Roles of Agencies and Supervisors in Effective Pay for Performance Systems, Lecture notes of Design

Download Roles of Agencies and Supervisors in Effective Pay for Performance Systems and more Lecture notes Design in PDF only on Docsity! Designing an Effective Pay for Performance Compensation System A Report to the President and the Congress of the United States by the U.S. Merit Systems Protection Board THE CHAIRMAN U.S. MERIT SYSTEMS PROTECTION BOARD 1615 M Street, NW Washington, DC 20419-0001 January 2006 The President President of the Senate Speaker of the House of Representatives Dear Sirs: In accordance with the requirements of 5 U.S.C. 1204(a)(3), it is my honor to submit this Merit Systems Protection Board report, “Designing an Effective Pay for Performance Compensation System.” Federal Government agencies are moving to better align pay with performance and create organizational cultures that emphasize performance rather than tenure. From our research, we have learned that agencies must invest time, money, and effort in the design and implementation of their pay for performance compensation systems in order to succeed. A credible and fair pay for performance system will require an effective performance evaluation system and supervisors who are able and willing to use it properly. Agencies will also need mechanisms such as training and systematic monitoring of pay decisions and outcomes to ensure that pay for performance systems operate as intended. Although the requirements listed above are universal, we believe the long-standing principles of providing “equal pay…for work of equal value” and “appropriate incentives and recognition…for excellence in performance” are best met by agencies designing pay for performance systems to suit their individual missions, workforces, and circumstances with respect to uniform guidelines and principles. Accordingly, this report discusses the critical choices that agency leaders will make during the design and implementation of a pay for performance compensation system. This discussion is intended to help agency leaders better understand how they can adapt pay for performance systems to their organizations and to help them choose wisely among alternatives for measuring and rewarding performance. I believe you will find this report useful as you consider the implementation of pay for performance across the Federal Government. Respectfully, Neil A. G. McPhie Table of Contents Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ix Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ix Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Conclusions and recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xi Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 A summary of pay for performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Benefits and risks associated with pay for performance . . . . . . . . . . . . . . . . . . . 2 Pay for Performance Decision Points . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 Is the agency ready for pay for performance? . . . . . . . . . . . . . . . . . . . . . . . . . . 3 What are the goals of pay for performance? . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Who should be paid for performance? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 What should be the timing for implementing pay for performance? . . . . . . . . . 8 What should be rewarded? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 How should employees be rewarded? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 How much pay should be contingent upon performance? . . . . . . . . . . . . . . . 18 How should performance-based pay be funded? . . . . . . . . . . . . . . . . . . . . . . . 20 How can costs be managed? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Who makes performance rating and pay decisions? . . . . . . . . . . . . . . . . . . . . 25 Who provides input to performance ratings? . . . . . . . . . . . . . . . . . . . . . . . . . 26 How can agencies facilitate pay system integrity? . . . . . . . . . . . . . . . . . . . . . . 28 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Conclusions and Recommendations. . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 A Report by the U.S. Merit Systems Protection Board vii ExECUTIvE SUMMARy Decisions that are made during the design and implementation of a pay for performance system are crucial. Therefore, agency decision makers should carefully consider their design options with full awareness of potential advantages and disadvantages. To assist agency leaders with these crucial decisions, we have prepared this report to address topics such as who should be covered, what should be rewarded, how to reward employees, and suggestions for preserving the integrity of the pay system. We hope that this discussion will assist agencies in anticipating important issues and tailoring their pay systems to their unique needs, which will facilitate their success. Background In the past, individual Federal Government agencies have obtained approval to introduce pay for performance systems in limited demonstration projects or in a few cases on an agency by agency basis. However, the transition of the Federal Government from the traditional General Schedule accelerated when the Department of Defense and the Department of Homeland Security obtained approval to implement performance-based pay systems. To further facilitate this shift from recognizing tenure to focusing on performance, legislation has been proposed to implement pay for performance across the Federal Government.1 However, moving from a pay system that rewards tenure to a pay system that emphasizes and rewards performance, will require more than legislation. Performance-based pay systems present unique opportunities and challenges, and the transition from tenure-based pay systems will be neither quick nor easy. With that in mind, we have prepared this report to support agencies who are planning to use (or are already using) pay for performance. Specifically, we discuss how agencies can design, implement, and operate a pay for performance compensation strategy. Past experience with the General Schedule suggests that in pay systems, “one size does not fit all.” Agencies need to customize compensation systems to their own unique circumstances. Based on this premise, readers should understand that although this report is presented as a “how to” guide, it is not our intent to prescribe a single solution. Instead, our goal is to help agency leaders anticipate and better understand some of the most critical decision points they are likely to face. 1 Working for America Act (draft proposal) as of February 2006. A Report by the U.S. Merit Systems Protection Board ix xii Designing an Effective Pay for Performance Compensation System Executive Summary believe that the organization will actually recognize and reward that performance. Those conditions are not likely to be achieved unless an agency meets certain requirements. These requirements include: 1. A culture that supports pay for performance; 2. Effective and fair supervisors; 3. A rigorous performance evaluation system; 4. Adequate funding; 5. A system of checks and balances to ensure fairness; 6. Appropriate training for supervisors and employees; and 7. Ongoing system evaluation. Research and experience indicate that agencies need to attempt to meet all of these requirements. For example, well-designed performance measures are not a satisfactory substitute for trained, conscientious supervisors; similarly, a high level of funding does not compensate for a lack of checks and balances. To make pay for performance successful, agencies need to make a substantial investment of time, money, and effort. To meet the requirements listed above, agencies will need to make investments that extend far beyond the money needed to fund bonuses and pay increases. For example, supervisors will need: training in designing performance measures and providing performance feedback; a performance evaluation system that enables them to accurately distinguish among levels of performance; and guidelines for determining pay increases or performance bonuses. The introduction of a pay for performance system can also identify weaknesses in other areas—such as training, communications, and employee relations—that should be addressed. Performance evaluation serves as the foundation of a pay for performance system. A pay for performance system links an employee’s pay to some measure of individual and/or pay organizational performance, usually through a formal performance appraisal. Consequently, performance standards and measures—and the application of those standards and measures—matter greatly to both the agency and the employee. Agencies should therefore ensure that (1) performance goals and measures are relevant, reasonable, and usable; (2) employees understand and participate in the performance evaluation process; and (3) performance is evaluated fairly and rigorously. Agencies considering a pay for performance system should be prepared to devote considerable effort to performance evaluation. One frequently cited challenge is that performance is often difficult to define and measure in public sector organizations. Even so, it is critical that this effort be made. A Report by the U.S. Merit Systems Protection Board xiii Executive Summary Agencies should select supervisors based on their supervisory potential, develop and manage them to function as supervisors rather than technicians or staff experts, and evaluate and pay them based on their performance as supervisors. Performance evaluation systems do not evaluate performance; supervisors do. Although agencies should take steps to define and measure performance as precisely and reliably as possible, they should recognize that it is not possible to predetermine or quantify every important aspect of employee performance. Therefore, agencies need to allow some room for supervisors to use discretion and judgment when