Lawler + Rewards Management

7/28/2019 Lawler + Rewards Management

1/9

Reward Practices andPerformance ManagementSystem Effectiveness

EDWARD E. LAWLER III

V irtually every organization has a per-formance management system that isexpected to accomplish a number of impor-tant objectives with respect to human capitalmanagement. The objectives often includemotivating performance, helping individualsdevelop their skills, building a performanceculture, determining who should be pro-moted, eliminating individuals who are poorperformers, and helping implement businessstrategies. There is little doubt that a perfor-

mance management system which canaccomplish these objectives can make a verypositive contribution to organizational effec-tiveness, but there is less clarity about whatpractices make a performance managementsystem effective.

There are a large number of design fea-tures that potentially can influence the effec-tiveness of a performance managementsystem, and many of these have been empiri-cally studied to determine their impact. For

example, there is considerable researchwhich shows that performance manage-ment effectiveness increases when there isongoing feedback, behavior-based mea-sures are used and preset goals and trainedraters are employed. There is one potentialdeterminant of performance managementsystem effectiveness, however, which hasreceived relatively little attention: howtightly the results of the performance man-agement system are tied to significant

rewards. The lack of attention to thisimpact is particularly pronounced when itcomes to the issue of using a performancemanagement system to systematically re-move lower-performing employees fromthe organization.

There are a number of reasons for believ-ing that systematically tying rewards tothe outcome of a performance managementsystem will make the performance manage-ment system more effective with respect to

Organizational Dynamics, Vol. 32, No. 4, pp. 396404, 2003 ISSN 0090-2616/$see frontmatter 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.orgdyn.2003.08.007www.organizational-dynamics.com

L E A R N I N G F R O M P R A C T I C E

396 ORGANIZATIONAL DYNAMICS


2/9

motivation, but there are also some that sug-gest it will make it less effective with respect

to development. In a well known 1965 articlethat is based on research done in GeneralElectric Co., Meyer, Kay and French arguethat when rewards are tied to performancediscussions, individuals tend to hear only thereward system part of the message. They donot hear the kind of useful feedback that willallow them to improve their performanceand develop their skills. This has led somecompanies, although not G.E., to separate thediscussion of rewards from the discussion ofperformance.

On the other hand, it is reasonable toargue that when rewards are tied to the out-come of performance appraisals it will leadto more effective performance managementsystems. Managers will be particularly con-cerned about doing a good job, since theoutcome of the appraisal will have a signifi-cant impact on their ability to allocaterewards based on performance and motivatethose individuals who work for them. Simi-larly, in the case of individuals, they knowthat how well the performance review goeswill effect rewards that are important tothem, so they may be particularly motivatedto prepare for the session and see that it goeswell. Further, there is a good possibility thatwhen appraisals are used to determinerewards, organizations will put more pres-sure on managers to differentiate among theemployees they are appraising since this iskey to rewarding individuals for their per-formance.

At this point in time no discussion of aperformance management system can com-pletely ignore the increased popularity of

forced distribution and seriatum rankingsystems. Ranking systems that force man-agers to place their employees in a 1 to Norder based on their performance haveexisted for decades. They are frequently usedin research laboratories (e.g., Sandia, Lawr-ence Livermore) and often place hundreds oreven thousands of employees in a singleranking. Forced distribution systems whichrequire that managers identify a predeter-mined percentage of their employees as poor

performers and outstanding performershave become more popular in the last dec-

ade. General Electrics forced distributionapproach has received a great deal of pub-licity. Other adopters of this approachinclude Intel Corp., Ford motor Co., Good-year Tire & Rubber Co., Enron Corp., CapitalOne Financial Corp., and Electronic DataSystems (EDS) Corp.

General Electrics former chief executiveofficer (CEO) Jack Welch is among the mostvocal and articulate advocates of forced dis-tribution performance management systemsthat force turnover of the lowest-performing

employees each year. At GE, its the bottom10 percent of employees who are supposed tobe eliminated. Mr. Welch feels so stronglyabout this practice that he highlighted it inhis 1999 letter to GE shareholders and advo-cated it again in his bookJack: Straight from theGut. In addition to arguing that a forceddistribution system is good for organiza-tions, he states it is good for individuals,because it takes people who are failing outof situations that are bad for them and thecompany.

