If you’re a change-minded senior executive looking for ways to improve performance, cut costs, or free up resources that can be redirected against important issues waiting in the wings, you might give serious thought to scrapping your company’s performance appraisal system. It devours staggering amounts of time and energy, it depresses and demotivates people, it destroys trust and teamwork and, adding insult to injury, it delivers little demonstrable value at great cost.
Here’s a quick and easy performance appraisal cost-estimating exercise you can apply to your own company: double the number of employees who receive appraisals, add three zeroes, and place a dollar sign in front of the resultant number. That’s the cost of the performance appraisal system at your company. If you think $2,000 per employee is too high, don’t double the number of employees, just add three zeroes. [ Lest you think this estimating methodology too cavalier, please know that two very thorough cost estimates were submitted by respondents to the e-mail survey that gave rise to this paper. One came in at $1,945 per employee and the other at $2,200. The latter included the costs of union participation so it was rounded down to $2,000.] But you probably won’t like that result either.
If you do scrap your company’s performance appraisal system you will accomplish more than realize a sizable cost savings; you will also, in one bold move, unfreeze your organization’s culture and eliminate one of the chief structural obstacles to any changes you and your management team might be contemplating.
The one thing you should not do is listen to those who will implore you to let them redesign the performance appraisal system to make it more supportive of the changes you have in mind.
Some reading these words will consider the previous comments outrageous and unfounded. But, as Craig Brooks, the director of a Winona, Minnesota human services organization and a 26-year veteran of performance appraisal sessions claims, “I could retire on the salary I earned during those meetings.” An exaggeration? Perhaps, but the point is clear enough: performance appraisals chew up a lot of time and money. On the other hand, they supposedly provide benefits that offset these costs. But do they? The Internet offered an unprecedented means of finding out and so, during the last two months of 1995, three e-mail messages were posted in quick succession to several discussion lists on the Internet (see the list at the end of this paper). The first message dealt with the form and function of the “classic” performance appraisal system — one where a manager appraises the performance of a subordinate or “direct report.” The second proposed an informal cost-benefit analysis of such appraisal systems. It included a starter list of costs and benefits as well as a call for additional costs and benefits from anyone interested in participating. The third presented a summary and synthesis of all responses and some rather obvious conclusions. The bottom line? Performance appraisal systems provide questionable benefits at amazingly high costs. In one company alone, the costs were conservatively estimated at almost $100 million.
The Accepted Mythology of Performance Appraisal Systems
From the viewpoint of senior executives, performance appraisal systems are generally regarded as a necessary part of the organization’s management system. This condition seems to exist because of the widely accepted, generally unquestioned benefits of performance appraisal systems. If asked, a typical executive might indicate the following benefits:
- The system provides employees with an opportunity to receive feedback regarding their performance, usually at least once a year and often on an interim basis during the year. This leads to reduced error and waste, increased productivity, improved quality and service for customers, as well as enhanced employee motivation, commitment, and a sense of ownership.
- The system provides an opportunity for performance related discussions that include the following aims: setting work objectives for the employee, aligning individual and organizational goals, identifying training and development needs, and discussing career progression opportunities.
- The system standardizes performance appraisals and makes them objective by providing uniform processes and criteria. This further results in a fair, valid, and legally defensible basis for rewarding and recognizing individual performance.
- The system affords the corporation legal protection against employee lawsuits for discrimination and wrongful termination.
A perhaps less typical but more candid executive might add a final benefit: the formal performance appraisal system shores up an organization’s hierarchical authority system. It gives the supervising manager control over the carrots and sticks in what is essentially a carrot-and-stick management system.
The list of benefits above, with the exception of the last one, represents an idealized view of performance appraisal systems, a view that is espoused by many but achieved by few, if any. Why? What is it about organizations that causes performance appraisal systems to fall short of this ideal? Can we or should we try to change things so that performance appraisal systems work the way we want? Or is there a better course of action? Answering these questions requires taking a rational look at the form and function of performance appraisal systems, and the effects such systems have on the very people they are intended to help.
Figure 1 – A Basic Performance Appraisal System
The general form of a basic performance appraisal system is depicted in Figure 1. A discussion follows.