evaluating employee performance—which is to say, some element of subjectivity. But they should also provide guidelines and training to equip supervisors to exercise that judgment responsibly because trust between supervisors and employees is critical to success. Supervisors also perform other tasks that directly affect employee performance and how employees will fare under a pay for performance system. For example, supervisors assign work; they translate organizational goals into work unit and individual performance goals; they identify training needs and provide access to training opportunities; and they provide coaching and feedback. Consequently, any agency considering pay for performance needs to ensure that its supervisors are willing and able to perform supervisory duties and should hold them accountable for how they discharge their responsibilities. Communication, training, and transparency are essential elements of a good pay for performance system. Pay for performance systems are most effective when they encourage employees to manage and improve their own performance—when they help employees to understand what is expected of them, to choose wisely among various courses of action, and to identify, seek, and obtain the resources (such as training and equipment) that they need to succeed. A pay for performance system cannot have these desirable effects unless employees understand the organization’s goals, their role in achieving those goals, and how the pay system works. Agencies should not assume that these matters are self-explanatory, but agencies should make a conscious effort to create this understanding. That includes describing how the pay for performance system operates; informing employees what behaviors and accomplishments will be rewarded; providing regular, constructive performance feedback; training employees on how they should present their efforts and accomplishments; and telling employees how they have fared under the pay for performance system and the reasons for the outcome. Checks and balances are necessary. As discussed above, a pay for performance system requires that agencies allow supervisors to exercise some degree of discretion and judgment in evaluating and rewarding employee performance. However, a pay for performance system also requires fairness—both actual and perceived—for it to have any credibility or motivating power. xiv Designing an Effective Pay for Performance Compensation System Executive Summary Consequently, agencies must also ensure that supervisory decisions concerning employee performance and pay are fair and reasonable. Training and guidance are necessary but not entirely sufficient. Agencies should establish internal checks and balances—such as limited delegations of authority, review panels, and internal appeals processes—to ensure that supervisors are, in fact, using their judgment and discretion appropriately. A pay for performance system needs sufficient funding to provide high- performing employees with meaningful pay increases and bonuses. Pay for performance represents a commitment from the agency to recognize and reward excellence. A pay for performance system will lose all credibility if high performance goes unrewarded. Therefore, agencies should ensure that adequate funds are earmarked for performance awards. Pay for performance systems should be evaluated regularly and modified when necessary. The General Schedule classification and pay system has remained in place, with relatively few changes, for more than 50 years. Although we expect that the practice of pay for performance will endure, the useful life of even the best-designed pay for performance system will most likely be measured in years, not decades. Pay for performance systems need ongoing attention to keep them functioning properly. Organizational goals will change; performance goals and measures will become obsolete; performance may improve or decline; managers may make errors in evaluating performance or allocating rewards. For all these reasons and more, agencies need to monitor the operation and effectiveness of their pay for performance systems and modify them accordingly. Only by giving the pay systems and related organizational requirements the ongoing attention that they warrant will agencies be able to obtain optimal results from their pay for performance systems. A Report by the U.S. Merit Systems Protection Board 3 Pay for Performance Decision Points Overview Understanding the theory behind pay for performance and its potential impact is critical to understanding the role performance-based pay can play in an organization. Nevertheless, agencies also need to pay attention to technical design points to ensure that the mechanics of the system are sound. The effectiveness of pay for performance in facilitating recruitment, retention, and motivation (and the resulting improvements in individual and organizational performance) depends heavily upon matching the approach to the situation. Thus, agencies need to carefully consider numerous decision points, such as those discussed below. To make it even more challenging, the various choices often have both advantages and disadvantages. Although it is tempting to simply transplant compensation systems from other organizations where they appear to be functioning well, agencies need to tailor pay systems to fit their unique circumstances and needs. Fortunately, agencies can learn from the dilemmas others have faced and base their decisions on experience gained elsewhere combined with information they glean from within. Questions agencies need to ask themselves range from the most basic—“Is the agency ready for pay for performance?”—to the more specific, questions such as who should be covered, what behaviors should be rewarded, and how bonuses should be distributed. Obtaining adequate funding and ensuring fairness can also challenge agencies, so these goals need to be pursued early in the planning process. Anticipating the substantial decision points and understanding the available options can help agencies make the best possible decisions. While exploring these issues requires some time and effort, it is worth the investment to avoid potential negative consequences in the long term. Is the agency ready for pay for performance? Given the appeal of paying for performance instead of tenure, many agencies have already moved past deciding whether to adopt a performance-based pay system and are rapidly moving towards implementation. Some of these organizations may already have in place an organizational culture conducive to pay for performance. However, many do not. Fortunately, these agencies do not have to passively wait for the conditions to improve. They can use pay for performance as a tool for organizational change to move the agency in the desired direction. For example, 4 Designing an Effective Pay for Performance Compensation System Pay for Performance Decision Points agency leaders can drive major organizational change by demonstrating commitment to a performance-based pay strategy through their words and actions. As time progresses, the emphasis on performance perpetuates itself as the components of an effective pay for performance system facilitate further evolution towards a performance-based culture. Figure 1 depicts the cyclical nature of this process. Figure 1. Changing Organizational Culture With Pay for Performance Since organizational culture may influence the ease with which pay for performance can be implemented, agencies may find it useful to do a self-assessment before deciding how to design and implement a new pay system. Table 2 displays relevant dimensions of organizational readiness and selected indicators to help agencies gauge where they currently are. A Report by the U.S. Merit Systems Protection Board 5 Pay for Performance Decision Points Table 2. Assessing Organizational Readiness for Pay for Performance Dimension Indicator Organizational Culture n Open, two-way communication is valued and pursued. n Trust exists between employees and supervisors/managers. n Human resources management (HRM) systems such as selection, training, and performance evaluation have clear and consistent objectives and support pay for performance. Supervisors n Employee efforts support organizational goals. n Work assignment, evaluation of performance, and distribution of awards are fair. n Discretion and accountability go hand-in-hand. Performance Evaluation n Assessment of employees is fair and accurate. n Employees receive timely, accurate, and meaningful feedback. Funding n Appropriate pay increases and bonuses are given. n Top leadership is willing to make difficult choices when allocating funds and awards. Fairness n Checks and balances are in place. n Transparency is valued and ensured. Training n Training is provided to both supervisors and employees. n Training covers both pay system philosophy and mechanics. System Evaluation n The organization evaluates how the pay for performance system is being administered and whether the pay for performance system is accomplishing its goals. n Employee attitudes are tracked. Agencies may use the results of their self-assessment on the dimensions above to determine what areas they need to improve. As we discuss in the following paragraphs, many of the indicators listed above will serve readily as tools for facilitating organizational change or for monitoring the progress of organizational change as pay for performance is implemented. Organizational Culture. The concept of organizational culture covers numerous “values and ways of behaving that are common in a community.”4 Alignment among human resources management (HRM) practices is one aspect of culture that can greatly affect the success of pay for performance in an organization. Aligning the pay system with other HRM practices, such as selection, training, and especially performance evaluation, serves to focus employees on shared goals and values and delivers a consistent message regarding the value of performance. It also provides supervisors with the tools to move the workforce towards the desired performance- based culture. If agencies devote the proper attention to the supporting HRM systems, they can optimize the effectiveness of their pay for performance system and strengthen or create a performance-based organizational culture. 