There are many criticisms of the forceddistribution approach, including the argu-ments that it represents an unrealistic man-date, and that it may force managers to makejudgments they cannot support with dataand do not believe in. Because of this itcanlead to decisionsthat cannot be defendedwhen there is an adverse impact. On theother hand, its proponents argue that it putsa rigor and discipline into the performancemanagement system that overcomes theall too common leniency tendencies ofmanagers. The key question, of course,

concerns whether or not using a forced dis-tribution system in combination with rewar-ding performance makes a performancemanagement system more effective or lesseffective.

R E S E A R C H M E T H O D

Questionnaire data were collected by mailfrom 55 Fortune 500 companies on their

397


3/9

performance management practices. Theresponse rate to the questionnaire was 32

percent. The typical respondents were headsof human resources (HR) or compensationand benefits for their company.

Survey questions were asked about boththe performance-management practices ofthe firm and the effectiveness of the perfor-mance management system. Performancemanagement system effectiveness was ratedon 17 items. When factor-analyzed, theseitems sorted into two major factors. The firstfactor contained items concerned with theeffectiveness of the system in terms of influ-

encing performance and driving the rightkind of performance behavior. The secondfactor contained items that involved the abil-ity of the system to differentiate between toptalent and performance and poor talent andperformance.

A P P R A I S A L A N D P A Y

C H A N G E S

The results concerning the relationshipbetween the use of appraisals for salaryincrease purposes and appraisal effective-ness are shown in Table 1. The results clearlyshow that when performance appraisalshave a great or very great influence on salaryincreases, they are more effective. This holdstrue both with respect to the performancemanagement system effectiveness items andthe differentiation effectiveness items. Eventhe item focusing on individual skill and

knowledge development shows that tyingappraisal results to salary increases has apositive effect, rather than the negative effectthat was suggested by Meyer, Kay andFrench. Thus, the results strongly supportthe argument that tying appraisal results to

TABLE 1 PERFORMANCE MANAGEMENT SYSTEM EFFECTIVENESS AND

TIE BETWEEN APPRAISAL AND SALARY INCREASES

EXTENT OF TIE BETWEEN APPRAISAL ANDSALARY INCREASESb

EFFECTIVENESSaNO TO MODERATE

MEAN (N 16)GREAT TO VERY GREAT

MEAN (N 39)

Performance Management System Effectiveness (Scale Score) 3.79 4.86*

Overall effectiveness 4.07 4.95*

Motivating performance 3.88 4.64Developing individuals skill and knowledge 3.63 4.49*

Helping the business be successful 4.00 5.00*

Developing a performance culture 3.63 5.08*

Supporting change efforts 3.19 4.32*

Providing useful feedback to individuals 4.50 5.16Supporting business strategy 4.25 5.33*

Supporting company values 4.00 5.51*

Providing accurate measures of performance 3.38 4.51*

Obtaining the commitment of managers who do appraisals 3.56 4.68*

Meeting the needs of individuals who are appraised 3.44 4.36*

Making good use of peoples time 4.00 4.84Linkage of individual performance to unit performance 3.63 5.21*

Differentiation Effectiveness (Scale Score) 3.84 5.12*

Identifying top talent 3.94 5.23*

Rewarding top talent 4.19 5.62*

Identifying poor performers 4.06 5.03Managing out poor performers 3.19 4.53*

a Response scale: 1 not effective at all to 7 very effective.b Response scale: 1 no, 2 some, 3 moderate, 4 great, 5 very great.* Significant difference (p :05) between categories.



4/9

salary increases is a positive with respect tothe effectiveness of the appraisal system.Table 2 shows the results with respect to

tying appraisals to bonuses. Not surprisingly,the results here are very similar to those forsalary increases. There is one difference, how-ever, the differentiation effects in this com-parison do not reach statistical significance,although they are in the direction of showingmore favorable results when there is a strongconnection between appraisal outcomes andbonuses. There is no obvious reason why theresults for differentiation are somewhatweaker on differentiation for tying bonusesto appraisal results than for tying salaryincreases to bonus results. Perhaps becausebonuses often involve larger budgets thansalary increases, and they are not a long-termcost, there is less pressure on managers todifferentiate among employees.

