当前位置:首页 >> 经管营销 >>

Do Financial Incentives Drive Company Performance An Evidence-Based Approach to Motivation an


HAR VA R D B U S I N E SS S C H O O L P R E SS

Do Financial Incentives Drive Company Performance?
An Evidence-Based Approach to Motivation and Rewards

E xc e r p t e d fro m

Hard Facts, Dangerous Half-Truths, and Total Nonsense: Pro?ting from Evidence-Based Management
By

Jeffrey Pfeffer, Robert I. Sutton

Harvard Business School Press Boston, Massachusetts

ISBN-13: 978-1-4221-2565-6

2565BC

Copyright 2007 Harvard Business School Publishing Corporation All rights reserved Printed in the United States of America This chapter was originally published as chapter 5 of Hard Facts, Dangerous Half-Truths, and Total Nonsense: Pro?ting from Evidence-Based Management, copyright 2006 Harvard Business School Publishing Corporation. No part of this publication may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form, or by any means (electronic, mechanical, photocopying, recording, or otherwise), without the prior permission of the publisher. Requests for permission should be directed to permissions@hbsp.harvard.edu, or mailed to Permissions, Harvard Business School Publishing, 60 Harvard Way, Boston, Massachusetts 02163. You can purchase Harvard Business School Press books at booksellers worldwide. You can order Harvard Business School Press books and book chapters online at www.HBSPress.org, or by calling 888-500-1016 or, outside the U.S. and Canada, 617-783-7410.

5
Do Financial Incentives Drive Company Performance?

T

here’s a lot of money in ?nancial incentives. Type the word compensation into Amazon.com and you get more than 47,000 book entries, with some 43,000 entries associated with the word incentive. Hordes of people are clearly interested in reading—and writing—about compensation and incentives. There are large compensation consulting companies such as Towers Perrin, Hewitt Associates, Mercer, and Watson Wyatt, plus scores of small ones, that make good money selling advice about how to design incentive systems that attract, retain, and motivate employees. Human resource executives devote huge chunks of time to designing pay systems and dealing with unusual cases and complaints from employees, contract workers, and consultants about compensation issues. Compensation committees of boards of directors devote vast energy to installing the right incentives for senior executives. These efforts typically involve “aligning the interests of executive of?cers with the long-term interests of the company’s stockholders.” 1 These tremendous efforts to “get the pay system right” are guided by several deeply held, widely shared, and intertwined beliefs and assumptions about

2 Dangerous Half-Truths About Managing People and Organizations

what motivates people in the workplace. Incentives are seen as the primary tool for aligning individual behavior with organizational objectives, because without effective incentives people would do nothing—the technical term from the economics literature is shirking, and the assumption is that work is aversive so people must be bribed to expend any effort. Or if they did expend effort, people are presumed to be almost certain to do things that undermine organizational or management goals. Underlying all this is the belief that people work primarily for money, and that because motivation is the most important factor affecting individual task performance, ?nancial incentives are the most important of all motivators. It follows that getting the incentive scheme right is critical for organizational success, for both motivating and aligning behavior. There are hundreds of books on pay in organizations that make exactly these points and emphasize the critical role of pay in determining organizational performance. The unremarkable and seemingly obvious assumption is that “basic to the effective functioning of any organization is its pay and reward system.” 2 This emphasis on the importance of ?nancial incentives is scarcely new. It goes back at least to the time of Frederick Taylor, the founder of scienti?c management at the turn of the 20th century. Taylor wrote in 1911, “What workers want most from their employers beyond anything else is high wages.” 3 In his classic pig iron handling experiments, ?nancial inducements based on productivity were used to persuade workers to accept scienti?c management’s prescribed methods. Pretty much the same view of ?nancial incentives is also seen in many well-regarded economics theories, as well as in?uential psychological theories. In economics, “a cornerstone of the theory in personnel economics is that workers respond to incentives . . . it is a given that paying on the basis of output will induce workers to supply more output.” 4 In psychology, Skinnerian learning theory argues that behavior is a function of its consequences, so if you want more of some behavior, like hard work, that behavior needs to be reinforced. And most operant conditioning theorists who study and intervene in organizational settings treat ?nancial rewards as the most potent form of reinforcement.5 Decisionmaking theory makes somewhat more complex assumptions about human behavior than reinforcement theory. But that theory also presumes that people choose actions—or at least want to choose actions—on the basis of the expected probability of obtaining valued outcomes.6 People generally desire more money, so if they believe that working harder will result in getting more money, they will expend the effort. In short, the belief that ?nancial incentives are the most powerful drivers of organizational performance

Do Financial Incentives Drive Company Performance?