Based on his or her perceptions, a manager prepares an appraisal of another employee. Appraisals typically have two components: text, and a number. The number is usually the basis for determining the employee’s merit increase (i.e., the size of the pay raise for the subsequent year). This is often quite modest and amounts to little more than a cost-of-living increase, an offset against inflation. Moreover, differences between the maximum and minimum increases are also quite modest. The merit carrot is not a very big one.
Perhaps the most significant aspect of the structure depicted in Figure 1 is that the appraisal has as its primary input the perceptions of the manager. Technically speaking, they are the only input. Given this model, it is obvious that if the system is to work effectively the manager’s perceptions must be objective, accurate, comprehensive, and free from any significant bias, distortion or undue influence; otherwise, the system is patently flawed. This leads to the following assertion:
The structure of the typical performance appraisal system makes managers who prepare appraisals the targets of efforts aimed at influencing, shaping, and just plain manipulating their perceptions and the appraisals based on these perceptions.
Several people have an interest in influencing a manager’s appraisal of a given employee’s performance. The most obvious is the employee. But there are others. These include other employees who are being appraised by the same manager, and anyone with a vested interest in having a given employee receive a good or a bad appraisal; for example, clients, customers, mentors, co-workers, and other managers whose own subordinates must compete for a finite pool of merit increase monies, plum assignments, and increasingly limited promotion opportunities. In a word, the politics of performance appraisal can be fierce. The preceding assertion may be elaborated upon as follows:
Many efforts to influence the perceptions of the managers who prepare appraisals, and the appraisals they prepare, are independent of and often have no relation to the performance of the person being appraised.
People and politics are not the only forces tending to negate the positive potential of performance appraisal systems. There are also important systemic or structural factors at work.
An appraisal leads to a merit increase. The size of the merit pool is limited and the distribution of these monies is typically according to some formula. Thus, in a performance appraisal system that allocates merit increase percentage on a five-point scale, not everyone can receive a five because there isn’t enough money available to support such an outcome. This is a restraint, a “can’t do.” By the same token, the numbers assigned must fit within the limits of the available pool of merit monies. This is a constraint, a “must do.” Restraints and constraints can also include EEO and affirmative action considerations. Because merit rating numbers must be adjusted to meet various restraints and constraints, the language and tone of the appraisals must in turn be adjusted so as to be consistent with the numbers. From this follows an inescapable conclusion: the honest, fair, valid, and objective assessment of all employees is literally impossible. The structure, restraints, and constraints of the system do not permit it.
The preceding discussion looks at performance appraisal systems mainly from a managerial perspective. But how does it look to employees, and what are its effects on them? People responding to the Internet queries provided the following answers to these questions.
Reductions in Productivity
Several people cited temporary reductions in productivity in the aftermath of the appraisal review sessions. One person estimated this period of reduced contribution lasts for about three months. An employee of the federal government said this period lasts at least six months. Even if it is assumed that such periods last no more than a few days or weeks, and that they represent a decrease in productivity of no more than 10 percent, the costs are still astronomical.
Erosion of Performance
Tauo Jokinen, a product development manager with Nokia, conjectured that performance appraisal systems actually erode performance over time as a result of people endeavoring to set goals that are achievable, thus ensuring themselves a decent appraisal. This might be viewed as a form of structural deflation regarding performance, and it is quite reminiscent of the late Kenneth Berrien’s view that management might control the lower limits of productivity but employees are clearly in control of the upper limits. [ Berrien’s comments were made in the context of a discussion about the balance of control between a supra system and its subsystems, and can be found in Chapter VII of General and Social Systems (Rutgers University Press: 1968).]
Creation of Emotional Anguish
Also cited were negative emotional states: worrying, depression, stress, and anguish (on the part of those giving as well as those receiving appraisals). After first acknowledging the “hard” costs of performance appraisals, Harry Heflin, an engineer with Intersys who is also chairman of the IEEE Engineering Management Society in Boston, wrote, “But I think the real cost is the emotional anguish as everyone anticipates, prepares for, and works the process.”
Damaging to Morale & Motivation
Closely related to the emotional factors cited above are the penalties paid in the form of decreased morale and motivation. These are deemed especially severe when the performance appraisal system is seen as “bad” or unfair. An element of unfairness cited by Charles Ladd, a TQM consultant, is the use of performance appraisal systems to reward or punish people for what are really natural variations in system or process performance. This means that people are praised and rewarded or cursed and punished for factors beyond their power to influence let alone control.