Table 3 shows the results for the relation-ship between appraisals and stock awards.

Again, the results show that having a closerelationship between appraisal results andrewards is associated with a more effectiveperformance appraisal system. The results arecloser to those for bonuses than for salaryincreases in terms of the strength of the rela-tionships. Here, too, the differentiation ofeffectiveness scale difference does not reachstatistical significance. Again, there is noobvious reason why this is true, but a possibleinterpretation is the same one as was offeredfor the non-significant relationship withrespect to bonuses. That is, with stock theremay be less pressure to differentiate amongindividuals than there is with salaryincreases.

Table 4 shows the results for the relation-ship between effectiveness and the degree towhich the lowest rated individuals are ter-minated. The results here show a strongrelationship between effectiveness and ter-mination of the lowest rated individuals.Both the performance management system


TIE BETWEEN APPRAISAL AND BONUSES

EXTENT OF TIE BETWEEN APPRAISAL AND BONUSESb



MEAN (N 34)

Performance Management System Effectiveness 3.95 4.90*

Differentiation Effectiveness 4.41 4.96



TIE BETWEEN APPRAISAL AND STOCK/STOCK OPTION GRANTS

EXTENT OF TIE BETWEEN APPRAISAL AND STOCKb



MEAN (N 24)

Performance Management System Effectiveness 4.23 4.92*

Differentiation Effectiveness 4.44 5.13


399


5/9


6/9

hardly surprising. It fits the common com-plaint against forced distribution appraisals

that too often the focus is on identifyingdifferent levels of performance, rather thanon helping individuals to develop andimprove their performance. In fact, theyoften put managers in the awkward positionof having to find poor performers to dismiss,and may discourage them from developingpoor performers because they know that theyneed to have people in the bottom end of thedistribution in order to meet their quota.

Perhaps the major argument for using aforced distribution approach concerns tying

rewards to performance. It forces managersto identify good and poor performers, thuspotentially leading to a stronger relationshipbetween performance and rewards. This, inturn, has the potential to lead to such positiveresults as higher levels of motivation andmore valid feedback. The key question, how-ever, is does using a forced distributionapproach actually lead to more positiveresults when rewards are tied to the outcomeof performance appraisal?

When those organizations who say theyfocus on tying appraisal results to salaryincrease data are looked at, there is a slightrelationship between the use of forced dis-tribution approaches and the effectiveness ofthe system. The impact is similar to the one inthe overall sample. Performance manage-ment systems are rated as more effectivewhen a forced distribution approach is notused, but the difference is not statisticallysignificant. Differentiation is slightly betterwith the use of forced distributions, but thedifference is not statistically significant.

The same result occurs when the effec-tiveness of performance management sys-

tems is looked at in the case of organizations that have a close relationshipbetween appraisal results and bonuses, andbetween appraisal results and stock options.In both cases, the performance managementsystems are slightly more effective when aforced distribution method is not used, butthe difference is not statistically significant.

The overall pattern is clear. When orga-nizations tie financial rewards to perfor-mance, performance management systemsare slightly more effective when forced dis-

tribution methods are not used. The one qua-lification to this has to do with differentiationeffectiveness. It tends to be slightly betterwhen forced distribution methods are used.

Thirty-three of the 55 companies in thesample to at least some extent tied termina-tion of individuals to their performance man-agement system. Of those, 12 used a forceddistribution methodology in their appraisalprocess. As can be seen in Table 6, theseorganizations had slightly lower scores whenit comes to overall performance managementsystem effectiveness. Further, 12 of the 14individual effectiveness items were lowerfor firms that used forced distribution. Thebiggest difference was in the area of devel-oping individual skills and knowledge. Non-forced distribution companies were signifi-cantly more effective. When it comes to dif-ferentiation effectiveness, there is no realdifference between firms that use forced dis-tribution ratings and those that do not. Thus,the data on the impact of terminating low-rated individuals does not support the use of


FORCED DISTRIBUTION WHEN TERMINATION EXISTS

FORCED DISTRIBUTION

EFFECTIVENESSa YES MEAN (N 12) NO MEAN (N 21)

Performance Management System Effectiveness 4.53 5.07Differentiation Effectiveness 5.02 5.11

a Response scale: 1 not effective at all to 7 very effective.

