Virtually every organization has a performance management system that is expected to accomplish a number of important objectives with respect to human capital management. The objectives often include motivating performance, helping individuals develop their skills, building a performance culture, determining who should be promoted, eliminating individuals who are poor performers, and helping implement business strategies. There is little doubt that a performance management system which can accomplish these objectives can make a very positive contribution to organizational effectiveness, but there is less clarity about what practices make a performance management system effective.
There are a large number of design features that potentially can influence the effectiveness of a performance management system, and many of these have been empirically studied to determine their impact. For example, there is considerable research which shows that performance management effectiveness increases when there is ongoing feedback, behavior-based measures are used and preset goals and trained raters are employed. There is one potential determinant of performance management system effectiveness, however, which has received relatively little attention: how tightly the results of the performance management system are tied to significant rewards. The lack of attention to this impact is particularly pronounced when it comes to the issue of using a performance management system to systematically remove lower-performing employees from the organization.
There are a number of reasons for believing that systematically tying rewards to the outcome of a performance management system will make the performance management system more effective with respect to motivation, but there are also some that suggest it will make it less effective with respect to development. In a well known 1965 article that is based on research done in General Electric Co., Meyer, Kay and French argue that when rewards are tied to performance discussions, individuals tend to hear only the reward system part of the message. They do not hear the kind of useful feedback that will allow them to improve their performance and develop their skills. This has led some companies, although not G.E., to separate the discussion of rewards from the discussion of performance.
On the other hand, it is reasonable to argue that when rewards are tied to the outcome of performance appraisals it will lead to more effective performance management systems. Managers will be particularly concerned about doing a good job, since the outcome of the appraisal will have a significant impact on their ability to allocate rewards based on performance and motivate those individuals who work for them. Similarly, in the case of individuals, they know that how well the performance review goes will effect rewards that are important to them, so they may be particularly motivated to prepare for the session and see that it goes well. Further, there is a good possibility that when appraisals are used to determine rewards, organizations will put more pressure on managers to differentiate among the employees they are appraising since this is key to rewarding individuals for their performance.
At this point in time no discussion of a performance management system can completely ignore the increased popularity of forced distribution and seriatum ranking systems. Ranking systems that force managers to place their employees in a 1 to N order based on their performance have existed for decades. They are frequently used in research laboratories (e.g., Sandia, Lawrence Livermore) and often place hundreds or even thousands of employees in a single ranking. Forced distribution systems which require that managers identify a predetermined percentage of their employees as poor performers and outstanding performers have become more popular in the last decade. General Electric’s forced distribution approach has received a great deal of publicity. Other adopters of this approach include Intel Corp., Ford motor Co., Goodyear Tire & Rubber Co., Enron Corp., Capital One Financial Corp., and Electronic Data Systems (EDS) Corp.
General Electric’s former chief executive officer (CEO) Jack Welch is among the most vocal and articulate advocates of forced distribution performance management systems that force turnover of the lowest-performing employees each year. At GE, it’s the bottom 10 percent of employees who are supposed to be eliminated. Mr. Welch feels so strongly about this practice that he highlighted it in his 1999 letter to GE shareholders and advocated it again in his book Jack: Straight from the Gut. In addition to arguing that a forced distribution system is good for organizations, he states it is good for individuals, because it takes people who are failing out of situations that are bad for them and the company.
There are many criticisms of the forced distribution approach, including the arguments that it represents an unrealistic mandate, and that it may force managers to make judgments they cannot support with data and do not believe in. Because of this it can lead to decisions that cannot be defended when there is an adverse impact. On the other hand, its proponents argue that it puts a rigor and discipline into the performance management system that overcomes the all too common leniency tendencies of managers. The key question, of course, concerns whether or not using a forced distribution system in combination with rewarding performance makes a performance management system more effective or less effective.
The results of the study provide clear guidance with respect to the impact of tying performance appraisal results to changes in pay and employment status. They strongly support the view that performance appraisal systems are more effective when there is a connection between the results of the performance management system and the reward system of the organization. This finding, hopefully, will end the debate about whether an appraisal system is damaged by using it to determine the amount of someone’s pay increase, the size of their bonus, or the stock options that they receive. It argues strongly that organizations make a mistake when they separate appraising performance from determining pay changes. It also makes it clear that terminating individuals on the basis of the appraisal system is associated with more positive performance appraisal outcomes.
The present study provides little direct evidence on why appraisals are more effective when they affect rewards, but there are a number of likely explanations for it. First and foremost, when rewards are involved it means that a performance management system is likely to be taken very seriously by both the appraiser and the appraisee. It assures that the appraisal will impact the individual’s relationship with the organization and potentially his or her compensation. That means he or she is motivated to see that it goes well—as is the appraiser, who is usually charged with doing a fair allocation of a limited amount of financial rewards.
