performance. When performance is high, higher pay increases should be granted; when performance is lower, pay increases should be lower or perhaps, if performance is unsatisfactory, there should be no pay increase at all. Essentially, the argument for this approach is that it will serve to motivate marginal performers to improve their productivity while good performers gain satisfaction from the knowledge that their efforts are recognized and rewarded. The most commonly implemented approach to pay for performance, known simply as merit pay, requires that individual employee performance (which is usually based on a supervisory rating) provide the basis for adjustments to individual pay. ²

In the United States, the federal government began an experiment with merit pay following passage of the Civil Service Reform Act of 1978. The program targeted mid-level federal managers (i.e., managers in General Schedule and Equivalent Grades 13-15). Studies of the impact of that effort found, unfortunately, that employees covered by the plan typically saw no credible link between pay and performance (Milkovich & Wigdor, 1991; Pearce, 1989; Perry, 1988-1989; Perry, Petrakis, & Miller, 1989; Siegel, 1987; U.S. General Accounting Office, 1984). Because the system was required to be budget-neutral in each agency (i.e., overall budget allotments for federal salaries were not allowed to increase simply because of the program), any large increases for selected employees had to be offset by smaller or no increases for others. This aspect of the program, coupled with the fact that supervisors often struggle with performance appraisal and find it difficult to draw sharp distinctions between employees, meant that pay increases under the system were usually not significantly different from what had occurred earlier. As a consequence, the motivational potential of the program was undermined (Brewer, Selden, & Facer, 2000; Perry, 1986). Studies of employee perceptions of the program found that many workers in the merit pay system believed that factors unrelated to performance were influencing ratings and subsequent pay decisions (Brewer & Walker, 2013; Milkovich & Wigdor, 1991).

Another example of merit pay was the program implemented in the state of Georgia following the 1996 reforms. This system was carefully planned and included a new performance appraisal process grounded on job-related performance standards along with written performance goals and expectations and new job descriptions (Kellough & Nigro, 2002). Nevertheless, a survey of employees conducted 5 years after the system was implemented showed that state workers had low levels of trust in the program and little confidence in its fairness. Employees and their supervisors both reported that it was not the best way to motivate state workers (Kellough & Nigro, 2002).

It is troubling to see the substantial effort expended on the development and implementation of these merit pay incentive systems when they produce such discouraging results. Notably, employee criticisms of merit pay often focus on perceived problems with performance appraisal processes. Of course, performance appraisal is the foundation of pay for performance, and if the incentive system is to be effective, the appraisal process must be viewed by workers as reasonable and fair. That goal is, however, far more difficult to achieve than one might expect (Kellough, 2012).

Annual performance reviews can be traced to the Hawthorne studies in the Western Electric factory near Chicago in the 1930s, when researchers Elton Mayo and Fritz Roethlisberger discovered that social structure is an important element in the workplace and suggested that supervisors could manage people more effectively by having formal meetings with them. In 1950, the U.S. government institutionalized this idea with the Performance Rating Act, which required annual performance reviews for federal employees. These reviews were later coupled with bonuses and salary increases. Despite nagging problems through the years, one compelling motive for continuing the practice in government was that successful businesses were doing it. But a quick scan of the literature reveals that many businesses were experiencing similar problems with the practice, and some high profile companies have since abandoned it, including General Electric, Microsoft, Accenture, Adobe, Google, and Yahoo (Mosbergen, 2015).

Nevertheless, if formal performance appraisal is utilized, valid criteria and standards for performance on specific jobs must be identified and used. Systematic job analysis and accurate job descriptions can provide an indispensable aid for selecting these criteria. One approach is to identify work outputs that are expected on a job. Such outputs could be identifiable products or services for which an employee may be held responsible. But it can be difficult to identify tangible outputs for many jobs. For example, consider the job of an office receptionist, or a budget analyst, or even a police officer. What are the specific work outputs they produce? Can we identify outputs (i.e., specific products or services) that reflect the full scope of their jobs? How do we handle situations where an individual’s accomplishments are dependent upon the work of others? What outputs do we assess for individuals whose job duties vary widely from day to day? How can we encourage employees to be creative and innovative in their work? In addition, even if we can adequately identify outputs for jobs, how do we specify standards for performance on those outputs? How should decisions about standards for performance be made? While these difficulties may not be insurmountable, they illustrate the complexity of performance assessment.