4 John P. Kotter and James L. Heskett, “Corporate Culture and Performance,” The Free Press, New York, NY, 1992, p. 141. 8 Designing an Effective Pay for Performance Compensation System Pay for Performance Decision Points work is more readily measured (and of more immediate importance to the public) than work performed by employees whose activities indirectly support organizational goals. In other cases, performance-based incentives may be reserved for those employees at the top levels of the organization. The logic behind this strategy is that accountability should be limited to those with the most control over results. In other words, since executives exert substantial influence over organizational success, they are entitled to significant recognition or blame for what they do or do not accomplish. In the private sector, Chief Executive Officers often receive sizable bonuses or performance-based pay increases that are linked to organizational outcomes, such as attainment of profit or other financial goals. Likewise, in the Federal Government, Senior Executives are eligible for annual bonuses and pay increases linked to their achievement of organizational objectives. Limiting pay for performance plans to select groups may enable the organization to highlight clearer links between employee behavior and outcomes, but doing so may create divisiveness. Depending upon the circumstances, such as whether the dual pay systems offer markedly different earning potential, coverage may be viewed as distinguishing between the “haves” and the “have nots,” creating some dissension between the two groups. This is particularly relevant if the benefits provided to one group are viewed as coming at a cost to the other group. What should be the timing for implementing pay for performance? Organizations often have the flexibility to decide whether the system should be implemented wholesale or in stages. If in stages, coverage can start with a single occupation or a limited group of occupations as a pilot test, and expand later to additional groups of employees once the system has proven itself. Starting with a small group of employees and later expanding coverage may enable the organization to demonstrate successes before rolling the system out to the entire organization. This also provides an opportunity to fine-tune the system and remedy any problems before they undermine its long-term success. In contrast, wholesale implementation may be preferable when the agency’s intent is to convey a dramatic organizational change message and foster a sense of solidarity. Implementing the new pay system wholesale also avoids the confusion and increased difficulty associated with administering multiple systems within an agency. What should be rewarded? An essential point to keep in mind is that pay for performance is a powerful tool, which must be used wisely. The axiom that “what gets measured, gets done” has particular relevance when measures are reinforced by monetary incentives. For A Report by the U.S. Merit Systems Protection Board 9 Pay for Performance Decision Points that reason, organizations must be very careful when deciding what to measure and reward, because they are quite likely to get what they measure—which may or may not be what they really want. In other words, agencies must be sure they are reinforcing desired behaviors associated with the most critical outcomes and not encouraging counter-productive responses. An effective performance appraisal system requires clearly defining expectations in advance, while recognizing that priorities may shift along the way. Enumerating specific goals gives employees a clear “road map” that they can use to decide how to allocate their time and efforts. Some jobs lend themselves more easily to this type of direction, while for others it is more difficult to specify in advance the precise accomplishments expected since the nature of the work is more complex or fluid. In these cases, flexibility regarding anticipated outcomes can be incorporated in the evaluation process. In the meantime, supervisors and employees should engage in continuing discussions so that expectations can be shared, despite necessary adjustments. Additionally, when circumstances outside the employees’ control determine outcomes, comparisons across employees in similar positions may assist the supervisor with evaluating employee performance. In such circumstances, supervisors’ subjective judgments necessarily play a role in evaluating performance. While subjective judgments cause some employees discomfort, supervisory discretion to evaluate performance is generally not something that can—or should be—orchestrated out of the process. Use of multiple measures. Given the complexity of work, multiple measures are often necessary to adequately capture accomplishment. To decide what to measure, agencies need to ensure that they focus on important outcomes without excluding other critical aspects of individual or organizational performance. One common problem in this area involves organizations that set quantitative goals only to find a negative impact on quality because important qualitative aspects of performance were not included in the goals. In other cases, organizations accurately identified top objectives, but overlooked subtle, yet important, priorities or activities. For example, by focusing employees’ attention only on part of a work process, such as timeliness, an agency can unintentionally instigate cutting corners and unsafe activities, which may serve to speed up the work process production at an unacceptable cost. While granting flexibility to employees who are pursuing difficult goals may encourage innovation, safeguards may need to be built in to ensure that necessary steps have not been inappropriately sacrificed. Additionally, it is important that the reward system does not undermine desirable aspects of performance, such as teamwork, that may not be explicitly recognized yet are important to organizational success. In situations that warrant looking at multiple facets of employee performance, a “balanced scorecard”5 perspective may prove to be very useful. Although notable 5 A “balanced scorecard” typically provides managers with a “fast but comprehensive view of the business” by featuring “financial measures” complemented by measures on “customer satisfaction, internal processes, and the organization’s innovation and improvement activities.” Robert S. Kaplan and David P. Norton, “The Balanced Scorecard—Measures That Drive Performance,” Harvard Business Review, 70(1), 1992, p. 71. 10 Designing an Effective Pay for Performance Compensation System Pay for Performance Decision Points differences between the private and public sectors impact what measures would be appropriate, a balanced scorecard approach for the Federal sector could include measures such as quantity and quality of output, teamwork, safety, and customer satisfaction, while focusing attention on the organization’s overarching mission.6 Alignment of organizational goals and measures. Supervisors frequently derive employees’ goals, at least in part, from high-level organizational goals. This “cascading” of goals is useful for aligning employee efforts with organizational objectives. To achieve this, employees need to understand how their individual performance supports organizational outcomes. However, supervisors should also recognize the value of a “bottom-up” approach that gives employees a voice in how they will be evaluated and some discretion in deciding how best to achieve the results desired. Excessive top-down control of goals, work methods, and job behaviors may stifle risk-taking and innovation by employees. In contrast, rewarding an open exchange of information may result in improved organizational outcomes over time as trust between the levels grows. Further, by aligning individual success with organizational success, there is a greater likelihood that agencies will be able to encourage employees to exert effort to achieve organizational objectives in concert with their personal goals. Standardized (organizational) vs. tailored (individual) criteria. Choosing between standardized organization-wide evaluation criteria and evaluation criteria tailored to individuals largely reflects an organization’s philosophy regarding the relative priority of what should be rewarded. Having a clear, overarching mission facilitates the use of standard criteria. Evaluating everyone against a common set of standards, linked to high level organizational goals, also serves to focus the attention of all employees on the highest level priorities. Likewise, using an agency- wide competency model reinforces agency values and promotes consistency across occupations and organizations. The downside of standardized criteria is that with their generality, the evaluation measures may not seem applicable to everyone. For example, front-line employees typically have a clearer line of sight to accomplishing mission objectives than administrative employees, who support the mission in a secondary manner. In other cases, the functions within an organization may be so diverse that it becomes difficult to use universal criteria. Providing leeway for tailoring criteria may be more appropriate in many organizations. This individuation of evaluation criteria may be by organizational subcomponent, occupation, grade level, or other categories. Individual vs. team vs. organizational performance. Similarly, the level at which performance is assessed for award purposes should reinforce the desired breadth of collaboration, although this must be balanced with the need to be able to identify