Incentives and Performance Management 

in the Public Sector

Nancy R. Katz

Kennedy School of Government, Harvard University

June 2000

When designing a performance management system, does it make sense to link goal accomplishment to monetary incentives? How does incorporating incentives into such a system affect workers’ performance, motivation, and attitudes? What shape should such incentives take? The purpose of this document is to summarize the highlights of empirical research that addresses these questions.

From the substantial body of research on goal-setting we know that workers who are given goals that are specific and difficult outperform workers who are given a "do your best" goal or no goal at all (Latham & Lee). Goals do four things: direct attention; mobilize task effort; encourage task persistence; and facilitate development of task strategies (Locke; Locke & Latham). In other words, goals provide us with a clear direction; inform us that we need to try hard; remind us that an end is in sight; and encourage us to think about the process of reaching that end.

From the body of research on feedback we know that workers given information about how they perform generally outperform workers who are not given such feedback (Enzle & Ross; Harackiewicz). Furthermore, we know that comparative feedback is especially useful. In studies that compare task feedback that allows a worker to compare his/her competence relative to others versus task feedback that allows a worker to assess his/her competence in isolation only, the comparative feedback has a stronger impact on workers’ feelings of effectiveness (Sansone). The research does not give a clear answer regarding the impact on actual effectiveness; the dependent variables in most feedback studies are related to motivation (self-perceived effectiveness) rather than performance (actual effectiveness).

We also know that a combination of goals and feedback has a more powerful effect on task interest and persistence than either goals or feedback alone (Bandura & Cervone). When goals and feedback are combined, we know whether we are on the right path, and we know how much farther we need to go to reach our goal.

What happens when monetary incentives are added to the mix? While goals and feedback clearly boost performance, adding incentives can enhance task interest and persistence further still (Locke, et al.). The key word is "can" – whether incentives will have a positive effect on motivation depends upon the nature of the incentives.

Early research suggested that when extrinsic rewards such as monetary incentives were linked to performance on interesting and appealing tasks, intrinsic motivation decreased. The reason for this effect was that when workers were rewarded for doing work they already enjoyed, they observed themselves accepting a reward and inferred that they must be working for the reward rather than for intrinsic enjoyment of the task. Extrinsic rewards thus dampened intrinsic interest (Deci; Lepper, et al.). This finding received a great deal of attention, but subsequent research, however, provided limited support. One review of 24 relevant studies found that while 14 reported a negative impact of extrinsic rewards on intrinsic motivation, 10 reported no such impact (Boon & Cummings). It is now clear that extrinsic rewards can impair or enhance intrinsic motivation, depending upon how the rewards are constructed and construed. Harackiewicz, Manderlink, and Sansone explain that rewards have three aspects: evaluation, performance feedback, and reward value. Each aspect can have a different impact on intrinsic motivation. The evaluation aspect promotes feelings of external control and thus reduces intrinsic motivation. The feedback aspect promotes feelings internal control and thus enhances intrinsic motivation. The reward value aspect – the incentive as a symbolic cue of achievement – makes competence salient and thus enhances intrinsic motivation. In a series of experiments, Harackiewicz and colleagues showed that introducing contingent rewards can either enhance, inhibit, or have no effect on intrinsic motivation, depending upon which of the three aspects is made most salient. Other researchers have obtained similar results (Enzle & Ross). Researchers are just starting to address the most interesting question: under what conditions will a given aspect be most salient?

The dominant model for understanding and predicting whether a reward is likely to affect worker motivation and effort is Victor Vroom’s expectancy model. Several decades of research have largely substantiated the accuracy and robustness of this model. Vroom asserts that the strength of a reward’s impact on worker motivation and effort are a function of three factors: expectancy, instrumentality, and valence. Expectancy is the worker’s perception of the strength of the link between effort and performance. If I work hard and put myself out, will that translate into enhanced task performance? Instrumentality is the worker’s perception of the strength of the link between performance and the reward. If my performance is strong, will I receive commensurate rewards? Valence is the value a worker places on the reward. Will the rewards I receive be things I really care about?