3

is seen in a host of enormously in?uential and widely held theories about human behavior in work settings. The problem is that these basic assumptions about ?nancial incentives and how they work are just that—assumptions. They are usually taken on faith rather than tested or even subjected to critical thought. The result is that companies build complicated and expensive incentive schemes that routinely fail to produce the behavior that leaders want or ever intended. But since companies don’t base their designs on sound evidence or logic, when they try to “?x” the system with “new” solutions that seem different on the surface, they often end up with solutions (and problems) that are nearly identical to the old ones. Or they change from one misguided plan to another with remarkably little learning along the way. This chapter provides ideas, data, and perspectives to help you think constructively about ways to stop, or at least reduce, the problems that so many ?nancial incentive systems in?ict. We do so by examining the evidence and underlying logic that form the foundation of ?nancial incentive systems.

What Incentives Can Do
There are three primary avenues through which incentives can enhance organizational performance, or if badly designed or misapplied, damage performance. First, ?nancial incentives could motivate more effort—a motivational effect. This is the effect usually sought and assumed when companies and consultants recommend instituting pay-for-performance schemes— people will work harder to achieve a greater ?nancial reward. Increasing people’s motivation can’t affect their ability (at least in the short run), only their effort. So interventions that focus on increasing effort presume, by de?nition, that if people would only try harder, they’d get better results. Yet interventions aimed at increasing motivation through incentives can be effective only if people have enough information to perform their work effectively and if other organizational systems and technologies are not the primary roadblocks to poor performance. Compensation consultants rarely acknowledge these limits of motivation, perhaps even to themselves. But most of the interventions they recommend presume that greater effort alone will bolster performance—without system redesign, information sharing, or upgrading people’s skills. Interventions that use money to drive motivation also presume that job performance is under the control of the people who are given the incentives, and that individuals’ actions then drive organizational performance. But

4 Dangerous Half-Truths About Managing People and Organizations

swings in performance aren’t always under an executive’s or an employee’s control. Take the case of a senior executive from Florida Power and Light, who told us, while attending a Stanford executive program, that his compensation was based on the pro?tability of the utility. The utility’s pro?tability, since in the short run most of its costs and rates were ?xed, depended mostly on the amount of electricity sold, and the amount of electricity sold depended mostly on the temperature. The hotter the summer in Florida, the more power was sold, and the more pro?table was the utility. That summer was a particularly hot one in Florida, so the executive got a big boost in pay during the month that he spent at the Stanford Executive Program in California. This executive noted that this incentive system made no sense— unless you believed he could control the weather in Florida. Incentives are effective motivators only when certain assumptions hold— including the idea that differences in employee ability and knowledge do produce performance differences—and when performance outcomes are under the control of people who receive the incentives. If these assumptions don’t hold—performance outcomes aren’t controllable and employee efforts don’t make a difference—then the motivation that comes from greater ?nancial incentives can’t and won’t positively affect performance. Instead, ?nancial incentives can undermine motivation and performance as a result of the frustration, unhappiness, and dismay that people experience when working harder in a system that makes it impossible for them to effect performances— even when they are getting rich at the moment, like that executive from Florida. This point about the futility of expecting better performance just by trying to bolster individual motivation has been made repeatedly by W. Edwards Deming and other writers in the quality movement, but is still forgotten by many executives and their advisers.7 Second, ?nancial incentives can provide people with information about what the organization values and what its priorities are, an informational effect. In a world where people can’t possibly give equal attention to every dimension of their jobs, and where companies often send con?icting messages about priorities—for example, pay attention to quality or customer service but also cut costs and increase ef?ciency—people tend look to the pay system to ?gure out what really matters to senior leadership. When Continental Airlines undertook a cultural and service transformation in the 1990s, going from worst to ?rst in on-time performance in a year, the company paid each employee $65 for every month that Continental ranked in the top half of the Department of Transportation rankings for on-time performance. Not only did this ?nancial reward motivate people to try harder, it signaled

Do Financial Incentives Drive Company Performance?