individual contributions. It is important to consider whether cooperation should be encouraged within a discrete work unit or across a broader context, such 6 Robert S. Kaplan and David P. Norton, “Transforming the Balanced Scorecard from Performance Measurement to Strategic Management: Part I,” Accounting Horizons, 15(1), 2001, pp. 98–99. A Report by the U.S. Merit Systems Protection Board 13 Pay for Performance Decision Points equipment and information, as shown in the following examples. An employee may work diligently toward a goal, only to have the priorities shift just before project completion. Supervisors also exercise control over some of the parameters of what an employee can accomplish by allocating job assignments and training. Another external variable is location, which often drives workload and therefore, could be factored into performance-based pay decisions. For example, one location, such as a large airport, may have a greater volume of traffic than a smaller facility. Similarly, the work done at a headquarters office is usually quite distinct from that done at a field office. The unavailability of needed equipment or information can also prevent employees from working at their optimal level. Therefore, many of the variables that determine individual productivity operate outside of employees’ control. Although some agencies may focus exclusively on outcomes, others may decide to keep external constraints in mind when making pay and award decisions to avoid penalizing high performers who are negatively impacted by circumstances beyond their control. At a group level, work units are sometimes dependent upon another unit for a critical step in the process and cannot proceed until that entity completes its role. There may also be uncontrollable external events that hamper successful completion of objectives, such as budget limitations or technical failures. Therefore, individuals or groups may find that they have worked hard and have done everything they can, but are impeded by factors outside their control. Again, in these cases, raters must consider how to evaluate performance—whether to rely entirely on outcomes or to give at least partial credit for efforts—keeping in mind that it can be quite de-motivating to employees if the rating system is so rigid that it does not take external factors into account. Should performance-based pay be tied to the performance appraisal system? Although annual award decisions in the Federal Government have often been tied closely to performance appraisal ratings, this is not always the case in pay for performance settings. For example, when going to a pay for performance system, the Federal Aviation Administration converted its performance appraisal process into an ongoing qualitative discussion between the supervisor and the employee regarding accomplishments and areas for improvement. Separate from the performance appraisal review, the supervisor’s recommendation triggers individual pay for performance increases, while all eligible employees receive the organization-wide pay increases, dependent upon organizational goal achievement. The purpose of de-linking the performance appraisal process and monetary rewards was to defuse the emotionally-charged annual process of assigning a numerical value to performance. However, the Government Accountability Office (GAO) has expressed concerns that this approach is insufficient to “provide enough meaningful information and dispersion in ratings to recognize and reward top performers, help everyone attain their maximum potential, and deal with poor performers.”14 14 U.S. General Accounting Office, GAO–03–612T, April 1, 2003, p. 10. 14 Designing an Effective Pay for Performance Compensation System Pay for Performance Decision Points In other cases, agencies tie pay decisions tightly to the performance appraisal process. As an example, some organizations operate a point-based pay for performance system, assigning individuals a score on a 100 point continuum that correlates with a performance level. Those above a certain level receive salary increases, with the amount corresponding to placement along the range. Although these systems appear precise and rigorous, their ability to make such fine distinctions between employees is questionable. Even if it may be easy to separate the “stars” from the “average performers,” it tends to be much more difficult to make fine distinctions among average employees given the complex nature of job performance and the inexact and subjective nature of many evaluation systems. As a consequence, employees may feel that their ranking relative to colleagues is incorrect, which undermines confidence in the overall system. How should employees be rewarded? Pay for performance can encompass a variety of rewards for above average performance. The two most common are bonuses, which are one-time cash payments, and performance-based pay, which provides a permanent increase to base pay. Each of these has advantages and disadvantages, which are explored further below. Bonuses. Bonuses represent an amount of pay that is “at risk” every year. In contrast to base pay, which is stable and primarily reflects an employee’s market value, bonuses should depend purely on performance and are not guaranteed. Employees in these types of systems frequently receive a base pay that is considered comparable to average market rate to facilitate recruitment and retention of a high- quality workforce, but additional dollars are distributed (often annually) on the basis of performance during the rating period. As a result, employees are guaranteed a certain salary, with the potential for earning more. The amount generally depends on a variety of factors, such as the available funding and the evaluation of the individual’s contributions, but the agency retains discretion over how much to spend each year. As illustrated by the example shown in Figure 2, bonuses serve to raise an employee’s salary above average market rate but only on an annual basis. In other words, each year an employee must earn an amount above the base rate of pay. Since each year’s bonus is independent of the bonus earned in prior years, total salary can fluctuate dramatically from year to year. When the employee excels, he may receive a sizable bonus, but if the employee’s performance is “average” or lower, he may not receive a bonus and his salary drops to the base rate. A Report by the U.S. Merit Systems Protection Board 15 Pay for Performance Decision Points Figure 2. Impact of Bonuses on Total Salary 1 60 50 40 30 20 10 Dollars in thousands 0 Base pay with bonus Average market rate (Base pay) 70 2 3 Time in Years 4 5 Advantages. Advantages of bonuses include the following: n Employee performance may vary from year to year so an employee may deserve a bonus one year, but not the next. Since the increase resulting from bonuses does not carry over to subsequent years, agencies can distribute a larger pool of bonus money each year rather than continue to fund performance-based pay increases from prior years. n Agencies may use bonuses throughout the rating period to facilitate timeliness, thereby strengthening the connection in employees’ minds between efforts and outcomes. n If agencies direct funding for increases into a bonus pool, they may provide sizable bonuses to top performers. Disadvantages. Disadvantages include the following: n Putting a large amount of pay “at risk” to be earned each year makes the employee’s salary less predictable. n If base pay levels are not fully competitive, reliance on bonuses as a reward may increase turnover. n At this time, bonuses are typically small relative to base salary in the Federal Government (but not necessarily in the private sector). Therefore, a culture change will be required to view bonuses as a mechanism for providing substantial amounts of money to recognize high performers. 18 Designing an Effective Pay for Performance Compensation System Pay for Performance Decision Points He does not receive an increase in the third year. In his fourth year, his performance warrants a $50,000 salary, which would exceed the $45,000 control point. Therefore, he receives a $5,000 increase in base pay and the additional amount is paid as a bonus of $5,000. In the fifth year, the control point holds his salary at $45,000 and he receives a $15,000 bonus. Figure 4. Impact of Control Points on Total Salary 1 60 50 40 30 20 10 Dollars in thousands 0 70 2 3 Time in Years 4 5 $5,000 Bonus $15,000 Bonus Year 1 $30,000 Year 3 $40,000 Year 2 $40,000 Year 5 $45,000 Year 4 $45,000 Bonus Control point (average market rate) Performance-based pay Average entry level pay rate As the example helps illustrate, control points ensure that employees do not receive pay above a certain level unless they sustain high performance levels. Employees at the control point may see this as a downside, since they cannot accurately predict what their salary will be each year, but at least they can be confident that their pay will not slip below the control point once they reach it. How much pay should be contingent upon performance? Although Federal agencies have long had the legal authority to provide high- performers with cash awards, within-grade increases, and/or quality step increases, the total amount of these awards or increases has typically been quite small— generally less than 5 percent of a Federal employee’s salary.15,16 Pay for performance compensation strategies should offer a larger percentage to ensure that the increases are large enough to be noticed by both recipients and nonrecipients. 15 U.S. Office of Personnel Management, Central Personnel Data File, 2003. 