Answer based on the article Orgainzational theory and design

OCTOBER 2013 / THE CPA JOURNAL60

By David Bukovinsky

Pay-for-performance (PFP) systems,once the purview of senior execu-tives, are becoming more common for workers at all levels of organizations. These systems provide a win-win situation for organizations and employees. Organizations want to elicit the greatest productivity from workers, and PFP sys- tems reward employees for excellent results. They focus employee attention on the goals and targets established by man- agement; the attainment of these goals translates into organizational success and greater remuneration. Organizations that tie compensation to performance produce bet- ter financial results than those that do not, research has shown. In general, the posi- tive effect increases as more employees are included and as the proportion of their salaries determined by performance increases. Oftentimes, however, programs do not live up to expectations, and results can be disappointing or even detrimental.

Deciding What to Measure For example, consider a real-life

employee who worked in the customer service call center of a major corpora- tion. Workers’ performance was measured based upon the number of calls they answered each hour. When asked by the author what this motivated workers to do, the employee replied, “End the call with the customer as quickly as possible so we can get on to the next call, whether the customer’s problem was solved or not. If they call back, it just further improves our performance statis- tics. Hanging up on them in mid-conver- sation can actually ‘improve’ our perfor- mance. If they call back, it just adds to the number of calls answered that hour.” She also stated that bonuses were based upon the number of calls fielded each hour.

In effect, employees were rewarded for efficiency, not effectiveness—and effi- ciency came at the expense of good cus- tomer service. Needless to say, this is not what management intended; the law of unintended consequences had been invoked. A more effective performance measurement would have considered the number of issues resolved per hour. This illustrates several potential, yet com-

mon, problems with performance-based com- pensation. PFP systems are well intended, but their design, implementation, and results often leave much to be desired. Once the dust set- tles, the heralded system is revealed to be ineffective or, even worse, detrimental.

Addressing Design Issues A fundamental mistake that organizations

frequently make is designing a PFP system in which the measures do not align with organizational goals. As a result, goal attain- ment does not produce greater net income, stock price, or customer satisfaction, as illus- trated by the previous example. Even if the desired number of calls per hour were achieved, it would be unlikely to impact profitability. One could argue that answer- ing more calls per hour would increase cus- tomer satisfaction, resulting in greater sales; however, the number of calls answered is too far removed from customer satisfaction, which is more closely tied to the quality of

Are Pay-for-Performance Systems Missing the Mark?

M A N A G E M E N T h u m a n r e s o u r c e s

Overcoming Design, Implementation, and Operational Challenges

responses that the customers receive. This system, as described by the employee, might actually reduce customer satisfaction. Thus, a good starting point in designing such a measurement system is to ask how, or if, an improvement in a specific activity will impact company profitability or some other strategic goal. The process of design- ing metrics should begin only after that link is understood. The call center example illustrates a sec-

ond common problem with a PFP mea- surement system: poorly designed, con- fusing metrics. In the previous example, management considered an increase in the number of calls answered per hour to represent an improvement in performance. But does an increase in the number of calls handled imply greater efficiency? Or does it just mean that call volume is increasing because more customers require customer service? Ideally, the call center would han- dle fewer calls per hour or even none at all; this would indicate fewer support issues, resulting in a decreased need for customer service capacity. This measure is poorly designed because

it measures efficiency, but not effective- ness. In addition, one can question whether improvement would be indicated by an increase or a decrease in the measure. If managers debate the meaning of a mea- sure, or don’t understand exactly what is being measured, a PFP system will be of little use, and the measurement process will waste resources. Another example of a poorly designed

measure can be found in the 2010 Transocean/BP Deepwater Horizon drilling rig explosion and oil spill in the Gulf of Mexico. The accident resulted in 11 deaths, billions of dollars in fines and cleanup costs, and an immeasurable impact on the two companies’ reputations. Despite this, senior executives at Transocean received significant bonuses for what the company described as its “best year in safety per- formance.” The measurement of safety per- formance was based upon the number of incidents and the potential severity of the incidents; it appears that the total number of incidents was sufficiently low to offset the potential severity of the Deepwater Horizon incident. A third common problem is basing