5

to employees that on-time performance was something Continental actually cared about, as did posting monitors that displayed the proportion of ?ights arriving on time each day. Research on customer service shows that sending clear signals can have powerful effects. One analysis of airline on-time performance reported that whether or not airline executives seemed to care about ?ying on time was among the most critical differentiating factors in whether airlines actually ?ew on time.8 The third way that differential ?nancial rewards may drive performance is that they presumably attract the right kind of people and repel the wrong kind, a selection effect. Think of recruits who face the choice between working for one company that offers performance-based pay and another one that offers pay based more on seniority. The idea is that motivated people who are driven to outdo their peers will choose workplaces where their superior performance will translate into more money in their pockets. And people who are less able or motivated will seek out less competitive workplaces where their lower performance will be penalized less. One argument that you hear about government or other quasi-civil service systems that don’t reward performance is that they can’t attract the best and most ambitious people; their failure to offer differential ?nancial incentives undermines the ability to attract and retain the best—presumably the most ?nancially ambitious—people. Stanford economist Edward Lazear believes that this selection effect is as important to organizational performance as any effect on motivation. “Pay that is only mildly related to output can be very powerful in sorting workers and providing information.” 9 As you will see in this chapter, each of these three mechanisms operates in every organization, but each also has unanticipated effects on the people it is meant to motivate, inform, and attract—effects that dampen performance with alarming frequency. Even when executives have the best of intentions, take great effort to study best practices, and bring in top compensation consultants, many still end up with pay systems that undermine performance— which makes the idea that ?nancial incentives drive company performance a particularly dangerous half-truth. Bad compensation systems are so damaging, in large part, because people use pay as a signal of whether the organization values them—their status and importance—and to ?gure out whether or not they are being treated fairly. Status and fairness both matter a lot to people. So making mistakes in pay can cause people to withhold discretionary effort, ideas, and information—and can fuel unwanted turnover. Financial incentives have a potent impact on performance, but not necessarily in the positive ways that executives and their advisers anticipate.

6 Dangerous Half-Truths About Managing People and Organizations

The Growth in Incentive Pay
There is no question that incentive pay is now ubiquitous and that the use of incentive-based pay systems has grown in recent decades, spreading from the United States to companies around the world. Even as early as the 1980s, surveys showed that over 80 percent of employees worked in organizations with merit pay plans, in which at least some employees received raises based on their rated performance.10 The prevalence of incentive or contingent pay has increased at all organizational levels during the last 15 years, with incentive schemes such as bonuses becoming particularly pervasive at more senior executive levels.11 Hewitt Associates, a compensation consulting ?rm, reported that in 1991, 51 percent of the companies participating in its salary survey offered at least one pay-for-performance plan. By 2003, that number had increased to 77 percent, after peaking at 81 percent in 2001. And 50 percent of the companies in 2003 had variable compensation plans that covered virtually every employee.12 Nor is this trend con?ned to the United States. A 2003 Hewitt survey of 115 Canadian organizations found that 81 percent offered pay-for-performance plans, up from 43 percent in 1994.13 There is also growing interest in using merit pay in noncorporate settings such as schools and government, places that have traditionally not used contingent pay. In Albuquerque, New Mexico, garbage truck drivers were put on an incentive pay plan, and more than $4 million was paid out to the 180 unionized drivers over a six-year period.14 Denver schoolteachers are subject to a pay-for-performance system that rewards them for their students’ progress, while in Houston, every school employee receives a bonus based on students’ test scores. In Florida, school districts are now required to create salary systems that reward teachers for student performance.15 In the U.S. government in 2003, the administration proposed having the Of?ce of Personnel Management administer a $500 million human capital performance fund to allow federal agencies to institute pay-for-performance practices. And the General Accounting Of?ce issued a report praising governmental efforts to rely more heavily on ?nancial incentives.16 In this spirit, the Of?ce of Homeland Security unveiled a new incentive system that did away with traditional seniority-based pay in January 2005. Of?ce of Personnel Management Director Kay Coles James boasted to reporters, “We really have created a system that rewards performance, not longevity . . . It can truly serve as a model for the rest of the federal government.” 17 The presumption is that merit pay ?xes performance problems, so incentive pay becomes a cure for any organization that might have them, including schools or government agencies.