If work outputs are not easily defined, we often rely instead on assessments of employee behavior or even judgments of employee traits that are presumed to be associated with job performance. Employees may be assessed, for example, on the timeliness with which they submit reports or the quality of their interactions with the public. Judgments may be made about traits such as honesty, initiative, or diligence. The process invariably contains large elements of subjectivity. Of course, there is nothing wrong with asking supervisors to exercise judgment, but the fact that the process may not be built upon objective criteria means that disagreements can easily emerge with employees who are dissatisfied with the result and the pay raise that it produces. Dissatisfaction with the outcomes of subjective performance appraisals can limit or undercut the motivational potential of pay for performance and erode workplace morale. In this way, perceptions can be more important than reality.

An additional problem can arise from what may be called the dynamics of the appraisal process. The act of handing a subordinate an unsatisfactory or weak performance rating can be unpleasant for many supervisors, and obviously, documentation is needed to justify lower ratings. As a consequence, supervisors may find it easier to inflate performance ratings. Supervisors who already believe they have more than enough to do on their jobs can avoid extra work and possible confrontations with unhappy employees who receive poor ratings by simply adjusting their ratings upward. Research has long ago demonstrated that in some federal agencies, nearly every employee was rated “fully successful” or higher (Perry et al., 1989). As a consequence, little variation in the performance rating scores of employees translates into little variation in pay raises. Most employees are likely to receive amounts close to what they would have received if percentage raises were distributed equally across the board.

Studies of merit pay systems typically fail to uncover any evidence that the reform significantly increases employee motivation or productivity (Bowman, 2010; Haga, Richman, & Leavitt, 2010; Jordan & Battaglio, 2014; Kellough & Lu, 1993; Rusaw, 2009). In fact, some studies report that pay for performance actually has negative effects in that it erodes trust between employees and their supervisors, promotes dysfunctional competition, and heightens levels of employee alienation from their organizations (Langbein, 2010; Pearce & Perry, 1983; Perry et al., 1989; Weibel, Rost, & Osterloh, 2010). Still, the concept of merit pay resonates well politically, especially when the public has a skeptical view of the efficiency of government operations and expects strict accountability for public employees. Under these circumstances, the reform becomes a potent symbol of politicians’ efforts to ensure that public employees are working hard and earning their keep. This may help to explain why political leaders embrace the concept without full consideration of what it might actually accomplish. In other words, instituting a pay-for-performance system has powerful symbolic meaning, and politicians may extract political value or advantage from their efforts to implement it—regardless of whether any substantive performance improvement is realized. Importantly, this view comports with our underlying theme: Civil service reformers often have ulterior motives for enacting reforms that promise greater economy and efficiency.

The Decentralization of Personnel Management Authority

A second theme popular among proponents of civil service reform initiatives in recent years is the idea that authority over public personnel management policy should be decentralized to the fullest extent possible. As noted earlier, in traditional civil service systems, the major responsibility for personnel policy rests in a central personnel agency. That agency retains responsibility for examining the qualifications of applicants, analyzing and classifying jobs, structuring the performance appraisal process, and determining pay and benefits. A major purpose of centralizing these functions is to ensure that all applicants and employees are treated equitably in the employment process. Nevertheless, the centralization of authority can rob line managers of needed flexibility in dealing with problems.

The past 35 years have seen significant experiments in the decentralization of virtually all functions of public personnel administration to strengthen managerial flexibility and allow line mangers to exercise more discretion (Brewer, 2000; Coggburn, 2001; Elling & Thompson, 2006; Feeney & Rainey, 2010; Jordan & Battaglio, 2014; Kettl, 1997; Tessema, Soeters, & Ngoma, 2009). Advocates for this reform argue that it will improve the government’s ability to recruit, select, manage, and retain employees. But with decentralization, there is a potential trade-off between the efficiency it can produce and the erosion of broader safeguards to promote consistency and equity in the treatment of applicants and employees across different units of government. In 2002, Jonathan West (2002) warned of the subordination of “traditional public sector values of employee rights, fairness, and equity” as a result of dramatic civil service reforms (p. 80). In general, as agency and managerial discretion increases and there is an absence of centralized oversight, opportunities for inconsistent or egregious personnel decisions will also increase. As non-merit factors increasingly influence hiring, firing, promotion, and pay decisions, traditional civil service values of neutral competence could be subrogated. Once again, these practices could raise the specter of impropriety, whether it exists or not.

