organizational rewards: practical, cost-neutral ...kering with the reward system. here's an...

Companies will benefit from rapid, inexpensive, dramatic improvements intheir reward systems from practice of nine principles.

Organizational Rewards:

Practical, Cost-NeutralAlternatives That You MayKnow, But Don't Practice

STEPHEN KERR

T his article is about organizationalrewards. It will not suggest how to spend

additional money you probably don't have,but it will present some principles that willhelp you identify cost-neutral alternatives toyour organization's current reward practices.

I predict that you will agree with, andprobably already know, most of what is pre-sented. Yet, paradoxically, I also predict thatif you look into the matter you'll discover thatyour own organization's reward practices areinconsistent with much of what is presented.Finally, I predict that if you are in a position todo something about it, and do decide to dosomething about it, you will benefit fromrapid, inexpensive, dramatic improvementsin your reward systems, and in the levels ofeffort and motivation of your people.

To some extent, this article is also about thefuture. Despite often-expressed truisms aboutthe pace, scop>e, and discontinuous nature offuture changes, there are stUl some fairly pre-dictable things that can be said about informationtechnology; global trade and sources of labor;organizational hiring and outsourcing practices;and demographics and other societal trends, andtheir Ukely consequences for the successfulimplementation of organizational rewards.

Principle 1: Rewards should be the thirdthing an organization works on; measure-ments should be the second; clear articulationof desired outcomes should be the first.

Principle lA: If you think you have arewards problem that can't be solved, you'rewrong; the problem is with your measure-ments, because anything that can be mea-sured can be rewarded.

Principle IB: If you think you have a mea-surements problem that can't be solved, you'restill wrong; you haven't defined and opera-tionalized what you're trying to accomplish.

LINKING REWARDSTO MEASUREMENTS

Rewards are an easy thing to persuade organi-zations to care about. Reward system consul-tants are eager to describe any number of exdt-ing new incentive and profit-sharingplans—^none of which you have, but all of whichyour competitors are apparently about toinstall—followed by an offer to help get yourreward system ready for the twenty-first cen-tury. Such consultants are usually well received.

Only the most conscientious among them

61

Steve Kerr is Vice President of CorporateLeadership Deveiopment and Chief LearningOfficer for Generai Electric, inciuding responsi-bility for GE's renowned leaderehip educationcenter at Crotonviiie. He was previously on thefaculties of Ohio State University, the Univer-sity of Southern California and the University ofMichigan, and was Dean of the Faculty of theuse business school from 1985-1989.

Kerr is a past-president of the Academyof Management, the world's largest associa-tion of academicians in management. Hiswritings on leadership, "Substitutes for Lead-ership" and "On the Folly of Rewarding A,While Hoping for B" are among the most citedand reprinted in the management sciences.Among his recent publications are TheBoundaryless Organization (Jossey-Bass,1995; coauthor); "Risky Business: The NewPay Game" (Fortune, July 22,1996), and Ulti-mate Rewards (Harvard Business SchooiPress, 1997; editor).

Recent interviews with Kerr have beenpublished in Fortune {November 13, 1995),The New York Times (February 4, 1996),Forbes ASAP (Marcti 10, 1996), Organiza-tional Dynamics (Spring 1996), Index Review(3"" Quarter 1996), Investor's Business Daily(November 19. 1996), Human ResourceExecutive (February 1997), Organization Sci-ence (May-June 1997), and /nfranef (August1997). He has also appeared on ABC, CBS,CNN, NBC, PBS, and cable and local (LosAngeles) TV.

will tell you that you shouldn't upgrade yourreward system until your measurements arereliable. These consultants are usually not wellreceived because, if your organization is typi-cal: (a) you've already revised your perfor-mance management/ assessment/ appraisalsystems more than once in recent years; (b)people seem to hate the new system about asmuch as they hated the ones before; so (c)nobody is interested in suffering through yetanother revision of measurements. Neverthe-less, it is essential that measurements bereviewed, and refined if necessary, before tin-kering with the reward system.