401


7/9

forced distributions. Indeed, it suggests thatit is better not to use a forced distribution

approach when the results of an appraisalwill be used for terminations.

C O N C L U S I O N A N D

I M P L I C A T I O N S

The results of the study provide clear gui-dance with respect to the impact of tyingperformance appraisal results to changes inpay and employment status. They stronglysupport the view that performance appraisal

systems are more effective when there is aconnection between the results of the perfor-mance management system and the rewardsystem of the organization. This finding,hopefully, will end the debate about whetheran appraisal system is damaged by using it todetermine the amount of someones payincrease, the size of their bonus, or the stockoptions that they receive. It argues stronglythat organizations make a mistake when theyseparate appraising performance from deter-mining pay changes. It also makes it clearthat terminating individuals on the basis ofthe appraisal system is associated with morepositive performance appraisal outcomes.

The present study provides little directevidence on why appraisals are more effec-tive when they affect rewards, but there are anumber of likely explanations for it. First andforemost, when rewards are involved itmeans that a performance management sys-tem is likely to be taken very seriously byboth the appraiser and the appraisee. Itassures that the appraisal will impact theindividuals relationship with the organiza-

tion and potentially his or her compensation.That means he or she is motivated to see thatit goes wellas is the appraiser, who isusually charged with doing a fair allocationof a limited amount of financial rewards.

It is also quite possible that organizationsspend more time developing and trainingindividuals to do performance appraisalswhen financial rewards and continuedemployment are on the line. At the organiza-tional level, it is obvious that when signifi-

cant financial rewards are at stake, they needto support the system and be sure that it is

well executed.The clear recommendation to organiza-

tions that follows from this research is that itis a good practice to tie the results of perfor-mance appraisals to the financial rewardsindividuals receive. There is a considerableamount of literature that suggest tying sig-nificant changes in pay to performance has apositive impact on motivation, and the pre-sent study suggests that it does not decreasethe effectiveness of the appraisal system.Quite to the contrary, it increases it and

makes it more effective as a developmenttool and at getting the organization and indi-viduals committed to doing a good job ofappraising and rewarding performance.

The results, which compare those organi-zations thatuse forced distribution techniquesin their appraisals and those who do not, alsoprovide guidance as to how organizationsshould do performance appraisals. Theyargue that using them does not improve per-formance appraisals and may, in fact, worsenthem.

Forced distribution approaches tend to beassociated with lower effectiveness in generaland, in particular, when appraisal results aretied to termination. There are a number ofpossible explanations for why forced distribu-tions may not have a positive impact, butperhaps the main one is that forcing ratersto conform to a pre-determined distributionleads to them being uncomfortable and, inmany cases, to make bad judgments. It alsomay lead to them being less committed to theprocess and to give mixed messages to theindividuals being appraised. For example,

they may discredit the result by saying, Ihad to rate someone low. Finally, it mayundermine the credibility of the system inthe eyes of those being appraised and thosedoing the appraisal. The net effect of this, un-doubtedly, is a lesser commitment on the partof those doing the appraisal, and ambivalenceor flat-out opposition to the whole process.

From the point of view of the individualsbeing appraised, they may well see the pro-cess as unfair because someone has to be



8/9

rated as a poor performer, and as a result theymay be less receptive to the process and to

hearing the results of it. It can also lead tounhealthy competition among peers. Whenemployees in a work area compete with eachother for ratings, knowing there is always apercentage at the bottom who will be forcedout, it creates fear and selfishness. People aremuch less likely to help each other, train eachother, share information, and operate as aneffective team. In todays flatter, knowledgework-driven, more team-based organiza-tions, excessive internal competition can takea significant toll on organizational perfor-

mance. Postmortems of Enron suggest thatsevere internal competition, which developedas a result of their system, led to its demise.