bonuses on items beyond an employee’s control. For example, an employee might

be able to improve his efficiency and effec- tiveness in answering calls, but only if there is a backlog of calls to be answered. If there are not enough calls, an employee cannot increase the number of calls han- dled per hour. Greater speed in handling calls would only result in more idle time. Similarly, manufacturing employees can-

not be expected to meet quotas if upstream operations are interrupted and the flow of products slows or stops. Company-wide bonuses based upon net income or stock price have little or no motivational force. Most employees work in jobs that are too far removed from such measures and are unlike- ly to be measurably impacted, regardless of how well the employee performs. Fostering a collaborative attitude does

not help, and employees are still likely to remain unmotivated by the plan. From the average employee’s viewpoint, any bonus based upon net income is akin to a random event, rather than a situation that they can influence. A cardinal rule of any measurement system, particularly a PFP system, is that employees should only be evaluated on factors they can control. In most cases, this will translate into localized measures that are directly related to an employee’s duties but that still move the organization toward its goals.

Performance Measurement Criteria Some criteria for evaluating performance

must be established. They may be objec- tive or subjective, both of which have advantages and disadvantages. Objective measures are clear-cut; employees know what performance will be measured and how it will be measured. But objective measures present two problems. First, employees might game the system by focusing on established criteria at the expense of other job duties or, worse yet, falsifying performance data. Second, not all aspects of job performance can be mea- sured by objective, quantitative data. This is one of the main issues confronting school districts wishing to base teacher salaries on performance: do the students’ scores on tests actually measure a teach- er’s ability to convey the joy of learning or to increase interest in subjects? Subjective performance evaluations

allow for the consideration of performance on a broader range of activities. The obvi- ous drawback is that they are subjective.

Because subjectivity requires management discretion in evaluating performance, it opens the door for employees to dispute or distrust the evaluations, and it often results in evaluations based upon personalities and organizational politics. In this case, employees might view the possibility and amount of the bonus to be more political than related to performance. Any PFP system must compare actual

performance to some target. A vague goal (e.g., “do your job better”) is useless, espe- cially if bonuses are to be based upon performance. Vague qualitative goals often result in bonuses awarded solely at man- agement discretion, not because of actual performance. Some quantitative criteria for evaluating performance must be estab- lished. This is where benchmarking against targets comes into play. Performance can be evaluated on an absolute or relative basis. Absolute performance relates to a specified target, which may be a goal (e.g., $X of sales) or a trend (e.g., three contin- uous months of improvements). But where do such targets come from?

For example, why is the goal a 10% improvement, rather than 8% or 17%? The chosen goals are often arbitrary, round numbers. This is not necessarily a prob- lem, but there are some circumstances in which specific goals can create problems. The first is when the goal is beyond

the capability of the employee or system. This will foster resentment and possible backlash as employees realize that no mat- ter how well they work, they cannot achieve the bonus. Employees might view this as an insult. To be offered the possibility of a bonus but have the goal be out of reach (either intentionally or unin- tentionally) is worse than not having a PFP plan at all. Another problem arises when conditions

beyond employees’ control favor some employees over others. For example, a goal of increasing sales by 10% would be eas- ier to reach for a salesperson whose terri- tory is an area that is displaying strong growth than for one whose territory is mired in recession. Setting goals too low is also problematic, because employees might relax their efforts after reaching the goal, especially if the bonus is a fixed amount. Countering this problem with a sliding-scale bonus has its own faults, particularly if there is a cap on the bonus

61OCTOBER 2013 / THE CPA JOURNAL

OCTOBER 2013 / THE CPA JOURNAL62

amount. A sliding-scale bonus will moti- vate continuous improvement only to the point where the bonus is capped. Reaching the cap will reduce employee motivation to improve because greater effort receives no additional reward. In either case, an employee might “bank” further achievable performance gains to be used against the following year’s target. Sales bonuses represent a prime exam-

ple. If a salesperson is relatively confident that she will not meet this year’s goal, she