Here's an illustration of what starting atthe wrong end—with the reward system—cando. For many years, one of the world's largestinsurance companies had a mediocre systemof rewards and measurements, of which oneelement was that high performers receivedvery small annual salary increases, and peoplewhose performance was unsatisfactory gotnearly as much. Fortunately, the poor mea-surements did not induce employees tobehave in dysfunctional ways: The rewardswere too insignificant. Unfortunately, the firmdecided to improve its reward system beforeattending to measurements and thus suc-ceeded in making its rewards more powerfuland attractive. This hurt rather than helped thecompany. Many employees established sub-optimal priorities and began to engage in dys-functional competition and to play games withthe numbers: It was now important to lookgood on the (still bad) measurements.

As another example, you may recall thatearly in President Jimmy Carter's term of office,he decided it would be a good idea to install amerit pay system in the Federal Civil Service. Onthe other hand, you're forgiven if you don'trecall this bit of Americana because the ideawent nowhere—not because it was a bad ideabut because it immediately became apparentthat no measurements existed to determinewhich Civil Service employees were performingmeritoriously. President Carter didn't break themeasurement system; they're never was one.

In general, measurements should bethought of as comprising the base of apedestal whose job it is to support the weight

62 ORGANIZATIONAL DYNAMICS

HOW JO OPERAJIOmUZE A MISSION OR AN miAJfVEIn GE's leadership center, Crotonville, we make frequent use of an exercise based on a valuable concept that has becomeknown by many names over tfie years, inclLtding visualization; backward imaging; and most recently by Covey, "start-ing with the end in mind." We ask our students to imagine themselves at a party six months or a year from now, thepurpose of which is to celebrate the achievement of some mission or corporate initiative. We ask them to describe. Invery specific terms, how their leaders', peers', and subordinates' behavior has changed—what they are actually doingmore of and less of—in the future compared to now. We use this process to ensure that our vision and mission state-ments are operational, rather tfian being merely sweet words or numbers. This makes it possible to determine whetherwe have true buy-in, and enables us to articulate roles and responsibilities for all employees in support of the mission,

We then test whetfier the behaviors we have identified are currently measured. For those that aren't, we devise mea-sures, because we know that any behavior that isn't measured can't be systematically rewarded, which means there'sa great chance it won't occur. Through exercises such as these, we teamed in GE that even such Impossible to mea-sure" concepts as empowerment, Work-Out, and boundarylessness could be measured and rewarded, but only afterwe became sufficiently clear in our own minds what we were trying to achieve. Only then could these concepts bearticulated in actionable, operational terms.

of the reward system. The more attractive anorganization's rewards, the more weight themeasurements will be asked to bear.

LINKING MEASUREMENTSTO DEFINITION

Principle IB posits that anything that can bedefined can bie measured. However, this is onlytrue when desired outcomes are defined inoperational, actionable terms rather than meresweet words ("We want to be the best businessin the whole wide world") or numbers {"We willbecome a billion dollar firm by the year 2001").

Although these two examples ("be the bestbusiness" and "be a billion dollar business") mayseem to reflect very different approaches toarticulating a mission or vision, they share threefundamental limitations. First, since it is impos-sible to oppose statements of this kind, no onewill oppose them, making it impossible to dis-tinguish between true and pseudo-buy in. Sec-ond, such statements fail to make clear whateach employee can do to support the mission.While senior executives will probably figure itout, the real proof that a definition is operationalis that supervisors and other low- and mid-levelworkers understand its implications for theirown tasks and responsibilities. If this fails to

occur, performance wil! prove extremely diffi-cult to measure and, therefore, to reward. Seethe box entitled "How to Operationalize a Mis-sion or an Initiative.") Third, expressed in suchgeneral terms, such statements don't differenti-ate the mission or vision from anybody else's.