Do Financial Incentives Drive Company Performance?

7

Why Money Is Used So Much: We Think Others Are Motivated by Money, Even If We Are Not
The growing use of variable pay and other ?nancial incentives and the belief in their effectiveness stems in part from something that researcher Chip Heath calls an “extrinsic incentives bias.” This is the tendency to overestimate how much employees care about extrinsic job features such as pay and to underestimate how much employees are motivated by intrinsic job features like being able to make decisions or having meaningful work.18 Heath’s research shows that individuals believe that others are motivated by money, even as they know that they are much less so. ? A survey by Kaplan Educational Centers of almost 500 prospective lawyers preparing to take the Law School Admissions Test revealed that 64 percent of the respondents said they were pursuing a legal career because it was intellectually appealing or because they were interested in the law, but only 12 percent thought their peers were similarly motivated. Instead, 62 percent thought that others were pursuing a legal career for the ?nancial rewards. ? In data collected from surveys over a 25-year period, respondents to the General Social Survey (a representative sample of people in the United States) rated “important work” that “gives a feeling of accomplishment” as the most important aspect of their jobs, with pay typically ranking third. But when people were asked about others, 73 percent thought that large differences in pay were necessary to get people to work hard and 67 percent agreed with the statement that people would not be willing to take on extra responsibility at work unless they were paid extra.19 Heath conducted an experiment demonstrating that people consistently overestimated how much other people were motivated by pay. Heath found that “participants listed an extrinsic incentive in the top position for themselves only 22% of the time, but they predicted that the [customer service representatives] would list an extrinsic incentive in the top slot 85% of the time.” 20 Participants in the experiment would have done a much better job predicting the motivation of others—and thereby earning a reward for their accuracy—had they simply extrapolated their own feelings about the importance of ?nancial incentives to others. Other evidence bolsters Heath’s ?ndings that management places excessive faith in the motivational magic of extrinsic rewards. A Watson Wyatt

8 Dangerous Half-Truths About Managing People and Organizations

2003–2004 survey of 1,700 high-performing employees—as identi?ed by their employers—from 16 organizations found that these top performers rated a desire to maintain a positive reputation as the most important factor in their motivation. These top performers ranked being appreciated second, belief that the work is important third, interesting assignments fourth, and expecting a signi?cant ?nancial reward ninth out of 10 items.21 A survey of 205 executives from diverse industries found that 68 percent reported their companies had executive bonus plans because senior management believed that such rewards would motivate executives.22 These same executives reported, however, that they did not make daily business decisions based on how such decisions would affect either their bonus or those of their people. Organizations ought to offer members suf?cient—and correct—inducements to motivate and direct their efforts, and to attract the right people. Heath’s experiments and other survey data “suggest that lay theories of motivation may hinder this process” by overemphasizing ?nancial incentives and underemphasizing the importance of work and its intrinsic interest.23 Heath’s ?ndings and the other evidence we’ve presented help explain why managers continue to rely too much on extrinsic rewards—they have ?awed assumptions about how others are motivated. Moreover, even when incentive systems do work more or less as intended, we may not like the consequences.

Incentives Signal What Is Important, but the Signals May Be Blunt
Incentives and measurements provide information, not just motivation, and the effects of the information alone on motivation can be pronounced. A classic demonstration of the power of external reinforcements was a study in the early 1970s at Emery Air Freight, a freight forwarder. Before the development of large package companies with their own airplanes, freight forwarders picked up packages and shipped them on airlines. They got a better rate to the extent the packages were placed in larger containers that were easier to handle. So Emery management wanted employees to put as many packages as possible into larger containers to cut freight costs. The company conducted a performance audit and found that, although managers thought they were using larger containers 90 percent of the time it was feasible, only 45 percent of the eligible packages were actually being put into larger containers.24 So the company announced a new program that provided rewards such as praise—not ?nancial rewards—for improvement. On the ?rst day, the

Do Financial Incentives Drive Company Performance?