It is also quite possible that organizations spend more time developing and training individuals to do performance appraisals when financial rewards and continued employment are on the line. At the organizational level, it is obvious that when significant financial rewards are at stake, they need to support the system and be sure that it is well executed.
The clear recommendation to organizations that follows from this research is that it is a good practice to tie the results of performance appraisals to the financial rewards individuals receive. There is a considerable amount of literature that suggest tying significant changes in pay to performance has a positive impact on motivation, and the present study suggests that it does not decrease the effectiveness of the appraisal system. Quite to the contrary, it increases it and makes it more effective as a development tool and at getting the organization and individuals committed to doing a good job of appraising and rewarding performance.
The results, which compare those organizations that use forced distribution techniques in their appraisals and those who do not, also provide guidance as to how organizations should do performance appraisals. They argue that using them does not improve performance appraisals and may, in fact, worsen them.
Forced distribution approaches tend to be associated with lower effectiveness in general and, in particular, when appraisal results are tied to termination. There are a number of possible explanations for why forced distributions may not have a positive impact, but perhaps the main one is that forcing raters to conform to a pre-determined distribution leads to them being uncomfortable and, in many cases, to make bad judgments. It also may lead to them being less committed to the process and to give mixed messages to the individuals being appraised. For example, they may discredit the result by saying, “I had to rate someone low.” Finally, it may undermine the credibility of the system in the eyes of those being appraised and those doing the appraisal. The net effect of this, undoubtedly, is a lesser commitment on the part of those doing the appraisal, and ambivalence or flat-out opposition to the whole process.
From the point of view of the individuals being appraised, they may well see the process as unfair because someone “has to be” rated as a poor performer, and as a result they may be less receptive to the process and to “hearing” the results of it. It can also lead to unhealthy competition among peers. When employees in a work area compete with each other for ratings, knowing there is always a percentage at the bottom who will be forced out, it creates fear and selfishness. People are much less likely to help each other, train each other, share information, and operate as an effective team. In today’s flatter, knowledge work-driven, more team-based organizations, excessive internal competition can take a significant toll on organizational performance. Postmortems of Enron suggest that severe internal competition, which developed as a result of their system, led to its demise.
Legal exposure is also a potential problem with quota systems. Numerous court cases in the U.S. have challenged the use of forced ranking or distribution systems. In 1975, I was an expert witness in a class-action age discrimination suit against the Sandia Corporation that focused on the validity of its forced distribution appraisal system. Sandia lost the case because it couldn’t prove that the rankings it created led to valid measures of individual performance. Recently, Ford Motor Company, reacting to the same problem, abandoned its forced distribution system after the company was threatened with lawsuits. In 2002, AARP filed a class action lawsuit against Goodyear, charging that its forced distribution performance management system discriminated based on age. Goodyear immediately abandoned its rating system.
My quarrel isn’t with the premise that companies need to identify and eliminate poor performers. Organizations certainly need methods and processes to accurately identify people who are not making the grade. Setting high standards, and dealing with those people who don’t meet or exceed them, is necessary in order to continuously improve organizational performance. The issue is how to do this effectively. Organizations need practices and methods for properly and fairly identifying poor performers, for motivating individuals and developing their talents, and for cutting losses for the individual and the company when someone is not working out.
Forced distribution approaches are an imposed bureaucratic solution to a real performance management problem, rater leniency. Often it is hard to get managers to identify poor performers when they know it may result in their being terminated or not getting a pay raise. But, in most cases, forced distribution methods are not be the best solution to this problem.
It is important to note that there are alternatives to the forced distribution approach that usually yield better results. Some organizations use guided distributions which give the raters more flexibility in terms of how many individuals they put in each of their categories, and work hard to influence the fairness of the ratings by training raters and by having “calibration meetings” in which several raters discuss how they rated individuals, why they rated them a particular way, and come to a mutual agreement about where different individuals fall. When these calibration methods are well facilitated and based on good data, they can provide results that are more credible to both the raters and the individuals being rated, and as a result lead to more positive performance management system outcomes.
Overall, the results of this study suggest that performance appraisal systems can have positive outcomes. The average scores on effectiveness are relatively high. This is encouraging since performance management systems are often seen as automatic losers for organizations. The evidence also clearly says, however, that if done incorrectly performance management systems can be a significant negative in an organization. What constitutes doing it right? The results suggest that using it to tie rewards to performance is a significant positive, and that using forced distribution technology is more of a negative than a positive, particularly when terminations are involved.