By way of illustration, sometimes when Iteach in university executive programs, I askparticipants to place their organizations' visionor mission statements in a shoebox. I then say,"I'm going to read one of the mission state-ments. If it's yours, raise your hand." I read thefirst one and a good number of hands go up,including people from an airline, a pharmaceu-tical firm, a plumbing supplies company, andthe U.S. Coast Guard. If I'm feeling sadistic thatday, I also ask participants to estimate tlie num-ber of senior management hours that went intothe statement's preparation. These numbers areoften mind boggling—175; 300; 500—yet thestatements tend to be so similar that most peo-ple can't distinguish their own from others:

If you've operationally defined what youwant and still have trouble measuring it, youshould employ an excellent tool that is proba-bly already in use somewhere within yourorgariization—360-degree appraisal. This tech-nique enables performance data to be gatheredfrom peers, customers, suppliers, and otherswho will be impacted if the goal, mission or ini-

63

dative is accomplished. If none of these peoplecan tell whether you've achieved it (or are pro-gressing toward it; sometimes lead indicatorsmust be identified), then either you reallyhaven't operationally defined it, or you shouldbe asking yourself why you think you want it.

In summary, when a problem seems to berewards-related, organizations should firstensure that they have operationally definedperformance, then ascertain that their mea-sures reflect their definition. They will then bein a position to address their reward systemproblem—although by then they may realizethat the problem wasn't really about rewardsat all, and has inadvertently been solved.

Principle 2: If a reward is unavailable,don't try to use it.

Principle 5: If you make people ineligiblefor a reward, you take away their motivationto strive for it.

AVAILABILITY AND ELIGIBILITY

Having examined the linkages among defini-tion, measurement and reward, let us now con-sider some of the characteristics of rewardsthemselves. First let's look at the most basic ele-ment of any reward—availability. It should gowithout saying that if you don't have some-thing, you shouldn't try to use it, but this prin-ciple is violated all the time. Organizations withvirtually no money to give often spend count-less hours rating, ranking, and grading theiremployees, thereby wasting time, raisingexpectations, and ultimately dispensing suchmeager increases that management is embar-rassed and employees are disappointed. Mypoint is not that financial rewards aren't valu-able; rather, if they aren't available, it's wise toaccept that fact and go on to other things—inparticular, to greater use of nonfinancialrewards, about which more will be said later.

Rewards are often made unavailable byfactors that are beyond anyone's control. (Forexample, since urxiversities are almost neverincorporated, deans normally cannot awardstock options.) Eligibility, on the other hand, isalmost always intentional. Organizationshave traditionally sought to motivate newer


and lower-level employees to aspire to higheroffice. Therefore, such attractive rewards aslarge salaries, profit sharing, deferred com-pensation, stock grants and options, executivelife and liability insurance, estate planningand financial counseling, invitations to meet-ings in attractive locations, and permission tofly first class or use the company plane, aretypically made available only to those whoreach the higher organizational levels. Dosuch reward practices achieve the desiredresults? In general, yes. Residents and internswork impossible hours to become M.D.s,junior lawyers and accountants do likewise tobecome partners, assistant professors publishso they won't perish, and Ph.D. students per-form many chores that are too depressing torecount here to obtain their doctorates.

However, as we look toward the future,the idea that large numbers of employeesshould be induced to lust after the top-leveljobs is fast becoming counterproductive.Numerous organizations have been delayeredand downsized in recent years, leaving themwith fewer positions at the higher levels. Con-sider also the demographic influence of the"baby boomers" and those who followedthem. These employees are now at the age—late forties and early fifties—when they mightreasonably expect to gain access to higher-level rewards. This circumstance of increasingnumbers of mid-level, middle-age employeeschasing decreasing numbers of executive posi-tions will create an unhealthy tension that, ifunchecked, may produce an entire generationof what has already begun to be labeled the"POPOs"— Passed Over, and Pissed Off.