9

proportion of packages placed in the larger containers increased to 95 percent in about 70 percent of the company’s of?ces.25 The speed of this overwhelming improvement suggests that the change in performance derived not just from the rewards that were offered, but also from the information provided that the current performance level was poor and this action—consolidating shipments—was important to the company. The intervention showed people how well they were doing, whereas before they didn’t have such information. It also helped people realize the importance of this speci?c dimension of their job. The Emery Air Freight experience demonstrates that letting people know how they are doing and what is important can have substantial effects on their behavior, even in the absence of ?nancial incentives. Unfortunately, signals from ?nancial incentive systems about what behaviors the organization values are not always what the company may desire or need. One day in August 2003, Jeffrey Pfeffer and his wife Kathleen went to look for a new car. They were inclined to buy a Toyota, but had heard good things about the Mazda 6 and the Nissan Altima, so they wanted to test drive all three on the same afternoon. They ?rst went to Putnam Toyota in Burlingame, California, where they were greeted by a typical, commission-based car salesman. When they explained what they were doing—test driving three vehicles to decide which they would choose—the salesman immediately concluded, correctly, that they were unlikely to purchase a car that afternoon. He didn’t want to waste his time with Jeffrey and Kathleen because he was paid on commission—a signal that the company believed his primary responsibility was to move metal. So the salesman sent them to the garage where the company’s new cars were housed, about a block away. When Jeffrey and Kathleen arrived, they learned that they needed a salesperson if they were going to test drive a car. This salesman achieved his goal—to get rid of two customers who were not ready to buy immediately. When Jeffrey and Kathleen ?nally decided to purchase a Toyota Camry, needless to say they did not return to Putnam. They bought it from Toyota 101 in Redwood City, California—a dealership where the salespeople are not paid on commission and the emphasis is on customer service. Putnam Toyota and many other organizations use pay systems to signal what is important, and the signals are not completely misguided. Car dealers want salespeople to generate revenue and not stand around and waste time. So the Putnam salesman did precisely what the pay system signaled him to do—he didn’t waste time with people who had a low probability of generating revenue immediately.

10 Dangerous Half-Truths About Managing People and Organizations

There is just one problem. The many companies that use similarly crude or blunt incentive systems aren’t interested solely in maximizing revenues during a single customer encounter. Having invested in advertising and other promotions to get people into the stores in the ?rst place, their leaders are interested in maximizing sales over long periods and, even better, building customer loyalty so that less marketing and advertising expenses are required to get people to return to their establishments. But using the wrong incentive systems sends employees signals that clash with the companies’ overall objectives. Consider another example. Marshall Industries was a $500 million electronics distributor before a major transformation and cultural overhaul turned it into a $2 billion competitive powerhouse in a few years. Prior to the transformation, which entailed eliminating sales commissions and other individual bonuses, people looked to the incentive system to learn what was important and behaved accordingly. Here are some of the bad results listed by then-CEO Robert Rodin: ? Our salespeople would ship ahead of schedule to make a number or win a prize. Our customers, on the other hand, were insisting on delivery in a window of one day early to zero days late. ? We held customer returns. We had to make sure that the returns coming in did not get counted against sales in the period for which we were trying to hit the numbers. So, if a customer returned items, sometimes our salespeople would put them in the trunks of their cars. ? We opened bad credit accounts. Any order was a good order as far as a salesperson paid on gross pro?t was concerned.26 Financial incentives do signal what is important and do focus people’s attention on those dimensions. But this is both good and bad news. It is good news in that incentives can be powerful in shaping behavior, but it is bad news if management doesn’t fully understand the implications and subtleties of the behavior shaped. The problem is that the typical ?nancial incentive system is too blunt and narrow a way of communicating what is important, unless the company has a very simple business model—where one or just a few behaviors matter. Incentive systems do have to be simple to be effective; people can only keep a relatively small number of things in their heads at any one time, so incentive schemes with multiple criteria are too complex to send straightforward signals that guide behavior. But simple signals cause damage when there are multiple, interrelated dimensions of individual performance—

Do Financial Incentives Drive Company Performance?

11

when judgment and wisdom are required to ?gure out the best ways to enhance overall organizational performance.