Vroom’s model highlights the fact that, in order for an incentive program to have a strong impact on worker motivation and effort, the worker must believe that effort will lead to performance, that performance will lead to rewards, and that the rewards will be desirable. A manager who wants to design an effective incentive system must take into account the worker’s perceptions along each of these three dimensions. Empirical research has shown that the strength of these three factors can in fact predict a worker’s effort and performance level (Mitchell; Katzell & Thompson). If any one of these factors is weak, the incentive system is not likely to have a meaningful positive impact.

The research I have cited until this point is primarily on the individual level. That is, it focuses on the impact of incentives on individual workers. (Actually, the research participants are usually college students rather than workers.) What about the organizational level? Do organizations that introduce incentives perform better? The evidence is mixed. Some researchers firmly conclude that linking pay to results leads to enhanced organizational performance (e.g. Lawler; Ehrenberg & Bognanno; Kahn & Sherer; Mitra et al.; Zenger & Marshall). Other researchers conclude that contingent pay has no appreciable impact on organizational performance (e.g. Milkovich & Wigdor; Pearce, Stevenson, and Parry; Pearce & Porter; Shay). Part of the reason for the lack of consensus is that these studies encompass a broad array of incentive systems, such merit salary increases, one-time bonuses, gain-sharing programs, and profit-sharing programs. Furthermore, these studies operationalize performance in very different ways, such as quality of output, quantity of output, financial status, worker perceptions, etc. Indeed, recent literature reviews (Heneman; Milkovich & Wigdor) note that it is too early to draw firm conclusions about the impact of incentives on firm performance.

In an ideal world, incentives lead to enhanced motivation, effort, and performance. In the real world, however, incentives can have dysfunctional effects. The dysfunctional effect that has received the most study is the worker’s lament, "It’s not fair!" When rewards are contingent on performance, workers are finely attuned to issues of fairness, and a distribution of rewards that is perceived as even slightly unfair can lead to significant problems (Greenberg).

Researchers have tried to understand when workers are likely to perceive their rewards as unfair, and how they are likely to react. A useful model for understanding these issues is Stacy Adams’ equity theory, which has been supported and refined through decades of research. Adams’ theory, simply put, is this: To assess whether I am being rewarded fairly, I compare myself to others. I compare not only the rewards I receive (known as "outcomes" in equity theory) but also my inputs, and the ratio of my rewards to my inputs. Inputs include such things as effort, talent, and tenure. If my ratio is smaller than yours, I perceive the distribution of rewards as unfair. I will try to redress this injustice by changing the elements of our two ratios to make them equal. Research has indicated that the most common approach to equalizing the ratios is decreasing my inputs – that is, reducing my effort (Campbell & Pritchard, 1976). When I try to affix blame for my unjust situation, I am likely to blame external factors, such as my supervisor, the organization, or the incentive system, rather than myself (Taylor & Pierce). Not surprisingly then, when workers feel relatively undercompensated they are more likely to engage in theft, sabotage, politicking, and turnover (Summers & Hendrix; Martin; Greenberg).

On the other hand, when assessing the fairness of the distribution of rewards, it is possible that I will discover that my ratio of rewards to inputs is bigger than yours. Adams’ theoretical framework predicts that in this situation of relative overpayment, I will react by increasing my inputs – effort – in order to bring our ratios in line. The evidence for this prediction is mixed. In the short term, workers may indeed react to feeling overpaid by expending more effort to justify their rewards. Over the long term, however, workers are likely to simply change their perception of their deservedness, rather than sustaining an increased effort level (Campbell & Pritchard).