Possible difficulties associated with greater decentralization and deregulation can be illustrated by considering proposals for the reform of position classification systems and procedures. Indeed, position classification (i.e., the system for placing jobs into pay categories) has been one of the personnel management processes most frequently targeted for reform. Position classification involves the evaluation of positions (or jobs) based on an analysis of job content and requirements, and the subsequent placement of jobs (and employees) into classes or grades that in turn determine their compensation levels. The standard criticism of these systems is that they are too rigid and hierarchical. Managers, it is argued, need more discretion to classify jobs as they see fit and to reclassify jobs quickly when necessary.

Oftentimes, a need for reclassification arises when a manager recognizes that an employee’s job has changed in important ways. Perhaps job responsibilities have been added or technology associated with performing the job has changed. At other times, pressure for reclassification arises when an employee has reached the top of a pay category and significant pay increases are not possible unless the position is placed in a higher class or pay grade. In both cases, it is argued that managers who would like to reclassify the position are often frustrated by personnel analysts who may not agree that the position has changed sufficiently to warrant reclassification. Certainly, personnel analysts who seek to protect the integrity of the classification system would not favor reclassifying a position simply to help a supervisor increase an employee’s pay.

A proposed reform of traditional classification systems aimed at giving line managers greater flexibility is known as “broad-banding.” This is, of course, an approach in which several pay grades are collapsed into a smaller number of broader grades or pay bands. As a result, a smaller set of wide pay classes is established containing a correspondingly wider range of jobs and salaries. Managers may then exercise greater discretion in determining compensation levels for positions within each pay band, provided, of course, that total salary levels remain within their overall budget limitations. This reform essentially supplants traditional approaches to classification in which rank is determined by position with a modified rank-in-person scheme within the various pay bands.

One potential problem with this reform, and certain other innovations that are designed to decentralize authority over personnel policy and increase managerial flexibility, is that differences in how employees are compensated may result from factors other than meaningful and substantive differences in their work. Within each of the larger pay bands established under a broad-banding scheme, there is a larger array of different jobs than would have existed previously in more narrowly defined pay grades. Presumably, the exercise of managerial flexibility under such a system will result in some jobs, which had similar pay levels in the past, being treated differently in the future. In addition, jobs that had been treated differently in the past might, under a broad-banding system, be treated similarly in the future. Indeed, this is the objective of broad-banding. But either of these situations can be perceived as inequitable and thus problematic. Unless management is extraordinarily careful, there is a significant risk of causing employee alienation and producing lower levels of employee job satisfaction and organizational commitment—an outcome that generally needs to be avoided. In the United States, federal experiments with broad-banding, conducted under authority provided by the Civil Service Reform Act of 1978, did not produce encouraging results. When employees working within a broad-banding system were asked if they perceived that individual differences in pay reflected real differences in levels of responsibility and job difficulty, only one fifth to one third agreed—depending on the experimental site surveyed (Schay, Simons, Guerra, & Caldwell, 1992). To the extent that employee perceptions of fairness and equity are important concerns, reform of the classification system should be approached cautiously. Even though traditional civil service systems are also ridden with problems, many employees seem to prefer them over radically different systems that are likely to be equally problematic, invoking the familiar anecdote “better the devil you know than the one you don’t.”

In the United States, the decentralization of personnel authority can also raise important legal questions. The Constitutional guarantee of equal protection of the law limits governmental authority to draw distinctions between people—including public employees. There is an extensive amount of jurisprudence on this issue that has evolved to the point where distinctions for which there is a “rational” basis are not usually questioned. Of course, the definition of what is “rational” ultimately falls to the courts, but it may well be that some distinctions drawn between government employees (e.g., different levels of pay for people doing essentially the same jobs in different agencies) have no rational justification. When individual agencies or bureaus are given authority to determine employee compensation levels, significant pay disparities between similar jobs across different agencies could result. Similar problems could arise in situations involving disciplinary actions that differ across agencies when the purported offenses are similar. Furthermore, if these kinds of discrepancies occur along racial or ethnic lines, the government will face a significant burden of demonstrating that the disparities serve compelling governmental interests and are narrowly tailored to meet those interests when such discrepancies are challenged on equal protection grounds. Claims of discrimination under Title VII of the 1964 Civil Rights Act could also arise if patterns of different treatment follow racial, ethnic, or gender differences. Whether such claims are upheld in the courts or not, the process of defending against them can be expensive and exhaustive. An awareness of the implications of equal protection and non-discrimination provisions may well point to a need for applying consistent standards across agencies, and thus, some notable degree of centralized authority over personnel policy.