Incentives Motivate—Sometimes the Wrong Behavior
There is no question that ?nancial incentives motivate people and, under the right conditions, can drive big increases in performance and productivity. Take, for example, Safelite Glass headquartered in Columbus, Ohio, one of the largest installers of automobile glass in the United States. Stanford economist Ed Lazear did a statistically sophisticated, detailed study of Safelite over 19 months when, under a new CEO and president, the company gradually moved from using hourly wages to paying employees based on how many windshields they installed.27 Because the company carefully tracked output per employee with a sophisticated computer system, and because some employees worked under both payment systems, Lazear obtained precise estimates of the effects of the new variable pay system on individual productivity and on turnover. Lazear estimated that there was a 44 percent gain in productivity—the number of windshields installed per day per worker—under the new incentive system. He further estimated that approximately half of the gain in productivity resulted from the same employees doing more work under the new system.28 Lazear also found that the effects of increased productivity persisted, actually increasing over time, which rules out the explanation that the effect was simply due to a shift in payment scheme from one system to another, a novelty or Hawthorne effect, which would diminish over time. Instead, “after workers are switched to piece rates, they seem to learn ways to work faster or harder as time progresses.” 29 Part of the productivity effects came from retaining and attracting better employees. Turnover at Safelite was high—close to 4 percent per month. After the new incentive system was installed, the average employee hired had higher productivity than those already in the plant and turnover was higher among the least productive people. The average wage went up about 7 percent, much less than the increase in productivity, so the cost per unit declined from an average of $44.43 under the hourly wage system to $35.24 under the piece rate system.30 Several characteristics of Safelite make it especially suitable for a successful variable pay or incentive system. First, the task was readily learned and, more important, involved little or no interdependence with other employees. It is possible that people learned from each other and shared tips on how to do the job more ef?ciently, but the task itself was done by one person

12 Dangerous Half-Truths About Managing People and Organizations

working alone. Individual incentives did not undermine teamwork because there wasn’t any need for teamwork. Second, it was easy to both measure and monitor quality, so employees could not simply work faster at the expense of doing a decent job. If a windshield broke, it was easy to identify the culprit. The company required the installer “to reinstall the windshield on his own time” and to “pay the company for the replacement glass before any paying jobs are assigned to him.” 31 Third, the company already had a sophisticated computerized work monitoring system, so the incentive system did not require new innovations for measuring employee productivity. And ?nally, employee goals were unambiguous and one-dimensional—to install windshields as quickly as possible, while doing a job of suf?cient quality to keep the windshield from falling out or breaking. Lazear is careful to explain why this setting was so ideal for using individual incentives: “output is easily measured, quality problems are readily detected, and blame is assignable.” 32 Unfortunately, few of the numerous other organizations that have implemented incentive systems have been as thoughtful about the scope conditions that make such systems effective. The literature is littered with tales of disastrous implementations, which often occur not because incentives don’t work, but instead because they work too well. Given an incentive to achieve some outcome, people do take that incentive seriously and work hard to obtain the goal that will earn them the ?nancial reward. The problem is that most organizations have more complex, multidimensional objectives and optimizing on just one thing creates other dif?culties. Consider a few of many possible examples. In Albuquerque, New Mexico, the city decided to put garbage truck drivers on an incentive system: if drivers ?nished their routes early, they could go home and still receive pay for their full eight-hour shift. The program was instituted to cut down on overtime by encouraging drivers to ?nish their prescribed routes on time or even early. This meant that “a driver who completes a route in ?ve hours would get ?ve hours of regular pay plus three hours of incentive pay.” 33 But an audit discovered numerous problems. “Fifteen of the twenty-four drivers who received the most incentive pay in 2002 consistently went to the land?ll in trucks over the legal weight limit.” Plus there was evidence that the incentive to complete the routes early resulted in more preventable traf?c accidents. Some drivers missed picking up all of the garbage on their routes and many were reluctant to stop using a truck that might need repairing. The audit of the program concluded, “the unintended results of the incentive

Do Financial Incentives Drive Company Performance?