Two additional factors argue for the princi-ple of eligibility. First, when you make peopleineligible for a reward, you take away theirmotivation to strive for it. For example, in moststate lotteries the odds against winning arehuge, yet millions of people buy tickets. Nowsuppose it was announced that people namedSmith were ineligible to win; that is, from thenon if someone named Smith had the winningnumber, another number would be drawn.How likely is it that anyone named Smithwould buy a ticket? The point is that eventhough offering people one chance in a million

is virtually the same as offering them no chanceat all, it generates entirely different mindsetsand patterns of behavior. Awarding stockoptions to even a few previously ineligibleemployees, for example, or inviting a merehandful of non-executives to the executiveretreat, will dramatically alter lower-levelemployees' perceptions of these rewards, andtheir motivation to try for them. (However, asmentioned previously, whether such anincrease in motivation is beneficial or harmful tothe organization will depend on whetherdesired outcomes are competently defined andmeasured. Offering stock options to thirty peo-ple may cause thirty thousand people to wonderwhy they didn't receive any. Thaf s why it's soimportant to address definition before mea-surements, and measurements before rewards.)

Second, when you put people in differentcategories, you give them a different perspec-tive, a different point of view. In today'sincreasingly diverse organizations, rallyingworkers around common priorities and ini-tiatives is hard enough without further subdi-viding employees into exempt vs. nonex-empt, union vs. nonunion, full- vs. part-time,permanent vs. adjunct, tenured vs. non-tenured, and other artificial distinctions.

Looking toward the future, organizationsthat understand the principle of eligibility arebeginning to dismantle some of the connec-tions between rewards and hierarchy, to helptheir employees see that they can reap attrac-tive rewards and have good careers withoutnecessarily reaching the top rungs of the orga-nization ladder. For example. General Electrichas sharply expanded its eligibility criteria forstock options, has gone from narrow to widebanding (from 29 pay grades to 6), and is exper-imenting with such things as knowledge-basedpay, dual ladders, and horizontally focusedcareers, in which employees progress toincreasingly central positions within key orga-nizational processes without necessarily mov-ing upward. Some of these techniques willundoubtedly prove more effective than others,but all constitute ways of distributing rewardswithout relying on hierarchical advancement.

Principle 4: For rewards to be powerful,they must be visible.

VISIBILITY

First and foremost, rewards should be visibleto those who receive them. Most rewards arevisible to recipients but not all. For example,investigation of its high turnover by a largeDetroit firm revealed that people were actuallyleaving for inferior packages. The problem wasthat the firm's benefits were described in suchactuarial doubletalk that even the recipientshad little appreciation of their value. The solu-tion was to create a cartoon booklet that, forthe first time, enabled employees to under-stand how much their benefits were worth.

As another example, some of theemployee benefits at a university I onceworked for were non-contributory. Sincethere were no deductions from salary, thereused to be no entry on the pay stub to showemployees what the university paid eachmonth on employees' behalf. All it took was asimple computer change to make these pay-ments visible.

Ideally, rewards should also be visible toother employees besides the recipient. If aspecial stock grant is awarded to a high per-former, for example, but no one else knowsabout it, the number of motivated people issomewhere between 0 and 1—not an effec-tive use of organizational resources.

Of all the principles of effective rewards,visibility, particularly where money isinvolved, is the one most often violated.Other than public sector organizations thatare required to do so, almost no firms publi-cize their disbursement of financial rewards.This is true even though research has shownthat when employees are asked how they'redoing vis a vis their colleagues, they almostalways report being worse off financially thanthey really are.

The principle of visibility is equally appli-cable to nonfinancial rewards. Many rewardsthat have no monetary component—forexample, inviting a team to share their ideaswith senior management, or recognizing anemployee's contributions outside her homeunit—are usually much more powerful if theyare made public.

Principle 5: If you want someone to per-

65

form, you should reward them when they doperform and not when they don't.

Principle 5A: A good reward says thankyou for the past, and invigorates the future.

Principle 5B: Most human beings makerotten martyrs.

Principle 5C: Never hurt your high per-forniers.

PERFORMANCE CONTINGENCY

In simplest terms, this principle states thatrewards should be based primarily on perfor-mance rather than on other factors. Results fromnumerous field studies and laboratory experi-ments suggest that, for people to be induced todo things they wouldn't do otherwise, a double-digit increase above base—typically 12-15%—isrequired. However, in most corporations, barelyhalf this percentage of total compensation isbased upon performance. In the not-for-profitsector the figure is usually even lower.