16 U.S. Office of Personnel Management, website, www.opm.gov/oca/pay/html/wgifact.htm. A Report by the U.S. Merit Systems Protection Board 19 Pay for Performance Decision Points The Senior Executive Service (SES) appears to be leading the way in terms of having more pay “at risk” based on performance. The SES pay for performance plan has eliminated pay levels within the SES pay band to create an open range and set a higher “aggregate compensation level.”17 Within this revised pay structure, the link between pay and performance will be emphasized, not only by tying pay increases to performance, but also by virtue of the size of the new performance-based pay increases. Whereas only a 2.2 percent increase was available for distribution to recognize performance by SES members in January 2004, future increases can draw from the sum of the general increase and locality pay increases.18 Therefore, future pay increase pools may contain more funding to appropriately recognize performance. In the private sector, the amount of performance-based bonuses has also grown in recent years. The average amount that companies spend on performance-based bonuses as a percentage of payroll gradually increased from 3.8 percent in 1991 to 10.5 percent in 2002, before declining slightly to 8.8 percent in 2003 and rebounding to 9.5 percent in 2004 and 11.4 percent in 2005, with these fluctuations due primarily to economic constraints.19,20,21 However, the amount devoted to these incentives varies widely by job characteristics, such as job type, level, industry, and geographic location. For example, WorldatWork reported that bonuses range from an average of 5.4 percent of base pay for hourly workers and 11.7 percent for exempt salaried workers to 29.1 percent for executives, for whom bonuses can equal millions of dollars.22 Of course, since many employees may receive small bonuses or no bonuses, these averages may mask much larger bonuses for a sizable portion of the private sector workforce. Several factors should influence the decision regarding size of the bonus or pay increase. First, the amount should be proportionate to the contribution of the employee. For example, those who contribute the most should receive the largest payoff. Those who made relatively minor contributions should receive little or nothing, to reinforce the message that rewards are reserved for those who exceed the evaluation criteria. Although many organizations like to spread bonuses over a large population, the resulting amount often ends up being insignificant and the failure to distinguish between high and low performers undermines the purpose of a pay for performance system. High performers may become demoralized and decrease their output, while the low performers have no incentive to change their behaviors. Second, bonuses should be large enough to get the attention of all employees, 17 U.S. Office of Personnel Management, Memorandum for Heads of Departments and Agencies From Kay Coles James, Director. Subject: New Performance-Based Pay System for the Senior Executive Service. December 16, 2003. 18 Stephen Barr, “Pay-for-Performance System Makes an Uneven Debut Among SES Corps,” Federal Diary column, the Washington Post, February 9, 2004, p. B2. 19 Hewitt Associates, press release, Sept. 9, 2003. 20 Hewitt Associates, press release, Sept. 8, 2004. 21 Hewitt Associates, press release, Aug. 31, 2005. 22 WorldatWork, salary survey from the website www.worldatwork.org, 2003. 20 Designing an Effective Pay for Performance Compensation System Pay for Performance Decision Points whether or not they received a bonus or performance-based pay increase. In other words, the amount of the cash bonus or pay increase should be sizable enough to be viewed as worth the extra effort and employees should understand what is required to earn such a reward. Finally, the organization must work within its short- and long-term budgets and communicate openly with employees to avoid creating false expectations regarding current and future salary levels. How should performance-based pay be funded? In the private sector, financial results (e.g., income and profits) frequently determine and fund bonuses and pay increases. In this manner, employees are able to share in the increased proceeds that their efforts have brought to the company. In good years, funding is readily available. In bad years, the employees often share the downturn in the company’s fortunes by receiving little or no salary increase or bonus, whether or not the company’s decreased profit is due to their efforts or to external economic factors over which they have no control. Funding is dramatically different in the public sector. Few Federal agencies obtain financial returns from their work. Even when they do (e.g., revenue collection, law enforcement), it is rarely appropriate to allow Federal employees to directly benefit from these returns. Furthermore, most agencies are funded through appropriations, and the relationship between performance and appropriations is tenuous at best. High performance and mission accomplishment does not necessarily lead to larger appropriations, and mediocrity or failure may have few, if any, financial consequences. Therefore, most Government agencies must allocate funding for performance-based salary increases and bonuses from their existing salary budgets (i.e., funds that might otherwise be used for pay increases that are not tied to performance) and prudently manage the long-range impact on salary costs when awarding increases to base pay. However, the amount available each year for pay increases is typically quite limited—regardless of whether agencies have met or exceeded their performance goals. For example, the general increase provided to Federal employees has varied from 2.1 percent to 3.8 percent over the past 7 years (2000 through 2006).23 Within-grade increases (WGIs) for employees provide another source of funding for performance-based pay, but amount to only about 3 percent of annual pay for those who are scheduled to receive them.24 Only employees in the first three steps can receive annual within-grade increases, while those at midrange steps progress every other year and those at the highest steps wait 3 years before their next within-grade increase.25 Therefore, the total amount available through pooling the general increase and the WGI funding is still somewhat limited and subject to fluctuations over time. Nevertheless, employee groups and unions have strongly opposed eliminating 23 U.S. Office of Personnel Management, website, www.opm.gov/oca/06tables/index.asp. 24 U.S. Office of Personnel Management, website, www.opm.gov/oca/pay/html/wgifact.htm. 25 5 CFR 531.405. A Report by the U.S. Merit Systems Protection Board 23 Pay for Performance Decision Points Figure 5. A Forced Distribution of Performance Ratings 1 30 25 20 15 10 5 Frequency rating 0 35 2 3 Performance rating 4 5 An advantage of the forced distribution system is that it forces supervisors to make distinctions among employees rather than taking the easy way out and rating everyone the same. The impact on employees may also be useful when the work environment is conducive to competition because this strategy increases the incentive to outperform one’s peers. However, the forced distribution approach has a number of flaws.31 First of all, performance in organizations rarely fits a normal distribution. Although everyone cannot mathematically be “above average,” one would hope that a distribution of actual performance levels using an anchored rating scale would be positively skewed—that is, the vast majority of employees would perform at an acceptable or higher level. See Figure 6 for an illustration of this scenario. 31 Edward Lawler, “Rewarding Excellence: Pay Strategies for the New Economy,” Jossey-Bass Inc., Publishers, San Francisco, CA, 2000. 24 Designing an Effective Pay for Performance Compensation System Pay for Performance Decision Points Figure 6. A Forced Distribution of Performance Ratings Contrasted with Actual Performance Levels 16 60 50 40 30 20 10 Frequency of rating 0 Actual performance level Performance rating 70 2 3 4 5 Under this scenario, rather than preserving the integrity of the system by limiting the number of employees who receive the highest rating, a forced distribution may undermine the accuracy of ratings by requiring a set number of employees to receive each rating. Consequently, their ratings will not necessarily correspond to their actual performance. For example, in some organizations, there are many who excel and should receive the highest ratings, but when the number of “outstanding” and “exceptional” ratings is limited, supervisors must make arbitrary distinctions between employees. In response, some supervisors resort to having employees “take turns” receiving the top ratings, which creates a situation where an individual’s excellent performance may not be recognized in the appropriate timeframe. This is de-motivating for the employee and fails to serve the purpose of recognizing role models. At the opposite end of the performance spectrum, it would seem illogical to specify a minimum number of very low ratings (i.e., a number of poor performers on board) because poor performers should have improved their performance to an acceptable level or been dismissed. Another potential drawback of the forced distribution model for ratings is the fact that it pits employees against one another as they compete for scarce resources. Since many organizational cultures value teamwork, the net impact may be decreased organizational performance as employees become less likely to cooperate with one another since they view each other as “competition.” Employees may also withhold information from supervisors if they are concerned that this feedback may result in negative ratings. A Report by the U.S. Merit Systems Protection Board 25 Pay for Performance Decision Points Reward top 5 percent, 10 percent, 20 percent, etc. Another method to limit performance-based pay costs requires designating only the top percentage to receive increases. Such exclusivity may be driven by a number of considerations. Choosing a larger percentage translates into more people sharing the wealth, while a smaller percentage means that fewer people will be recognized, but they will be likely