The Erosion of Merit System Protections

Traditional merit systems for public employment are intended to ensure the integrity and effectiveness of the public service by structuring selection procedures so that jobs go to qualified applicants, and by helping to ensure that employees are protected from unnecessary political interference or abuse. Since the late 19th century, protection from wrongful discharge has been a central element of civil service systems in the United States and elsewhere. Typically, when such rules are in place, employees who have successfully completed a probationary period may only be dismissed for just cause. Such cause could include unsatisfactory performance, insubordination, or other forms of malfeasance in office; and in these situations, the burden is on the public employer to document evidence of wrongdoing, to notify the employee of the charges, and to give the employee an opportunity to refute the charges—often before an impartial arbiter. In essence, these procedures work to ensure that employees are treated fairly by their employers, but they do make the removal of unproductive employees more difficult than it would be otherwise. In addition, the legal shroud of protection covering employees may discourage managers from tackling many cases where disciplinary action or removal may be needed for the good of the service.

In the U.S. context, an important point to bear in mind when considering these kinds of merit system protections is that once a government jurisdiction chooses to establish a policy under which employees will only be terminated for just cause, the kind of procedures that must be followed are not merely a matter of federal, state, or local regulation; ultimately, the procedures must conform to U.S. Constitutional requirements and court interpretations of those requirements. When government promises that adverse action will be taken only for just cause, an employee may then be said to have a reasonable expectation of continued employment provided that his or her performance is satisfactory. From a legal perspective, such an expectation creates for the employee a property interest in the job. Once that interest is established, the due process provisions of the Fifth and Fourteenth Amendments come into play. Specifically, the Constitution prohibits government from depriving an individual of life, liberty, or property without due process of law, and the courts have generally determined the types of procedures that are necessary to meet due process obligations. In an influential case decided in 1985, the U.S. Supreme Court indicated that minimal due process requires that employees be notified of pending adverse actions and have the opportunity to hear and respond to allegations against them prior to their termination (Cleveland v. Loudermill, 1985). These requirements are designed to prevent the mistreatment of employees. The Court noted that government may not create a property interest for employees and then determine unilaterally the procedures necessary to meet the requirements of due process. Rather, such requirements emanate from the U.S. Constitution and court interpretations.

Of course, a state or other government entity need not create a property interest in the first place, or if it has been created, the appropriate government authority may rescind it. This latter option is what civil service reform advocates are often seeking when they promote the withdrawal or reduction of merit system protections. To date, the two major jurisdictions that have done so on a wide scale are the states of Georgia and Florida. Georgia abolished merit system protections for all employees hired or promoted by the state after July 1, 1996 (Nigro & Kellough, 2006). Essentially, this reform was an attempt to shift public policy toward at-will employment, and today, the vast majority of the state’s employees are at-will workers (Kim & Kellough, 2014). In Florida, the property interest in employment was removed for all supervisory or managerial employees in 2000 (Bowman, West, & Gertz, 2006). In essence, these public servants, and those covered by the reform in Georgia, are no longer promised that dismissal will be for just cause only. Indeed, they can be dismissed without a reason being specified. Numerous other states have also expanded at-will employment, but on a scale less sweeping than what was accomplished in Georgia and Florida (see, for example, Battaglio, 2010; Battaglio & Condrey, 2009; Coggburn, 2006; Coggburn et al., 2010; Goodman & Mann, 2010; Kim & Kellough, 2014).

Reformers should recognize, however, that the abolition of property interests in employment might not be simply a matter of legislative change in merit system provisions. A property interest may be established also by the existence of an implied contract that could be grounded in an employer’s oral representations or written job offers that promise no adverse actions without just cause, or personnel policies or employment handbooks issued by the employer that make similar representations (Markowitz, 1995). Notably, employers may tend to characterize their systems as fair and equitable—in other words, prospective employees may be told that they have nothing to worry about so long as they do their jobs. Yet it might be wise for such employers to keep in mind another popular anecdote: “be careful what you wish for.” Such provisions may be viewed by a court as constituting a contractual agreement between the employer and the employee and, in turn, such an agreement could be sufficient to establish a property interest for the employee. Once that interest is created, due process requirements must be followed in any termination procedure regardless of the language of merit system statutes. Consequently, those who favor eliminating merit system protections must be certain that actions taken by innumerable managers and agencies do not inadvertently create an implied contract.