As a quick test of the extent to whichfinancial rewards are performance-contingentin your organization, consider that compensa-tion is determined primarily by three factors:What people do (clerk, manager, vice presi-dent; assistant professor, full professor, dean),how long they've done it, and how well they'vedone it. Now imagine that some outsider isallowed to ask two questions and must thenguess how much various employees are earn-ing. Which of these three questions—Whatjob do they have? How long have they doneit? How well have they done it?—is least use-ful in predicting people's pay? In many, manyorganizations the answer would be "Howwell," which is inconsistent with the principleof performance contingency.

Distributing rewards without regard forhow well people have performed makes littlesense in general, and has a particularly nega-tive impact on high performers. Rewards areamong the most powerful tools an organiza-tion has to thank high performers for pastefforts; noncontingent rewards don't do this.Rewards can also play a major role In stimu-lating future performance; noncontingentrewards don't do this either.

Not only does such a system under-reward your best workers, so it doesn't saythank you for the past, it also fails to invigo-rate the future, alienating those who are mostable to find employment elsewhere.

Although organizations never intend tohurt their high performers, the effect of manypolicies is to do just that. For example, across-the-board budget cuts are taken in stride bythe corporate politicians (who build slack intotheir budgets as a hedge against such cuts) andby inefficient managers (who have so much fatin their operations that there's plenty of roomto cut). The real victims are those who run theirunits in a highly effective manner. Similarly,poor performers usually can enroll in courses,attend seminars, and take vacations wheneverthey want, whereas high performers are luckyto get an occasional day off when their work-loads permit. And, in your organization, whois more likely to be offered financial incentivesto retire early? Your t>est performer, or the per-son you're happy is leaving?

Principle 6: A long-deferred reward losesmost of its power.

TIMELINESS

Making rewards contingent upon performanceis unlikely to be effective unless the recipientreceives them, or at least is informed that he willreceive them soon after the action beingrewarded. If a rat in a cage pulls a lever, forexample, and nine months later (on his anniver-sary date) a lump of sugar appears, no learningoccurs. In the same vein, employees who arerequired to wait a long time to be rewarded arelikely to conclude that they probably won't stillbe there, or the boss won't be there, or themoney won't be there, or something bad willhappen in the meantime that will prevent thereward from being received. Even in caseswhere receipt of the reward is not in doubt, thelonger the interval between performance andreward, the less likely it is that the recipient willunderstand that the two are connected.

One primary cause of untimely rewardsis the existence of policies that delay dis-bursements until an employee's anniversary


or some other arbitrary calendar date.Another common cause of delay is the exis-tence of so many sign-offs and one-over-onereviews that by the time the reward is finallyapproved, no one can remember what it's for.(By way of illustration, in the early days ofGE's Work-Out program, a friendly rivalryemerged among the consultants as to whocould find the longest trail of approvalsrequired before a reward could be distributed.I found six in one of the businesses I con-sulted to, but I lost the contest to someonewho found nine. Not surprisingly, simplifyingand rationalizing the approval trails became apriority Work-Out topic in GE.)

Perhaps in the days of tall hierarchies, slowcomputers and large batch processing, it wascost-effective to require multiple sign-offs and totie appraisal and reward cycles to predetermineddates, permitting data from many employees tobe processed at the same time. Given today'sstate of information technology, and increasin^yin the future, it will be possible, and highly desir-able, to more closely connect orgaruzationalrewards with the behaviors being rewarded.

Principle 7: The best rewards are thoseyou can take back if necessary.

REVERSIBILITY

One aspect of being human is that, wheneveryou make a decision, there's a decent chanceyou've made a bad decision. Consequently, anice property of any decision is reversibility—being able to undo an outcome you don't like orat least cut your losses and prevent its repetition.In general, reversible mistakes are not overlyexpensive; irreversible ones can be deadly.