When I compare my reward-input ratio to yours, it is much more likely that I will perceive my ratio as too small rather than too big. This is because people generally have exaggerated perceptions of their performance (Meyer); this is a fundamental cognitive bias. On top of that, people generally are prone to compare their pay to others who are perceived as comparable in performance, but who earn more (Martin). Given these tendencies, the odds are good that I will be dissatisfied with my rewards and perceive myself as unjustly undercompensated.

How can a manager cope with the seemingly inevitable dissatisfaction that performance-contingent rewards will thus produce? One approach is to reduce the intensity of the pay plan – that is, to reduce the proportion of a worker’s pay that is contingent on performance. But this approach also reduces the positive impact of incentives on motivation and performance (Zenger & Marshall). Research suggests the most effective way to cope with a worker’s sense of "distributive injustice" is by establishing "procedural justice." Distributive justice concerns the relative size of my rewards; procedural justice concerns the process by which the size of my rewards was determined. Greenberg studied the simultaneous impact of distributive and procedural justice, and discovered a fascinating interaction. Workers perceived a high pay level as fair regardless of the process by which the pay level was determined. Workers perceived a low pay level as fair only when the process by which the pay level was determined was fair. In other words, workers tolerated a distribution of rewards that they felt was unfair so long as the process of determining the distribution seemed fair.

What contributes to workers’ sense that a process for determining rewards is fair? A process is more likely to be perceived as fair when it is open and transparent, and when workers can contribute to the process by providing relevant information (Lind & Taylor; Kanfer et al).

Given the level of interest in contingent pay in both the public and private sectors, there is surprisingly little research into how best to structure an incentive system. There are very few studies that compare the effectiveness of different incentive plans (Milkovich & Wigdor). This may be due to the difficulty of conducting comparative research. Comparing the impact of different incentive structures requires recruiting many organizations into a study. This is costly in both time and effort. It is much easier for a researcher to focus on the incentive scheme at one particular organization, although such analysis does not yield direct information about how one type of incentive scheme compares to another, or the conditions under which a particular type of incentive scheme makes sense. As noted earlier, Zenger & Marshall have found that schemes with greater incentive intensity (percentage of pay that is contingent on performance and thus at risk) have a larger positive impact on motivation and performance than schemes with lesser incentive intensity. We also know that, when designing group-based incentives, the smaller the group, the greater the impact of the incentives on motivation (Zenger & Marshall). But only one feature of incentive system design has received sustained attention :whether and when incentives should be individual or team-based.

Should incentives be based on individual or on group performance? Research shows that both approaches have benefits and both have costs. Basing rewards on individual performance is generally associated with increased pressure on individuals to perform and to accept responsibility for their own actions, and increased risk-taking behavior (Milkovich and Newman). When individualistic schemes successfully distinguish between high and low performers, such systems provide a valuable source of performance feedback, and foster the sense of a meritocracy. Individual rewards can be especially useful in large organizations where workers might otherwise feel lost in the system, according to Lawler.

When incentives are based on group performance (which typically means every group member receives an equal reward) group members report greater liking and respect for one another, enhanced self-esteem and perceptions of control, lower anxiety, and more task enjoyment (Johnson & Johnson). Slavin found greater communication among team members when rewards were group- rather than individual-based, even when the task did not require any interaction. Mesch et al. found higher levels of learning and information sharing among group members when rewards were based on group performance. Several investigators have found that group-based rewards foster cooperation and helping (Milkovich & Newman; Miller & Hamblin).

Both the individualistic and the egalitarian, group-based approaches have serious shortcomings. Under the individualistic approach, resources and information are more likely to be hoarded than shared. Individualistic systems can exacerbate the sense of a two-tiered society of organizational winners and losers. Outstanding performance appraisals are, at least in theory, reserved for a select minority. This can alienate the very people who most need to improve (Drennan). Instead of trying harder, low performers may rationalize their poor performance evaluations as merely a sign of incompetence or bias on the part of those conducting the performance appraisals. The organization can produce a residue of disgruntled people who feel they owe it nothing; indeed, they may wish it ill (Gabris & Mitchell). High performers also can suffer under individualistic pay schemes. Several classic case studies of incentive plans, from that of Roethlisberger & Dickson to William H. Whyte, have documented the ostracism and other negative social sanctions high performers must sometimes endure.