ro receive sizable increases. The organization’s culture often drives the decision regarding how many people should be included. In a strong team environment, broader coverage may make more sense, while more stringent limits may work in an individualistic culture. Some private sector corporations have begun implementing a restrictive policy to reward only the very top performers. Research into return on investment in these companies has found that the cost of providing sizable monetary incentives to the “stars” can be recouped as a direct result of their subsequent improved outcomes. Nevertheless, such elitism may leave the majority of the workforce demoralized. Federal agencies should carefully consider the implementation and effects of such a strategy, lest they turn the large group of average and above average performers into disgruntled employees. This is especially true since it is likely that the employees at the very top of the distribution will often be there year after year and others will never see positive monetary feedback no matter how hard they try. Control points in pay bands. As discussed earlier, control points in pay bands limit individual pay progression and thus serve as a “brake” on growth in total salary costs. Since they are frequently established in the middle of the pay band (i.e., below the band maximum) and usually correspond to the average market salary for the relevant job family or occupation, they help keep costs down although employees at this level are still eligible for annual cash bonuses to recognize their outstanding contributions. Who makes performance rating and pay decisions? Although supervisors typically rate employee performance and those ratings are usually linked with the actual distribution of pay increases, organizations may choose to retain higher-level control over both decisions. For example, at one extreme, control over rating and pay decisions may be centralized at the top level. This strategy works best in small organizations, and may be used to focus attention on the organization’s values, especially when all components and/or employees are rated on the standardized criteria or when organizational components are evaluated on whether they accomplished their unique objectives (and everyone within the component then receives the same award). A central board of senior managers may also exercise sole discretion over pay decisions when rating criteria are not standardized (i.e., a variety of individual competencies and outcomes are assessed), consistent treatment of all employees remains a primary goal, and the senior managers have enough perspective to be able to accurately judge performance across the organization. However, a potential hazard inherent in centralized evaluation of performance and distribution of pay increases relates to the likelihood that senior managers will not be knowledgeable about performance within subunits of the 28 Designing an Effective Pay for Performance Compensation System Pay for Performance Decision Points Employee (Self-rating). With increasing supervisory ratios and the need for supervisors to devote time to tasks other than observing the work of their employees, it is understandable that supervisors may not be familiar with all of an employee’s accomplishments during a rating period. Hence, it is advantageous for employees to provide their supervisors with a summary of accomplishments during the rating period. This enables employees to explain extenuating circumstances that prevented achievement of all the established objectives and to highlight accomplishments supervisors may otherwise overlook. Although some employees may embellish or underestimate their achievements, effective supervisors will use this exchange of information as an opportunity to clarify actual accomplishments and discuss with employees past, present, and future goals and ways the employees can improve their performance. Peers. Peers often occupy a position that provides them with insight into the day- to-day performance of their coworkers. Peers see coworkers on a regular basis and may be best able to judge the effort a colleague makes. Since their work is often very similar, peers are able to assess whether the outcomes reported by the employee are reasonable and if other constraints are operating that prevented the employee from accomplishing the desired objectives. However, peer input must be given weight cautiously, especially in a pay for performance setting where employees may view each other as competition because this may lead employees to undermine others to build themselves up or to establish mutually beneficial pacts with colleagues to rate each other positively. Further, opinions may be clouded by irrelevant biases if employees do not receive extensive training in proper evaluation techniques, which should raise their awareness of any discriminatory tendencies. Customers. Many Federal agencies have customers outside the Government, although the definition of customer may not correspond with what we typically view as a customer in a retail or service environment. In these agencies, employees need to treat their customers in a way that accomplishes the desired interaction in a professional, effective, and efficient manner. Additionally, Federal employees who work in support functions, such as human resources, finance, and information technology, have customers within their agency. In some cases, customer satisfaction can be traced directly to individual employees and may be an appropriate factor to consider in evaluating performance. How can agencies facilitate pay system integrity? Organizations can take a number of steps to evaluate the integrity of their compensation systems, even when significant supervisory discretion has been built into the structure. Differential treatment of employees—which, in the Federal Government, will require abandoning the practice of basing individual pay decisions on tenure and a low threshold level of performance—is both a primary strength and a potential weakness of a pay for performance system. Therefore, building safeguards in the system from the very beginning is critical to the acceptance and long-term viability of a performance-based pay system. Possible safeguards are discussed next. A Report by the U.S. Merit Systems Protection Board 29 Pay for Performance Decision Points Employee/union involvement in the development process. To obtain input from those who know the work best, agencies should consult their in-house experts—employees and the representing unions and other employee groups—early in the process. A truly collaborative process enables the organization to benefit from the perspectives of those who will be most affected by the new pay system and also facilitates buy-in by having all levels involved from the outset, rather than trying to pull everyone onboard after the plan has been finalized. Such involvement also sets the stage for participation in the future evaluations, revisions, and other steps that will be needed to fine-tune the pay for performance system. Supervisor’s role in the performance evaluation process. Having a fair and effective performance evaluation process is absolutely essential to preserving the integrity of a pay for performance system. However, a great process is meaningless without capable first-level supervisors who are able and willing to properly evaluate performance. Unfortunately, many supervisors do not feel prepared for the demands of the supervisory role, particularly those relating to performance evaluation. This is not surprising since some supervisors have been selected for technical ability rather than supervisory skills and have not been subsequently trained for significant components of their new role. Complicating matters further is the tendency for some of these supervisors to gravitate toward what they do well—serving as technical experts—leaving little time for the important responsibility of effectively managing their employees. Implementing a pay for performance system often magnifies the demands of the supervisor’s job because of the greatly increased significance of the supervisory responsibilities. Those responsibilities include communicating with the employee throughout the appraisal cycle, observing and documenting employee behavior, assimilating those details, and evaluating the employee’s performance in comparison to established standards and to other employees. Often, these responsibilities must be discharged in an increasingly complex and fast-paced work environment, for an increasing number of subordinates. Although none of these responsibilities is new, basing pay on performance evaluations ups the ante for employees. That, in turn, ups the ante for supervisors. Supervisors may experience even greater trepidation about the already contentious appraisal process, knowing that their decisions will affect employees’ livelihoods. Supervisors need the training, the tools, and the time to be able to function effectively in this difficult role. In general, change often creates an uncomfortable degree of uncertainty for everyone involved, but challenges to the status quo regarding pay can be particularly unsettling. Over time, supervisors and employees may develop mutual trust as both sides become more familiar with the pay for performance process. Holding supervisors accountable for their decisions can encourage them to make the right decisions despite the discomfort associated with efforts to change their own behaviors and those of their subordinates. 