In terms of organizational rewards, undo-ing an undesired outcome means being able toreclaim a reward that's already been given;e.g., taking back a promotion or a company car.If this is impossible, the next best thing is toreverse the decision to give the reward so thatthe same individual doesn't receive it again.

Some rewards are virtually or literallyirreversible. For example, although policymanuals may contain procedures for recap-turing base pay, the combined weight of com-

pany tradition, endless appeals, and moun-tains of paperwork usually deters anyonefrom attempting to do so. Therefore, raisingan employee's salary creates an annuity forhis or her organizational lifetime. Further-more, since future increases are normally cal-culated as a percentage of salary, erroneouslyincreasing someone's pay will tend to becomegeometrically more expensive over time.That's why reversible compensation by what-ever name—bonuses, incentive pay, compen-sation at risk—is such an attractive alterna-tive. Such rewards are consistent with theprinciple of performance contingency and,being reversible, don't commit you to futurepayments unless high performance recurs.Reversible compensation also can serve as akind of shock absorber, making it possible toreduce payroll without taking out people.

Perhaps the biggest problem with reversiblerewards is their tendency to become irreversibleover time, by being perceived by recipients asentitlements. Numerous companies have had toscrap or rebuild their gainsharing, profit sharing,or bonus plans because their payouts became soexpected that the "bonus" or "incentive" com-ponent eventually became just another name forbase pay. In one well publicized case, seniormanagement decided to thank the workforcefor an excellent year by giving every employee aChristmas turkey. They did the same thing ihenext year, but when the third year wasn't verygood they withheld the turkey, only toencounter great outrage from employees whocomplained that they were being denied their"traditional" Christmas turkey. In this case, ittook only 24 months for a generous gift to beperceived as an entitlement!

While some portion of this phenomenonmay be attributable to human nature, organi-zations sometimes bring this upon them-selves, either by avoiding hard decisions andlabeling nearly everyone a high performer orby allowing the non-bonus compensationportion to erode to such an extent that onlyby adding the bonus can the total package besaid to be competitive.

Principle 8: Don't underestimate theimportance of non-financial rewards.

Principle 8A: Stop using the term "reward

67

and recognition"; it implies that "rewards"refer to money, and "recognition" is all thatcute other stuff that organizations do. It is farbetter to speak about financicil and nonfinan-cial rewards.

Depending on if and when you went tobusiness school, you may recall being taughtthat "intrinsic" rewards are more importantthan extrinsic ones: that money is merely a"hygiene factor," not a motivator, and thatwhat human beings truly crave is either "selfactualization" or "job enrichment" but cer-tairJy not anything as crass as money.

That isn't my message. To me the evi-dence is clear that money is, potentially, awonderful reward. Nobody refuses it,nobody returns it, and people who have morethan they could ever use do dreadful thingsto get more. When financial rewards are dis-tributed in ways consistent with the princi-ples we've been discussing, you purchasemotivation and energy to pursue organiza-tional objectives; that's a good investment.

The trouble with money, as notedthroughout this article, is that it is so often dis-tributed in ways that violate most of the prin-ciples. When this occurs, you haven't pur-chased motivation, energy, or anything elseyou can use. Spending money in this case isnot an investment, just a foolish expense.

Viewed in this context, the most attrac-tive aspect of nonfinancial rewards is not thatthey are self-actualizing or intrinsicallyenriching but that their distribution tends tobe consistent with the principles of effectiverewards. This makes them more powerful,and therefore of greater use to most organiza-tions, than financial rewards.

To demonstrate this, let's take a fastinventory of some of the principles we've dis-cussed, starting with availability. We notedearlier that financial rewards are necessarilylimited, and sometimes are unavailable. Onthe other hand, nonfinancial rewards arenever unavailable, because you can literally cre-ate your own supply. You can give recognition,or praise, or performance feedback to oneindividual, then give it to someone else. Youcan give more freedom, challenge, or respon-sibility to someone today, then give her more


of the same next week.You can, if you choose, make all your

employees (or at least, all your nonunionemployees) eligible for nontinancial rewards.You can also make these rewards visible ifyou like, and performance-contingent, andyou needn't wait for high level sign-offs andanniversary dates, because nonfinancialrewards don't derive from the budget or theboss, and are seldom mentioned in employ-ment contracts and collective bargainingagreements. Furthermore, in comparisonwith our earlier discussion with respect toincreases in pay, if you inadvertently givesomeone more freedom or challenge than hecan handle, you can take it back. Therefore,organizations can be bold and innovative intheir use of nonmonetary rewards, becausethey don't have to live with their mistakes.