might intentionally defer sales to the fol- lowing year to improve her chances of a bonus in that year. Alternately, the sales- person could accelerate sales or “stuff the channel” at year-end to meet the goal, knowing full well that some of the sales will be returned in the following year or will reduce her sales for the following year. Customer relations might be damaged if salespeople imply that customers should buy now because of “possible” price increases in the future, ship unordered goods to customers, or delay delivery of purchases. Relative performance compares one

employee’s performance to another or to some other metric. For example, salesper- sons at a car dealership might be compared against each other, or the productivity gains of a manufacturing department might be compared to industry-average productivity gains. Care should be taken with systems

that compare employees or groups to each other. For example, at an educational institu-

tion that the author is familiar with, bonus- es are based upon a faculty member’s performance rating, compared to the aver- age rating of all other faculty members in the same organization. Three problems are evident. First, the individual has no con- trol over the performance of others; supe- rior performance by the individual might not receive the recognition it deserves if other employees are also performing well. The second problem is a corollary to the first: a mediocre employee might be rewarded for being the “best of the worst” if other employees are performing at low levels. Similar issues arise when comparing performance to an external benchmark like an industry average. The final problem is the lack of a consistent means of rating employees. Because department managers rate their own employees, workers in a department with a diligent manager who gives honest evaluations are at a disadvantage when compared to employees in a department with a manager who maintains his popu- larity by giving everyone high ratings, regardless of whether they are deserved. Relative performance measures might also motivate employees to sabotage the work of others to make their own performance appear better. The period covered by the performance

appraisal must also be considered. The ubiquitous annual evaluation feeds into the criticism that most organizations, especial- ly publicly held ones, are too narrowly focused on annual results. Rewarding employees for reaching short-term goals can have detrimental long-term impacts. For example, discretionary costs represent the low-hanging fruit employees and man- agers will often reach for. Consequently, cutbacks are often focused on such areas as employee training, research and devel- opment, customer development, and other activities that create long-term value for an organization. Motivating employees to achieve short-

term goals also forces them to bypass projects with long lead times before ben- efits are realized. Short-term goals might foster an “I’ve met this year’s goal, so I’ll save my other ideas for next year” men- tality. As a result, improvement projects

are not implemented as quickly as possi- ble and results are deferred until later periods. Creativity and risk-taking are sti- fled. Moreover, achieving short-term goals might not add up to long-term success. Longer time horizons can reduce these

problems, but they can also create other ones. How long should the evaluation period extend? If it is too short, the problems asso- ciated with annual performance reviews will not be mitigated. If it is too long, an employee might not be overly concerned because she believes she will have moved on to another position within the organization or to another organization entirely before her performance is evaluated and rewarded. Longer time horizons also force employees to accept deferred gratification, which might reduce employee satisfaction and, conse- quently, performance.

Understanding the System The bonus allocation formula in the edu-

cational institution example above illus- trates another common problem with PFP systems—understandability. The complex allocation formula included variables, vari- ables with subscripts, and Greek symbols. Some employees in the organization could, with a little analysis, understand how their bonus amounts were calculated. Others found the calculations indecipher- able and, thus, could not understand the linkage between their performance and the resulting bonus. If the measurement system and bonus

allocation scheme are unintelligible to employees, it will not produce the intend- ed effect because it will be ignored. Complexity does not necessarily translate into effectiveness. The same can be said if employees do not understand why per- forming certain activities, or performing them at a certain level, is important. Job responsibilities should not have meanings that are hidden from the employees. Instead, employees must understand why something needs to be done and why it is important to do it well. Additional challenges occur when PFP

systems are used with teams. Teams can outperform individuals because they take responsibility for an entire project, rather than just some portion of it, and because team members have a common goal. This can motivate them because they can see how their efforts contribute to the end

Motivating employees to

achieve short-term goals

also forces them

to bypass projects with

long lead times before

benefits are realized.