30 Designing an Effective Pay for Performance Compensation System Pay for Performance Decision Points Supervisor and employee training. Training serves as a critical though often overlooked step in the transition process of implementing a pay for performance system. Organizations must be willing to invest the necessary resources, even though training often seems superfluous compared to the need to fund the pay pool. Supervisors and employees need to be trained in the mechanics of the pay for performance process. Even more important, they should be educated on the philosophical underpinnings of the new pay system so they understand what the organization is trying to accomplish through it. Effective supervisor training is particularly essential, since the success of the system depends upon the ability of the supervisors to effectively communicate with employees, whether setting performance expectations, sharing ongoing feedback, or providing feedback during performance evaluation discussions. Training should be delivered via a variety of media to ensure that everyone understands the message. For example, when Honeywell’s Commercial Avionics Division implemented an incentive compensation plan, it educated their employees with two major training tools: a traditional HR manual to explain the new system to employees and a comic book that showed employees discussing pay.32 These two tools provided employees with information in the formal language needed to explain the details of the system, plus a more user-friendly version to which employees may have been better able to relate. Since rating performance usually relies heavily on subjective appraisals, learning how to fairly rate employee performance remains one of the most critical skills for supervisors (and for employees when they provide input on themselves, their peers, or their managers). Given that Federal Government work generally does not involve piecework or sales, rating Federal employee performance typically requires a great deal of judgment. Training is essential to ensure that judgment is not impaired by the failings described below: Leniency. In order to maximize benefits for their staff, supervisors may be inclined to be lenient and rate their employees too favorably. When a team is competing with other teams for a fixed pot of money (and especially if the team accomplishments are rolled up to reflect the supervisor’s accomplishments), a supervisor is likely to rate all of his employees as “outstanding.” Even when money has been divided among teams and the supervisor must decide how to allocate a limited amount of money among team members, many supervisors find it less stressful to rate everyone as basically the same to ensure that everyone receives a benefit. It may actually make sense to do this if the amount of money at stake is small, because to do otherwise could engender ill feelings without a good reason. However, there are downsides of spreading around award money, including: (1) each individual receives an amount so small it is not influential, and (2) it fails to distinguish between “average” and “better than average” performers, which means the motivational message is lost. 32 Shari Caudron, “How Pay Launched Performance,” Personnel Journal, 75(9), 1996, pp. 70–75. A Report by the U.S. Merit Systems Protection Board 33 Pay for Performance Decision Points outcomes) to obtain bonuses and pay increases. While most organizations provide some direction to employees, clarifying what is expected and what the organization will provide in return is absolutely essential in a pay for performance system if the bonuses are to have any value in motivating employees or focusing work efforts across the organization. The necessary communication must be at the organizational and individual supervisor level so employees receive a consistent message. Employees also need to understand the basis for pay decisions and how those decisions are made to bolster their perceptions of fairness, which is critical to job satisfaction.40 The third aspect relates to openness about pay outcomes: the disclosure of pay data at the organizational and individual levels. Some organizations publish information on their intranet websites regarding the distribution of performance ratings and accompanying bonuses and pay increases, without revealing any identities. Employees can view the data to see how they compare to others on the performance continuum and to learn the full range of monetary consequences associated with improving their performance (or allowing it to decline). Even fuller disclosure would reveal pay increase amounts by individual, although this appears to be a much less common practice despite the fact that the specific details on Federal employees’ pay and awards are considered public information. Revealing individual outcomes may generate jealousy and misunderstandings, especially when employees are not in a position to accurately evaluate their peers’ contributions. Releasing information on the aggregate level enables organizations to manage the message by clarifying what needs to get done to achieve a certain level of reward and demonstrating that the organization delivers what is promised. Ongoing evaluation of the system. To ensure that the pay for performance system is operating as intended, organizations should conduct ongoing data analyses to evaluate the system’s impact. These analyses should provide a comprehensive perspective on the effects of the pay system at various points in time, comparing, for example, pre-implementation measures with data for the system as it progresses and when it becomes firmly entrenched in the organizational culture. The focus of the analyses will vary according to organizational emphases, but issues such as fairness, cost, and the distribution of funds should be relevant to every agency. At a minimum, organizations should conduct some basic analyses of data from the human resources management data files to ensure that the system is operating in a fair, efficient, and effective manner. For example, agencies should compare performance ratings, salary levels, and pay increases by various demographic groupings, while keeping other factors in mind, such as employees’ tenure, education, and job series. They should also monitor the frequency (as a percentage of the population) of bonuses or increases and their amount to ensure that the distribution is consistent with organizational philosophy. Finally, agencies need to develop objective measures to help examine the impact of the pay system on outcomes. 40 L. Alan Witt and Lendell G. Nye, “Gender and relationship between perceived fairness of pay or promotion and job satisfaction,” Journal of Applied Psychology, 276(4), Dec. 1992, pp. 910–917. 34 Designing an Effective Pay for Performance Compensation System Pay for Performance Decision Points To assure the fairness of pay for performance systems, organizations should also evaluate the relationship between performance-based pay increases (and other pay increases) and sex, race and national origin (RNO), grade, occupation, and similar variables.41 Assuming that these analyses do not reveal patterns of apparent discrimination, publicizing them should further bolster employees’ perceptions that the process is fair. Review process. To facilitate consistent treatment of employees, some organizations have instituted a review process to examine ratings (and anticipated pay increases) within work groups or across parts of the organization. The reviewers may utilize demographic data, especially on sex and RNO, to identify disparate treatment of individuals by group membership. Managers may be required to justify individual ratings with summaries of employee accomplishments, although review panels are generally hesitant to overrule ratings because this would undermine the rater’s authority.42 The review panel may also question patterns of ratings that show a higher average for one group or organization than for others. In response, a manager may need to demonstrate that the team has made above average contributions to organizational goals, thus warranting greater bonuses. Appeals process. The credibility of a pay for performance system may be greatly enhanced by establishing an appeals process to provide employees a means to challenge rating or pay decisions that they believe to be unfair. That process could involve employees presenting their cases before an impartial panel tasked with reviewing these disputes. Summary Clearly, no one model exists for how to design, implement, and operate a pay for performance system. While agencies can learn from the experience of others, ultimately, each organization must consider the issues carefully in order to make the best decisions given their unique circumstances. Although paying for performance requires attention to an extensive list of serious issues, considering them in advance of implementation enables organizations to lay the groundwork for a successful performance-based compensation system. Since many of the decisions are interrelated, it is critical that they be considered simultaneously. While one best answer may not be readily apparent, agencies must consider the best information that they can gather during the design phase. Upon implementation, an ongoing effort to evaluate results and adjust the system as necessary should be viewed as a logical and critical aspect of supporting the system. 41 Michael M. Harris, Brad Gilbreath, and James A. Sunday, “A longitudinal examination of a merit pay system: Relationships among performance ratings, merit increases, and total pay increases,” Journal of Applied Psychology, 83(5), 1998, pp. 825–831. 42 Michael O’Donnell, “Creating a performance culture? Performance-based pay in the Australian public service,” Australian Journal of Public Administration, 57(3), 1998, pp. 28–40. A Report by the U.S. Merit Systems Protection Board 35 Pay for Performance Decision Points Change—especially change as momentous as introducing pay for performance— creates stress in an organization. When agencies embark on significant changes with low levels of trust in place, employees frequently experience anxiety about how they will be impacted. However, properly building, implementing, and operating a pay for performance system can actually serve as a tool for developing trust between supervisors and employees. For example, a review of the Department of Defense demonstration projects typically found that after a few years, many employee concerns regarding pay for performance were proven unfounded as employees became comfortable with the new system.43 Therefore, the fear of the unknown may be overcome if the right approach is taken. 