Beyond this technical or instrumental difficulty in accomplishing reform of this nature, there are other potential problems. Important questions can be raised as to whether such change is really desirable in the long run. In addition, efforts to undercut employee rights could erode the motivational bases for public service. Employee protections and due process requirements exist, as noted previously, to ensure that employees are treated fairly in their interactions with management and are protected from unwarranted political intrusion. This is the primary purpose of procedural safeguards in employee discipline and termination processes, and a considerable body of research has demonstrated that positive benefits accrue to organizations where employees perceive they are treated fairly and equitably. Indeed, several years ago, Rainey (1997) pointed out that a number of organizational behavior researchers contend that “well developed procedures for justice in organizations contribute to productive management” (p. 245). Rainey (1997) also pointed out that various scholars have found that employees who perceive greater levels of procedural justice in their organizations also exhibit increased levels of trust in management and higher levels of job satisfaction (e.g., see Alexander & Ruderman, 1987; Greenberg, 1996). Some argue that this parity is an element of the psychological contract between public workers, their organizations, and the government. It should be noted that many also believe government should aspire to be a model employer, setting high standards of fairness and justice worthy of emulation by other, nongovernmental organizations. Research has shown that higher levels of job satisfaction are associated with positive behaviors such as lower absenteeism and decreased turnover (Rainey 1997). Alternatively, there is also evidence to suggest that when employees believe they are treated unjustly or inequitably, they are less productive. Adams’s (1965) equity theory posits that when employees feel they are treated unfairly, they may throttle down their levels of output to compensate for the perceived inequity. Of course, it is conceivable that organizational justice could be promoted in agencies where formal procedures for employee protection have been significantly reduced or eliminated, but in such cases, justice would depend only on the idiosyncratic behavior of numerous individual managers, supervisors, and agency executives. Certainly the goal of standardization would fall victim to these reforms and the perception of fairness and justice might suffer. Whether organizational justice or, perhaps more importantly, employee perceptions of justice will exist within public organizations under those circumstances is an open question.

Conclusion

Civil service reform has been a major item on the government agenda for quite some time. Many of the most common reforms have several things in common: They tend to target the bureaucracy when the problems may be more systemic in government, they often require the implementation of pay-for-performance systems, and they aim to weaken or dismantle traditional civil service systems by deregulating personnel policy and eliminating or reducing merit system protections afforded to employees. The stated objective of such reforms is to improve organizational performance and productivity by streamlining personnel procedures, but oftentimes, these reforms are proposed with other goals in mind as well. For example, elected officials may be seeking to augment their authority, or they may be pursuing these reforms for ideological or symbolic reasons. Pay for performance, for instance, appears to be a logical and compelling idea at first glance, but closer inspection suggests that it may not be an effective mechanism for motivating employee performance. Similarly, arguments can be made for empowering line managers and giving them more discretion over personnel functions such as job classification. Many proposed reforms accomplish that objective, but along with the increased managerial discretion, there will be a greater opportunity for inequity and an erosion of the integrity of various personnel management processes. Even under the best of circumstances, it is not clear that a broad-banding system for job classification will make public organizations more productive, for example. One might also argue that merit system protections for employees are excessive. Such protections do significantly constrain managerial prerogatives, but again, when managerial discretion is increased, care must be taken to ensure that it is not used irresponsibly. As Woodrow Wilson (1887) argued nearly 130 years ago, “There is no danger in power, if only it be not irresponsible” (p.213). Entrusting more discretion in public managers suggests that we must also have mechanisms to ensure that they use their discretion responsibly. Clearly under a system of reduced employee protections, there is a significantly heightened need for supervisors and managers to act ethically. The problem is finding the right balance between the need for public managers to have the flexibility necessary to guide their organizations effectively and the simultaneous need to ensure fairness and equity in the treatment of job applicants and employees.

In sum, we promote more careful and comprehensive consideration of civil service reforms in government. The difficulty is in finding a way to accomplish that goal. Public administration is infused with politics, and civil service reform is, at its core, a political process. Nevertheless, we urge that reform not be grounded solely upon the dictates of ideology or the demand for enhanced political control of the bureaucracy. Reform should also be informed by data and knowledge developed through the application of systematic and non-partisan analysis. We urge proponents of “good government” across the political spectrum to heed this suggestion. Researchers should continue to examine both the intended and unintended consequences of reforms, and report on the broader implications of their findings. Such findings will contribute to the expanding knowledge base of Public Administration and provide further impetus for an evidence-based approach to reform. Reformers should avoid faddish, fever-driven reforms and opt for more carefully considered ones. In a pluralistic political system, all political actors should work tirelessly to ensure the honesty and integrity of all others. Civil service reformers should be held to a high standard in this regard. This change will, of course, result in a less radical but ultimately more effective civil service reform agenda. We firmly believe that this approach will best serve the long-term interests of politicians, public employees, and citizens.