Group-based rewards can have dysfunctional effects as well. Group-based incentives can promote regression to the mean rather than outstanding contribution (Milkovich and Newman). Low performers may have little incentive to obtain training and raise their contribution. So as not to be taken advantage of, high performers may hold back from exerting themselves (Harkins) or leave the organization altogether. Alternatively, high performers may become vigilant police officers who pressure the low performers to try harder (Drabman, Spitalnik, & Spitalnik). As a result, low performers may feel tremendous pressure and scrutiny from other group members (Ames), which may further inhibit improvements in low performers’ output. Furthermore, the group product could suffer if low performers feel their low status gives them little right to exert influence or express their individual perspective.

Given the compelling data both pro and con, what can be inferred from this body of research? Incentives should be team-based when cooperation and knowledge-sharing are critical to task success, such as in cross-functional product development (Balkin & Gomez-Mejia; Milkovich & Wigdor). Task complexity is likely to shape the need for cooperation and the extent of interdependence among workers. When task success hinges on individual excellence, individual incentives are appropriate. This is likely for tasks that are simpler and less interdependent. The nature of the work should drive the design of the incentive system.

How should the design of an incentive system vary based on the type of worker? Several decades ago, sharp distinctions were drawn between the types of pay plans appropriate for senior managers, middle managers, and line workers. Those distinctions have now largely fallen by the wayside. Only a small amount of research speaks to this question (and the research addresses the question only indirectly). Gomez-Mejia & Balkin found that performance-contingent pay is less appropriate for workers who have a low willingness to take risks. Placed under a variable compensation regime, such workers are likely to withdraw, either cognitively or behaviorally. Igalens & Roussel report that exempt employees are more likely to experience contingent pay as motivating than are non-exempt employees. Bushman, Indejejikian, & Smith found that incentive intensity (the percentage of pay at risk) is greater at higher levels in an organizational hierarchy than at lower levels. Bushman et al. argue that this is appropriate, since people at higher levels have greater influence over the organization’s success.

Adams, J. 1963. Toward an understanding of inequity. Journal of Abnormal and Social Psychology, 67: 422-436.

Adams, J. 1965. Inequity in social exchange. In L. Berkowitz (Ed.), Advances in experimental social psychology, vol. 2, p. 267-296. New York: Academic Press.

Ames, C. 1981. Effects of group reward structure on children’ attributions and affect. American Educational Research Journal, 18: 273-287.

Balkin, D. & Gomez-Mejia, L. 1984. Determinants of R&D compensation strategies in the high tech industry. Personnel Psychology, 37:635-650.

Bandura, A. & Cervone, D. 1983. Self-evaluative and self-efficacy mechanisms governing the motivational effects of gal systems. Journal of Personality and Social Psychology, 45:1017-1028.

Boone, K. & Cummings, L. 1981. Cognitive evaluation theory: An experimental test of processes and outcomes. Organizational Behavior and Human Performance, Dec. 1981, 289-310.

Bushman, R., Indejejikian, R., & Smith, A. 1996. CEO compensation: The role of individual performance evaluation. Journal of Accounting & Economics, 21:161-194.

Campbell, D. & Pritchard, R. 1976. Motivation theory in industrial and organizational psychology. In M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology. P 63-130. Chicago: Rand McNally.

Deci, E. 1971. Effects of externally mediated rewards on intrinsic motivation. Journal of Personality and Social Psychology, 18:105-115.

Drabman, R., Spitalnik, R., and Spitalnik, K. 1974. Sociometric and disruptive behavior as a function of four types of token reinforcement systems. Journal of Applied Behavior Analysis, 7: 93-191.