43 U.S. General Accounting Office, “Implementing Pay for Performance at Selected Personnel Demonstration Projects,” GAO–04–83, Washington, DC, January 23, 2004, and U.S. Office of Personnel Management, “Summative Evaluation 2002 DoD S&T Reinvention Laboratory Demonstration Program,” Washington, DC, August 2002. 38 Designing an Effective Pay for Performance Compensation System While many of these requirements relate to effective human resources management practices that are important to any organization, pay for performance further increases their necessity. Attending to these human resources management issues provides agencies with a much greater likelihood of achieving a fair and effective pay for performance system. To make pay for performance successful, agencies need to make a substantial investment of time, money, and effort. Pay for performance systems require substantial initial and continuing investment. These resources must be carefully spent on building and maintaining a system that suits the organization’s mission and objectives. Performance evaluation serves as the foundation of a pay for performance system. An effective performance evaluation system is a fundamental prerequisite of pay for performance. Agencies must be able to communicate with employees regarding what the organization values and how it will accurately measure employee contributions to these goals. Without this information, agencies would be unable to appropriately distribute performance- based pay increases and bonuses. Agencies should select supervisors based on their supervisory potential, develop and manage them to function as supervisors rather than technicians or staff experts, and evaluate and pay them based on their performance as supervisors. Because supervisors play a pivotal role in pay for performance systems, it is essential that they be able and willing to perform the important supervisory functions inherent in performance-based pay systems. To achieve this goal, agencies must select, train, and pay supervisors based on their demonstration of qualities that are suited to a pay for performance environment. Communication, training, and transparency are essential elements of a good pay for performance system. The key to the effectiveness of a pay for performance system rests with clarifying the mission and objectives of the organization, how these are linked with employees’ efforts, and consequently, what competencies, behaviors, and/or outcomes the organization values. Open communication regarding goals and progress; training in the philosophy and mechanics of the pay system; and transparency regarding how the system operates can mobilize the workforce in the desired direction. Checks and balances are necessary. Agencies can greatly facilitate the real and perceived fairness of the pay system by building in appropriate checks and balances. Although knowledge about the agency’s pay for performance plan and transparency regarding its outcomes can help supervisors and employees understand how the system should work, other mechanisms to ensure fairness are needed to further raise and maintain confidence in the system. Conclusions and Recommendations A Report by the U.S. Merit Systems Protection Board 39 A pay for performance system needs sufficient funding to provide high performing employees with meaningful pay increases and bonuses. Being able to provide high performers with meaningful pay increases is critical to operating an effective pay for performance system. Therefore, agencies need to have adequate funding to support pay increases for those who deserve them. Pay for performance systems should be evaluated regularly and modified when necessary. Agencies should conduct an ongoing evaluation of the compensation system to help them ascertain whether organizational goals are being met and identify ways to improve the process. Summary In this report, we have discussed requirements of a successful pay for performance environment. Table 3 summarizes the characteristics that organizations should demonstrate to most effectively support pay for performance. However, taken as a whole, these qualities represent an ideal setting that exists in few, if any, organizations. Given that organizations cannot (and should not) wait for perfect circumstances to begin to implement pay for performance, agencies should keep these features in mind as a goal, and work towards them. Fortunately, if done correctly, implementing a performance-based pay system can also help agencies move in the direction of these critical success factors—if they understand how pay for performance can help them reach these goals. Agency leaders need to understand that they have enormous discretion and can select from a multitude of options to build a performance-based pay system that will work well under their unique circumstances. The greatest challenge for many decision makers is likely to be making the choices that will best enable their pay system to help them to achieve their goals. Conclusions and Recommendations 40 Designing an Effective Pay for Performance Compensation System Table 3. Features of a Successful Pay for Performance Environment Organizational culture 4 Organizational culture supports the concept of pay for performance. 4 Organizational leaders demonstrate commitment to a performance- based pay strategy through their words and actions. 4 Open communication is present. 4 A high degree of trust exists betw+een supervisors and employees. 4 Human resources management systems support pay for performance. For example, an effective performance management system is critical to support a pay for performance system, and both should be built on the same values. Training 4 Supervisors and employees receive advance and ongoing training that covers both the mechanics of the system and its underlying philosophy. Supervisors’ role 4 Supervisors treat employees fairly when assigning work, evaluating performance, and allocating rewards. 4 Supervisors monitor and document employees’ performance on a regular basis throughout the assessment cycle and provide feedback that is timely, accurate, and meaningful. 4 Supervisors assist poor performers, through measures such as feedback and training, to help them to improve their performance. If their performance does not improve, the employee may be subject to an adverse action, such as a pay reduction. 4 Supervisors have the necessary discretion to make personnel decisions. Performance evaluation 4 Supervisors and managers are held accountable for ensuring that performance ratings distinguish between levels of performance. 4 Performance measures have been carefully designed to encompass the most critical outcomes. 4 Performance is evaluated at the appropriate level. 4 Employees understand how their individual performance supports organizational outcomes. Fairness 4 The system includes checks and balances to ensure fairness. 4 Pay and bonuses are distributed according to performance. Funding 4 Adequate funding ensures that employees receive compensation that corresponds to their contributions. Evaluation 4 The compensation system is evaluated on an ongoing basis to ensure that it is accomplishing the desired objectives in a fair and cost-effective manner. 4 Measures of employee attitudes, such as employee engagement and motivation, as well as outcome indices, are tracked and analyzed to determine the impact of pay and other factors on employee satisfaction and performance. Conclusions and Recommendations A Report by the U.S. Merit Systems Protection Board 43 U.S. General Accounting Office. (2003). “Results-Oriented Cultures: Modern performance management systems are needed to effectively support pay for performance.” GAO–03–612T, Washington, DC, April 1, 2003. U.S. General Accounting Office. (2004). “Implementing Pay for Performance at Selected Personnel Demonstration Projects,” GAO–04–83, Washington, DC, January 23, 2004. U.S. Merit Systems Protection Board. (2003). “The Federal Workforce for the 21st Century: Results of the Merit Principles Survey 2000.” Washington, DC. U.S. Office of Personnel Management. (2002). “Summative evaluation 2002–DoD S&T Reinvention Laboratory Demonstration Program.” Washington, DC. U.S. Office of Personnel Management. (2003 and 2004). Civilian Personnel Data File (CPDF). U.S. Office of Personnel Management. (2003). Memorandum for Heads of Departments and Agencies from Kay Coles James, Director, Subject: New Performance-Based Pay System for the Senior Executive Service, December 16, 2003. U.S. Office of Personnel Management. (2003). Memorandum for Human Resources Directors from Ronald P. Sanders, Associate Director for Strategic Human Resources Policy, Subject: Recent Legislative Changes, December 24, 2003. U.S. Office of Personnel Management. (2004). General schedule within-grade increases from the website www.opm.gov/oca/pay/html/wgifact.htm. U.S. Office of Personnel Management. (2006). Salaries and wages from the website www.opm.gov/oca/06tables/index.asp. Wayne, Sandy J., Shore, Lynn M., Bommer, W. H., and Tetrick, Lois E. (2002). “The role of fair treatment and rewards in perceptions of organizational support and leader-member exchange.” Journal of Applied Psychology, 87(3), 590-598. Witt, L. Allen, and Nye, Lendell G. (1992). “Gender and relationship between perceived fairness of pay or promotion and job satisfaction.” Journal of Applied Psychology, 77(6), 910-917. WorldatWork. (2003). Salary survey from the website www.worldatwork.org. Bibliography U.S. Merit Systems Protection Board PRST STD 1615 M Street, N.W. U.S. POSTAGE PAID Washington, DC 20419 WASHINGTON DC OFFICIAL BUSINESS PERMIT NO. G-113 PENALTY FOR PRIVATE USE $300 Designing an Effective Pay for Performance Compensation System