Is all of this talk of nontinancial rewardsmerely textbook stuff, or does it work in thereal world? To answer this question, let me askyou one. What do the following occupationshave in common: soldiers and sailors; police-men and tiremen; priests and ministers; pri-mary school teachers; registered nurses; rehabtherapists; and volunteer workers who raisemoney for charity? My response is: The finan-cial rewards for doing these jobs are not veryhigh, and the work is not particularly glam-orous. Yet the nontinancial rewards—chal-lenge, responsibility, interesting and meaning-ful work, etc.—tend to be high, and, notcoinddentally, people in these occupations areoften highly committed and motivated.

Principle 9: Get peers, subordinates andcustomers involved in your reward and mea-surement systems

Another vital aspect of successful rewardsystems pertains to who does the rewarding(and the measuring). In a traditional bureau-cracy, this question is easily answered: If youare my hierarchical superior, you do myappraisal and recommend my rewards; if Ioutrank you, I appraise and reward you. In allorganizations, however, opportunitiesabound for peers, subordinates, customers,and others to contribute to the reward andmeasurement process.

In one well-publidzed example involving

appraisal by peers, a division within Wells FargoCorporation distributed Monopoly money toemployees, telling them to "award" these fundsto whichever fellow employee had been mosthelpful in enabling the employee making theaward to achieve his or her objectives. Thecompany then converted this make-believecurrency into real money. For another exampleof peer-based rewards, you have to look no fur-ther than the last World Series or Super Bowl.While the total pool available to the winningand losing teams is based on television revenueand gate receipts, the amount each playerreceives is determined by his fellow teammates,based on whatever objective and subjectivedata they deem appropriate. Many otherawards (e.g. Nobel Prizes, Academy Awards)are also based on judgments by peers.

Customers can also be heavily involved inthe reward and measurement process. Mar-riott hotels, for example, rely heavily on guestsatisfaction surveys when making decisionsabout rewards, and Northwest Airlines hasdistributed cash-convertible certificates to itsmost frequent flyers, inviting them to awardthese certificates to deserving Northwestemployees. As another example, note thatbaseball's starting All-Star teams are selectedby the fans—which is not only an honor, hasfinancial consequences for many players dueto incentive clauses in their contract.

Principle 10: All principles have excep-tions (except this one).

motion, or a trip to Hawaii while others did not.In theory, such pressure is desirable because itcatalyzes explanation and debate. In practice,particularly if your measurements are so unreli-able as to make decisions indefensible, you mayprefer to operate outside the glare of publicity.

Making rewards visible may also be unde-sirable when the recipient wishes to avoidbeing singled out for attention, for fear ofincurring jealousy or disrupting workgroupharmony. In some cases, recipients may wishto appear to be out of favor with management.I once saw a division vice president, followinga successful negotiation, attempt to hug aunion representative on the factory floor, onlyto have the beneficiary-victim angrily wrigglefree of the intended embrace. In this case,while a private expression of gratitude by thecompany officer might have been welcome,public recognition most certainly was not.

TIMELINESS

I once saw a CEO give no more than a per-functory thank you after receiving a briefingby a mid-level manager, but then telephoneher the next day to say "I really appreciatedyour input yesterday." In this instance, the"theatrical pause" created by permitting timeto elapse between the action and the recogni-tion for that action undoubtedly enhanced itsvalue.