13

program could be an increase in safety risks, cost of operations, legal liabilities, and customer dissatisfaction.” 34 In New Orleans, once the murder capital of the United States, the police felt pressure to make the city safer. The city instituted a program under which “districts that showed improvement in crime statistics got awards that could lead to bonuses and promotions, while districts that didn’t faced cutbacks and ?rings.” 35 To further encourage improvement, districts were put in competition with each other to see which one could cut crime the most. It turns out there are at least two ways to cut serious crime—by reducing the incidence of serious crime or by reclassifying crimes that do occur as less serious. Faced with pressure and incentives to cut serious crime, you can imagine what happened. In the First District, the top cops cut serious crime by simply reclassifying crimes. An “investigation found that in the last year and a half in that one district, 42 percent, nearly half of all serious crimes, were classi?ed as minor offenses and never fully investigated.” 36 The result: when the reclassi?cations came to light, the police chief ?red ?ve senior of?cers, including the district commander, for falsifying crime statistics. These fudging problems aren’t con?ned to the public sector. An analysis of the overbooking of oil reserves by Royal Dutch/Shell that resulted in the resignation of the chairman of the company and its head of exploration, followed shortly by the departure of the chief ?nancial of?cer, pointed the blame at incentives. Shell compensated its executives with stock options, so there were incentives to keep the stock price up. One way of maintaining the stock price was to overstate reserves. “In a 2001 report, Houston consultants Rose & Associates noted the pressure on managers at publicly traded energy companies ‘to push the envelope of credibility in an effort to buoy investor con?dence and thus increase stock value.’ Among other things, the consultants pegged the overbooking to incentive programs that offered bonuses for big reserve estimates.” 37 The lesson here is a variant on an old adage: be careful what you pay for, you may actually get it. When the tasks that people do are even modestly complex, it is often impossible to think up every possible way that they might achieve those goals. And even if it were feasible to imagine every contingency, the long and convoluted list of rules and conditions would render the system incomprehensible and ineffective. Financial incentives are best applied when there are simple, clear, agreed-upon measures that make cheating almost impossible, or perhaps, when the powers that be care only about

14 Dangerous Half-Truths About Managing People and Organizations

optimizing performance on those measures, regardless of what it takes for people to hit the numbers.

Incentive Systems Do Attract Talent—Often the Wrong Kind
There is little doubt that the level of ?nancial incentives offered and the system used—for instance, individual pay for performance—attracts different people to different organizations. Incentive systems are a big part of an organization’s culture. Some organizations explicitly select for cultural ?t, so they recruit people who ?t their values, including those re?ected in the incentive system. Even for those organizations that are less systematic about selecting for ?t, prospective employees will try to determine if they are likely to succeed in the company—after all, who wants to work at a place where they will fail? So candidates use the incentive system as an important way of diagnosing the organization’s or occupation’s culture and values. There are occupations—investment banking among others—where most people are in it mostly for the money. A few years back, Pfeffer listened to videotapes of focus groups conducted anonymously for a large San Francisco law ?rm in which 2-Ls, law students between their second and third year of law school, discussed the advantages of working for a particular ?rm that paid well. But the ?rm also required long hours and provided poor feedback and supervision for associates, as is typical of many large law ?rms. The students discussed working at the ?rm mostly in terms of the high salaries that would make paying back their student loans easier and faster. So Lazear’s ?ndings about which existing employees left Safelite Glass and which new employees took jobs are not anomalous; incentive systems and the amount of ?nancial incentives offered do attract different people to different companies. This ?nding does not, however, answer the question of whether you should want people who come to your organization because of the ?nancial incentives you offer. Years ago, when James Treybig was CEO of Tandem Computers (a company since merged into Ungermann-Bass), the company would not tell people their salaries before they were hired, even at the most senior executive levels. After extensive recruiting and many meetings, people would be told (if it were true) that there was a good ?t between them and the company and that there was every expectation that they would be successful at Tandem and that the company wanted them to join. If they asked about their salary, they would be told that Tandem paid a competitive salary and offered a competitive ?nancial package. If people insisted on knowing their precise salary, they would not be offered the job. To paraphrase Treybig, “if people

Do Financial Incentives Drive Company Performance?