Emphasizing Individual vs. Team and Task vs. Process
One factor the author was sure would be cited, but wasn’t, is that the classic performance appraisal system emphasizes individual or task-level performance instead of team or process performance. Appraising individual performance can be a divisive factor in an environment where genuine teamwork is required. Consequently, in times of change, retaining an appraisal system that focuses on individual task performance sends at best a mixed message when management calls for teams or wants to focus on business process performance instead of individual task performance.
Fostering A Short-term View
Another factor the author thought would be cited and wasn’t is the short-term view that is inherent in annual performance appraisal systems. Essentially, annual performance appraisal systems ask of employees, “What have you done for us this year?” Employee contributions over time — past or future — do not enter into the equation. Little wonder, then, that Mike Hammer, the famed reengineering guru, could be heard lamenting the lack of a long-term view in one of his recent seminars. [ Mike could be heard uttering this lament in Boston, on December 4, 1995, during the first offering of his new seminar, “The Process-Centered Organization.” Mike was lambasting the media and the educational establishment for churning out young people with a short-term, self-interested view.]
Institutionalizing Existing Values & Biases
A military officer with a Ph.D., who is stationed at the Pentagon and who wishes to remain anonymous, observed that performance appraisal systems serve to institutionalize the values and prejudices of those in power — and to protect these values and prejudices from challenge. Consultant Charles Ladd made this same observation independently of the officer. Both argued that this aspect of performance appraisal systems forms a structural impediment to cultural change, that it acts to maintain the status quo.
Fostering Fear and Lack of Trust
Directly related to the factors cited above is the degree of fear associated with the appraisal system. This ties to a lack of trust in one’s boss, and management in general, and leads to a phenomenon known as “malicious compliance,” that is, a passive-aggressive stance of “tell me what you want me to do and I’ll do it” on the part of an employee. As one might expect, Deming’s dictum to drive fear out of the workplace was frequently cited in this context.
A Carrot-and-Stick Management System
The source of the fear cited above owes to the fact that the carrot-and-stick nature of appraisal systems is mostly stick. Performance appraisals become a permanent part of the employees’ personnel folders. There, many people have access to them including prospective employers elsewhere within the company, the human resources department (HR), and other executives and senior managers. Past appraisals exert a significant influence over status and standing, future assignments, and promotions. Thus, although performance appraisal systems do not distribute much in the way of rewards, they can inflict great damage. Control of appraisals is largely in the hands of the employee’s supervisor. Savvy employees know that success hinges in large part on “psyching out the boss.” They also know that when senior executives call for change, the marching orders, if any, will come from their supervisors.
Redesigning Performance Appraisal Systems Is A Sisyphean Task
In short, for political, structural, and systemic reasons, performance appraisal systems cannot function as intended. Worse, they seem to have an almost exclusively negative impact on the very employees they are meant to help. This view of performance appraisals squarely contradicts the mythology of performance appraisal systems. Further, the reasons performance appraisal systems fail to provide the benefits claimed for them seem firmly rooted in the nature of organizations and the behavior of people. Trying to change these factors so that performance appraisal systems will work the way they are intended is truly a modern-day version of Sisyphus’s legendary task.
All things considered, one would hope that performance appraisal systems are relatively inexpensive. This is not the case.
The costs cited by the Internet respondents run the gamut from obviously quantifiable or hard costs to soft costs or costs that are much more difficult to quantify. The preceding discussion can be viewed in large measure as a discussion of the soft costs. The following discussion focuses on the hard costs. These consist chiefly of the direct and indirect expenses associated with the following activities:
- preparing appraisals
- setting goals and objectives
- conducting interim and annual performance reviews
- reviewing at higher levels appraisals written at lower levels
- designing, printing, copying, filing, and distributing appraisal forms
- designing and communicating the appraisal process
- training supervisors, managers, and executives in the appraisal process
- handling post-appraisal appeals, grievances, and lawsuits
A manager who works as a reengineering specialist for a large overseas retailing operation used only the first three cost categories above to derive an annual per-employee cost of $1,945. He then reduced his company’s 130,000 employee count to 85,000 full-time equivalents (FTEs). He further assumed that only 60 percent of those FTEs actually receive a performance appraisal. Even so, his calculations place the cost of his company’s performance appraisals at a staggering $100 million. As he pointed out, these are not out-of-pocket costs, but they do represent the cost of the time used.