EXCEPTIONS TO THE PRINCIPLES:VISIBILITY

At the risk of adding a note of confusion, Iwould like to note that the principles of effectiverewards, like most other canons in life, aresometimes worth violating. For example, weargued that visibility adds to the power ofrewards. This is always true, but it does nottherefore follow that making rewards more vis-ible is always desirable. As previously noted,one key factor ought to be the quality of yourmeasurements. Publicizing the distribution ofrewards piUs pressure on an organization to saywhy some people received a bonus, or a pro-

PERFORMANCE-CONTINGENCY

To reward their sales force, many firms offercommissions in direct proportion to how muchproduct is sold. Such a practice is consistentv ith the principle of performance-contingency,but it offers no incentive for high performers toprovide assistance to low performers. To pro-vide such an incentive, one large consumerproducts company dramatically changed thepayout formula so that all sales people in thesame branch received the same commission,with the size of the pool dependent on saleswithin that branch. The new system was per-formance-contingent at the branch level but

69

not from the standpoint of individual high per-formers This change goes against most of theprinciples pertaining to performance-contin-gency. The new system hurts the high per-formers, and is less effective at thanking themfor past performance. However, it may prove tobe a superior way to invigorate the futurebecause, for the first time, high performers havean incentive to assist low performers. In gen-eral, high performers have the most knowledgeabout how to sell, and also make the most cred-ible role models. If the change doesn't cause thehigh performers to leave or become alienated,the new system may prove to be superior.

CULTURAL AND NATIONALDIFFERENCES

Where reward practices are concerned, it isextremely important to be sensitive and payattention to cultural and national differences.In some cultures, for example, ineligible classesare part of the fabric of society, and it is proba-bly inevitable that reward practices will reflectthese societal distinctions. In other cultures,performance-contingent rewards, at least atthe individual level, encounter philosophicalopposition. A Japanese manager once told me,after my lecture on performance-contingency,that he was offended by my remarks: "Youshouldn't bribe your children to do theirhomework, you shouldn't bribe your wife toprepare dinner, and you shouldn't bribe youremployees to work for the company."

In still other cultures, opposition to perfor-mance-contingent rewards is founded less inphilosophy than in economic realities. In coun-tries where marginal tax rates are very high,employees are generally uninterested in finan-cial rewards, preferring leisure time, access tovacation villages, and any perks that don'tcome with imputed income attached. In thesecountries the reward principle "availability"definitely refers to after-tax availability.

This does not mean that everything we knowabout rewards becomes invalid in other setting.However, it does mean that, in the increasinglyglobal business environment of the 21^* Century,we must be willing to modify the principles in

ways that enable them to provide useful guide-lines in different countries and different cultures.

CONCLUSIONS

Let me close with a few optimistic observa-tions. First, I've argued that your organiza-tion's reward practices are probably inconsis-tent with many of the principles we'vediscussed, so your improvement opportuni-ties are tremendous. Second, I can assure youthat your competitors' practices are at least asflawed as your own (since you obviouslyappreciate the importance of the topic or youwouldn't have read this far.) Therefore,upgrading your rewards can be a very effec-tive way to gain competitive advantage.

Third, 20 years of studying reward sys-tems has convinced me that there is virtuallyno relationship between the power ofrewards and their cost. Throwing additionalresources at this problem is not the answer.The key is to modify your present practices sothat they become more consistent with theprinciples that have been discussed.

Fourth, I'll wager that not only do youknow and agree with most of the principlesdiscussed, you're probably already doingmost of what we've been recommending — athome! When, for example:

• You tell your kids that they can't playoutside until their room is clean

• You suggest to the other couple at thestart of the meal

• that you split the check at the end• You agree to pay the gardener $25 to

do the yard, and another $20 after you inspectthe yard

• You inform your highly competitiveson, who is about to cut the cake, that his sisterwill select the first piece you're putting intopractice many of the principles that have beendiscussed in this article. Organizations arelarger, more heterogeneous, and infinitely morecomplex—but the principles are the same.

•I

To order reprints, call 800-644-2464 (ref. number10502). For photocopy permission, see page 2.


organizational rewards: practical, cost-neutral ...kering with the reward system. here's an...

Documents