Drennan, D. 1988. Motivating the majority. Management Today, March: 88-92.

Ehrenberg, R. & Bognanno, M. 1990. The incentive effects of tournaments revisited: Evidence from the PGA tour. Industrial and Labor Relations Review, 43:74-88.

Enzle, M. & Ross, J. 1978. Increasing and decreasing intrinsic interest with contingent rewards: A test of cognitive evaluation theory. Journal of Experimental Social Psychology, 14:588-597.

Gabris, G. and Mitchell, K. 1988. The impact of merit raise scores on employee attitudes: The Matthew effect of performance appraisal. Public Personnel Management, 17: 369-382.

Gerhart, B. & Milkovich, G., 1975. Employee compensation: Research and

practice. In M. Dunnette and L. Hough (Eds.), Handbook of Industrial and Organizational Psychology, 3:481-569. Palo Alto, CA: Consulting Psychologists Press.

Gerhart, B. & Milkovich, G. 1990. Organizational differences in managerial compensation and financial performance. Academy of Management Journal, 33:662-691.

Gomez-Mejia, L. & Balkin, D. 1989. Effectiveness of individual and aggregate compensation strategies. Industrial Relations, 28:431-445.

Greenberg, J. 1982. Approaching equity and avoiding inequity in groups and organizations. In J. Greenberg & R. Cohen (Eds.), Equity and justice in social behavior, p. 389-436. New York: Academic Press.

Greenberg, J. 1986. Determinants of perceived fairness of performance evaluations. Journal of Applied Psychology, 71: 340-342.

Greenberg, J. 1987. Reaction to procedural injustices in payment distributions: Do the means justify the ends? Journal of Applied Psychology, 72:55-61.

Greenberg, J. 1990. Employee theft as a reaction to underpayment of inequity: The hidden cost of pay cuts. Journal of Applied Psychology, 75: 561-568.

Harackiewicz, J., 1979. The effects of reward contingency and performance feedback on intrinsic motivation. Journal of Personality and Social Psychology, 37:1352-1363.

Harackiewicz, J., Manderlink, G. & Sansone, C. 1984. Rewarding pinball wizardry: Effects of evaluation and cue value on intrinsic interest. Journal of Personality and Social Psychology, 47:287-300.

Harackiewicz, J., Sansone, C. & Manderlink, G. 1985. Competence, achievement orientation, and intrinsic motivation: A process analysis. Journal of Personality and Social Psychology, 48:493-508.Lind, E. & Tyler, T. 1988. The social psychology of procedural justice. New York: Plenum.

Harkins, S. 1987. Social loafing and social facilitation. Journal of Experimental Social Psychology, 23: 1-18.

Hayes, L. 1976. The use of group contingencies for behavioral control: A review. Psychological Bulletin, 83: 628-648.

Heneman, R. 1992. Merit pay: Linking pay increases to performance ratings. MA: Addison-Wesley.

Igalens, J. & Rousel, P. 1999. A study of the relationships between compensation package, work motivation and job satisfaction. Journal of Organizational Behavior, 20:1003-1025.

Jenkins, G., Mitra, A., Gupta, N. & Shaw, J. 1998. Are financial incentives related to performance? A meta-analytic review of empirical research. Journal of Applied Psychology, 5:777-787.

Johnson, D. and Johnson, R. 1989. Cooperation and competition: Theory and research. Edina, MN: Interaction Book Company.

Kahn, L. & Shere, R. 1990. Contingent pay and managerial performance. Industrial and Labor Relations Review, 43:107-120.

Kanfer, R. 1990. Motivation theory and industrial and organizational psychology. In M. Dunnette (Ed.), Handbook of industrial and organizational psychology, p.75-170. Palo Alto, CA: Consulting Psychologists Press, Inc.

Kanfer, R., Sawyer, J. Earley, P. and Lind, E. 1987. Fairness and participation in evaluation procedures: Effects on task attitudes and performance. Social Justice Research, 1:235-249.