Legal exposure is also a potentialproblemwith quota systems. Numerous court cases inthe U.S. have challenged the use of forcedranking or distribution systems. In 1975, I wasan expert witness in a class-action age dis-crimination suit against the Sandia Corpora-tion that focused on the validity of its forceddistribution appraisal system. Sandia lost thecase because it couldnt prove that the rank-ings it created led to valid measures of indi-vidual performance. Recently, Ford MotorCompany, reacting to the same problem,abandoned its forced distribution systemafter the company was threatened with law-suits. In 2002, AARP filed a class action law-suit against Goodyear, charging that its forceddistribution performance management sys-tem discriminated based on age. Goodyearimmediately abandoned its rating system.

My quarrel isnt with the premise thatcompanies need to identify and eliminatepoor performers. Organizations certainly

need methods and processes to accuratelyidentify people who are not making thegrade. Setting high standards, and dealingwith those people who dont meet or exceedthem, is necessary in order to continuouslyimprove organizational performance. Theissue is how to do this effectively. Organiza-tions need practices and methods for prop-erly and fairly identifying poor performers,for motivating individuals and developingtheir talents, and for cutting losses for the

individual and the company when someoneis not working out.

Forced distribution approaches are animposed bureaucratic solution to a real per-formance management problem, raterleniency. Often it is hard to get managersto identify poor performers when they knowit may result in their being terminated or notgetting a pay raise. But, in most cases, forceddistribution methods are not be the bestsolution to this problem.

It is important to note that there arealternatives to the forced distributionapproach that usually yield better results.

Some organizations use guided distributionswhich give the raters more flexibility in termsof how many individuals they put in each oftheir categories, and work hard to influencethe fairness of the ratings by training ratersand by having calibration meetings inwhich several raters discuss how they ratedindividuals, why they rated them a particu-lar way, and come to a mutual agreementabout where different individuals fall. Whenthese calibration methods are well facilitatedand based on good data, they can provideresults that are more credible to both theraters and the individuals being rated, andas a result lead to more positive performancemanagement system outcomes.

Overall, the results of this study suggestthat performance appraisal systems can havepositive outcomes. The average scores oneffectiveness are relatively high. This isencouraging since performance managementsystems are often seen as automatic losers fororganizations. The evidence also clearly says,however, that if done incorrectly perfor-mance management systems can be a signif-

icant negative in an organization. Whatconstitutes doing it right? The results suggestthat using it to tie rewards to performance isa significant positive, and that using forceddistribution technology is more of a negativethan a positive, particularly when termina-tions are involved.

403


9/9

SELECTED BIBLIOGRAPHY

The article by Meyer,Kay andFrench is titledSplit Roles in Performance Appraisal andappeared in Harvard Business Review, 1985,43, 123129. Jack Welchs book is Jack:Straight From the Gut (Warner Books, 2001).An article of mine contains an extensive

critique of forced distribution appraisals,The Folly of Forced Ranking, Strategyand Business, 2002, 28, 2833. For a furtheranalysis of what impacts the effectiveness of

performance management see E. Lawler andM. McDermott, Current Performance Man-agement Practices, WorldatWork Journal,2003, 12, 6167. For a review of performancemanagement, see J. M. Smithers (Ed.), Per-formance Appraisal (San Francisco: Jossey-

Bass, 1998.) An overview of rewarding per-formance can be found in E. Lawler, Reward-ing Excellence (San Francisco: Jossey-Bass,2000).

Edward E. Lawler III is distinguished professor of business and directorof the Center for Effective Organizations in the Marshall School ofBusiness at the University of Southern California. He joined USC in 1978and during 1979, founded and became director of the University s Centerfor Effective Organizations. He has consulted with over one hundredorganizations on employee involvement, organizational change, andcompensation and has been honored as a top contributor to the fields oforganizational development, organizational behavior and compensation.He is the author of over 300 articles and 35 books, and his articles haveappeared in leading academic journals as well as Fortune, HarvardBusiness Review and leading newspapers. His most recent books includeFrom The Ground Up (Jossey-Bass, 1996), Rewarding Excellence (Jossey-Bass, 2000), Corporate Boards: New Strategies for Adding Value at the Top(Jossey-Bass, 2001), Organizing for High Performance (Jossey-Bass, 2001),Treat People Right (Jossey-Bass, 2003), and Creating a Strategic HumanResources Organization (Stanford Press, 2003).


Lawler + Rewards Management

Documents

Transcript of Lawler + Rewards Management