Johann Hanekom, CEO of a 500-person subsidiary of a South African telecommunications company, indicated that the labor costs required to recover all operating expenses and burden at his company are pegged at $85 per hour. Using his estimate of hours, the annual per-person cost of performance appraisals at his company comes to $2,200. This amounts to a company-wide cost of $1.1 million per year. (Hanekom’s estimate includes the cost of union participation.)
Hanekom also suggested that the linkage between merit increases and the performance appraisal system contributes to a form of structural inflation. Whether through union bargaining or the uncoordinated demands of people for decent ratings, wages and salaries increase more than would be the case if there were no performance appraisal system or if it were not linked to the annual pay increase. In short, performance appraisal systems provide financial leverage to employees and unions as well as to management.
A reasonable person would be hard pressed to argue that the benefits of performance appraisal systems outweigh their costs. The costs are extraordinary and many of the supposed benefits cited do not withstand serious scrutiny. Of what benefit or credibility, for instance, are career discussions in organizations where downsizing and layoffs are the order of the day?
Performance-related discussions between bosses and subordinates do not require a formal, full-blown performance appraisal system. Indeed, it can be argued that the real coaching and counseling sessions that shape and improve employee performance occur informally, outside such systems. The same can be said of goal setting and feedback. It is also questionable if much of what passes for feedback in formal performance appraisal sessions is deserving of the term. [ For a quick look at how the concept of feedback in the social sciences differs markedly from its technical origins, see my article, “Feedback About Feedback,” in the Human Resources Development Quarterly (Jossey-Bass: Fall 1995). For a scholarly and well-researched examination of the evolution of feedback thought, see George P. Richardson’s marvelous book, Feedback Thought in Social Science and Systems Theory (University of Pennsylvania Press: 1991).] If feedback is viewed as information about actual conditions compared against a set of reference conditions, and if the results are measurable and measured, what role does the manager really play? If they are neither measurable nor measured, what role can the manager play?
The need for a basis on which to allocate the annual merit increase presupposes the need for an annual merit increase. If this is indeed the case, the reasons for it need to be examined. If it serves as a cost-of-living increase, a hedge against inflation, this can be done on a flat-rate basis without the need for a performance appraisal system.
If pay increases are warranted for other reasons, it is unlikely that they require a performance appraisal system to administer. Bonuses or other special increases can and should be tied to very specific, very visible, very measurable results, and this doesn’t require a performance appraisal system. Profit sharing is another case in point.
Poor performers probably constitute less than 10 percent of the work force, so why incur the expense of keeping book on the entire work force? Moreover, keeping book on poor performers does not require an elaborate, formal performance appraisal system. Worse, the books kept on those who are not poor performers can backfire in court. As Craig Brooks wrote, “The typical, traditional performance appraisal is worthless and, in fact, lawyers have told me the appraisal itself quite often is management’s worst enemy in disciplinary grievances and court challenges.” Brooks is not alone. Several people pointed out that performance appraisal systems might increase, not decrease the costs of appeals, grievances, and lawsuits. The legal protection provided by performance appraisal systems seems questionable.
Perhaps the greatest cost of all is that performance appraisal systems silently mock senior executives who call for change. This is embarrassingly apparent when initiative after initiative pleads to have the performance appraisal system changed to support its aims. Such pleas offer compelling testimony that performance appraisal systems are seen as a basic device for getting individuals to comply with the aims of management. This emphasis on compliance is the status quo that such systems maintain, no matter how much they are redesigned. This emphasis is out of place in a world where the ability to elicit contributions from employees matters more than the ability to ensure compliance. As a purely practical matter, given the time lag between changes in the aims of management and changes to performance appraisal systems to support those new aims, it seems unlikely that changes to performance appraisal systems can keep pace.
In conclusion, performance appraisal systems could be eliminated with no harm done and with great economic and emotional benefit. Consequently, change-minded executives should not listen to pleas to redesign their company’s performance appraisal system but should instead give serious thought to scrapping it.
The Internet discussion lists are named below:
- firstname.lastname@example.org (business process reengineering)
- email@example.com (change management)
- firstname.lastname@example.org (consulting)
- email@example.com (human resources)
- firstname.lastname@example.org (learning organizations)
- email@example.com (TQM in manufacturing and service companies)
- firstname.lastname@example.org (training and development)
Rating: 4 out of 5