result. There is also social pressure to per- form well and not let the team down. Ideally, these two factors will steer indi- viduals toward performance that is in the best interest of the team and organization. The primary, and obvious, problem arises when individuals do not contribute even- ly. As a result, other team members must pick up an underperformer’s duties to successfully complete a project. Under such circumstances, basing compensation upon team performance becomes problematic. The free rider is rewarded based upon team performance, not on his own merit; thus, such individuals are overcompensated and the remainder of the team is under- compensated for their efforts. The composition of a team can also

affect the performance of otherwise com- petent employees. Team dynamics might bring out the best or worst in the team members. At best, more experienced or knowledgeable members can help improve the performance of others. At worst, per- sonality conflicts, politics, or other rifts can result in intended or unintended interfer- ence with member performance. Team-based compensation schemes

might also affect mobility and movement in and out of teams. A high-performing team might resist the addition of new mem- bers if they believe it will weaken the team and consequently reduce their compensa- tion. Alternately, members who believe the rest of the team holds back their perfor- mance might be motivated to leave one team for a stronger one. The combination of these two tendencies can eventually result in a concentration of excellent per- formers in certain teams, leaving other teams populated with weaker performers.

Implementation and Operational Issues Although design plays a large role in

PFP systems, other issues arise during a system’s implementation and operation. One of the most important steps in the implementation stage is communication. Employees must understand what is expected of them, which activities will be evaluated, how those activities will be eval- uated, the goals their performance will be judged against, and how the PFP system impacts their compensation. Poorly designed systems allow for unin-

tended consequences. In the call center example, management unwittingly moti-

vated the employees to do something detri- mental. Admittedly, the employees in that situation consciously acted in a way they knew was counter to management’s intention. Poorly designed systems also open the door for outright manipulation, especially when recurring, short-term tar- gets are used. As previously noted, the employees were not only performing these injurious activities, they were being paid to do so. But that is not always the case. Employees

might act in good faith, trying to improve performance, only to later discover that their well-intended actions caused undesirable out- comes. Management must be alert to these possibilities and take steps to prevent them. Good communication can alleviate many of the unintended negative consequences that can result from the actions of well-inten- tioned employees. Intentional gaming of the system for per-

sonal gain is more difficult to control. Expectations of ethical behavior that per- meate all levels of the organization and are constantly reinforced, upheld by example, and enforced when necessary might serve to overcome some of the temptation to com- mit self-serving actions. An ethical envi- ronment is more likely to align employee actions with the organization’s well-being. A lax environment allows employees to revert to their own self-interest. The composition and size of bonuses can

also be an issue. The type of bonus—cash, additional time off, trips, prizes—might be important to employees. Different people are motivated by different means. Regardless of its composition, the bonus pool in a PFP system must be of suffi- cient size to motivate employees. Employees will not respond to a system in which the potential bonuses amount to lit- tle more than a pat on the back. The author dealt with a situation in

which the difference between being a supe- rior employee in an organization of infe- rior performers and being an inferior employee in an organization of superior performers was virtually nil. Using some hypothetical data in the organization’s bonus allocation formula, the author found that the difference amounted to approxi- mately $50 per month for employees whose annual base salaries ranged from $70,000 to $100,000. Obviously, there was little monetary reason to perform at a

high level. It is also worth noting that employees working long hours or in excep- tionally stressful jobs, such as hospital emergency room staff or CPAs during busy season, might be more motivated by addi- tional vacation days or some other form of nonmonetary compensation. Allowing indi- vidual employees to choose their type of bonus compensation can go a long way in making the PFP system successful.

Implementation Suggestions PFP systems can be powerful motivators

and an important component of a continu- ous improvement system; however, such

plans are not without pitfalls. Management must carefully consider what it hopes to achieve by rewarding performance, and it must design systems to promote desired behaviors consistent with organizational goals. The PFP system must be based on activities within an individual’s control that can be evaluated with some degree of cer- tainty and can be compared to some bench- mark. Finally, the amount and type of com- pensation awarded must be sufficient to motivate employees to perform as desired. PFP systems are very effective in theory, but are not as easy to implement in reality as is often assumed. q

David Bukovinsky, PhD, CPA (inactive), is a professor of accountancy at Wright State University, Dayton, Ohio.

63OCTOBER 2013 / THE CPA JOURNAL

The PFP system must be

based on activities within

an individual’s control

that can be evaluated

with some degree of

certainty and compared

to some benchmark.

Copyright of CPA Journal is the property of New York State Society of CPAs and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.