Katzell, R. & Thompson, D. 1990. An integrative model of work attitudes, motivation, and performance. Human Performance, 3:63-85.

Latham, G. & Lee, T. 1986. Goal setting. In E. Locke (Ed.), Generalizing from laboratory to field settings, p. 101-118. Lexington, MA: Lexington Books.

Lawler, E. 1971. Pay and organizational effectiveness. New York: McGraw-Hill.

Lawler, E. 1981. Pay and organizational development. Reading, MA: Addison-Wesley.

Lepper, M., & Greene, D., 1978. The hidden costs of reward. Hillsdale, NJ: Erlbaum.

Lind, E., & Taylor, T. 1988. The social psychology of procedural justice. New York: Plenum.

Locke, E. 1967. Motivational effects of knowledge of results: Knowledge or goal setting.

Journal of Applied Psychology, 51:324-329.

Locke, E., Feren, D., McCaleb, V., Shaw, K., & Denny, A. 1980. The relative effectiveness of four methods of motivating employee performance. In K. Duncan, M. Gruenberg, & D. Wallis (Eds.), Changes in working life, p.363-388. New York: Wiley.

Locke, E., & Latham, G. 1990. A theory of goal setting and task performance.

Englewood Cliffs, NJ: Prentice-Hall.

Martin, J. 1981. Relative deprivation: A theory of distributive jsutice for an era of shrinking resources. In L. Cummings & B. Staw (Eds.), Research in Organizational Behavior, v. 3, p. 53-107. Greenwich, CT: JAI Press.

Mesch, D., Lew, M., Johnson, D. and Johnson, R. 1988. Impact of positive interdependence and academic group contingencies on achievement. Journal of Social Psychology, 128: 345-352.

Meyer, H. 1975. The pay-for-performance dilemma. Organizational Dynamics, 3: 39-50.

Milkovich, G. and Newman, J. 1990. Compensation. Homewood, IL: BPI/Irwin.

Milkovich, G. & Wigdor, A. 1991. Pay for performance: Evaluating performance appraisal and merit pay. Washington, DC: National Academy Press.

Miller, L., and Hamblin, R. 1963. Interdependence, differential rewarding, and productivity. American Sociological Review, 28: 768-778.

Mitchell, T. 1982. Expectancy-value models in organizational psychology. In N. Feather (Ed.), Expectations and actions: Expectancy-value models in psychology, p.293-312. Hillsdale, NJ: Erlbaum.

Pearce, J. & Perry, J. 1983. Federal merit pay: A longitudinal analysis. Public Administration Quarterly, 43:315-325.

Pearce, J. & Porter, L. 1986. Employee responses to formal performance appraisal feedback. Journal of Applied Psychology, 71: 211-218.

Pearce, J., Stevenson, W., & Perry, J. 1985. Managerial compensation based on organizational performance: A time series analysis of the effects of merit pay. Academy of Management Journal, 28:261-278,

Roethlisberger, F. and Dickson, G. 1929. Management and the worker. Cambridge, MA: Harvard University Press.

Schay, B. 1988. Effects of performance-contingent pay on employee attitudes. Public Personnel Management, 17:237-250.

Slavin, R. 1983. Cooperative learning. New York, NY: Longman.

Summers, T., & St. Hendrix, W. 1991. Modeling the role of pay equity perceptions: A field study. Journal of Occupational Psychology, 64: 145-155.

Taylor, P. & Pierce, J. 1999. Effects of introducing a performance management system on employees' subsequent attitudes and effort. Public Personnel Management, 28:423-452.

Vroom, V. 1964. Work and motivation. New York: Wiley.

Zenger, T. & Marshall, C. 1995. Group-based plans: An empirical test of the relationships among size, incentive intensity and performance. Academy of Management Best Paper Proceedings, p. 161.

Additional Readings and Resources

 Performance Management Project  Home Page

Visions Project Home Page