15

come for money, they will leave for money.” He recognized that ?nancial incentives are the most fungible resource, available to almost any organization. People who came only for the pay package could be lured away by the next company that offered a little bit more, and since turnover was disruptive, Tandem wanted people who would have a greater likelihood of staying with the organization. Treybig and Tandem believed that if people joined because they liked the work, the business, the management, the culture, and their coworkers, they would be more likely to stay than if they came just for the money. This lesson about joining an organization for the right reasons is also relevant for the law ?rm just described. The students in the focus groups were quite clear that they would stay with the ?rm until their student loans were paid off or greatly reduced, and then leave. And, indeed, this particular law ?rm lost more than 50 percent of its associates by the end of their third year of employment, which was precisely the point at which they had enough experience to become pro?table for the ?rm. Contrary to some popular lore, the hardest-working people are often those working in the government. Because they are doing their jobs out of a sense of service and to make a difference, you often see more, not less, commitment and effort among public service employees.38 Our colleague James Baron makes a similar point about not necessarily wanting to rely on ?nancial incentives when teaching his MBA course on human resource management. He poses the following hypothetical question to the students (and you can answer it for yourself). If you had a choice, when confronting a serious, possibly life-threatening illness, of going to see one of two doctors, which would you choose: (a) a doctor who had entered medicine primarily to make a lot of money, or (b) a doctor who had entered medicine because he or she was interested in the subject matter and had a desire to serve people? Think about which doctor you would choose and why. Not surprisingly, the majority of the MBA students, who are more oriented toward ?nancial incentives than most people, choose the second doctor. The reasons the MBAs provide are often consistent with the sociological concept of professionalization— that is, a professional is an individual who puts the clients’ interests ?rst, who has an obligation to do right and do well by that client, regardless of that professional’s own self-interest. We expect doctors and other professionals to take our interests and needs into account, and not to think only or even mostly about what will enrich them. We want professionals serving us to choose treatments or courses of action based on the best available evidence for ef?cacy, not what will make them the most money in the short term.

16 Dangerous Half-Truths About Managing People and Organizations

Even in nonprofessional contexts, there may be good reasons why you should be cautious about hiring people who join your organization only or mostly for the money—especially when employees are in high demand. Let’s return to the example of Safelite Glass for a moment. As Lazear points out, the average pay per employee went up 7 percent, while ?rm productivity went up 44 percent.39 How can this gap between pay and ?rm productivity be explained—with ?rm productivity increasing much more than pay? Safelite is in Columbus, Ohio, where there is an oversupply of workers clamoring for decent-paying jobs. So workers lacked the leverage to get more than a fraction of the value produced by their greater efforts—Safelite could exploit an imperfect, somewhat “sticky” labor market. But such is not always the case. When employees hold the upper hand, and companies battle for top talent with money alone, then their best people will keep leaving for more money, as they are working for nothing else. Consider major league baseball. The advent of free agency and de-emphasis on team loyalty has produced rapidly increasing salaries for players and a system in which most teams lose money on an operating basis. These examples demonstrate part of James Treybig’s “if they come for money, they’ll leave for money” concern. Playing a ?nancial incentive strategy in a labor market where people can and do search for all available opportunities and move for higher compensation will do precisely what economic theory predicts: ensure that employees will receive their marginal revenue product and further ensure that companies will be forced to pay wages that are fully at market, thereby reducing the company’s pro?ts from what they might have been had they offered employees a less wage-centric employment bargain. If you believe that people work only for the money, such departures would be impossible to stop. But it would also be impossible to explain why people ever join the Peace Corps, enlist in the military when they have betterpaying jobs (as professional football player Pat Tillman, who later died in Afghanistan, did), and why even in professional baseball, players still accept lower salaries to stay with teammates and managements they enjoy working with—as San Francisco Giants player J. T. Snow did in 2003. And there is one other reason not to select those people who want to be in your organization only or mostly for the money they can make. There is evidence, at least from student populations, that people who choose a place for more instrumental reasons are much more likely to engage in dishonest behavior. Don McCabe and his colleagues have conducted numerous studies of college student cheating over the years.40 They have found that students who are in school or have chosen a major for instrumental reasons—in order

Do Financial Incentives Drive Company Performance?

17

to get a better job or to make more money—are much more likely to cheat than students who have chosen a course of study because of their interest in the subject matter. This result makes perfect sense if you think about it. If I am trying to master a subject because of my intrinsic interest, cheating makes no logical sense—it defeats my desire to learn the material. If I am, on the other stand, studying just to get a credential, than what matters is the credential—getting out with the piece of paper—not necessarily what I learn. The implications for companies are clear. If people are there for the money— at Enron, for instance, or anywhere else—then they will do what it takes to get the money, regardless of what that is. Much better, it would seem, to have people who actually have some interest in the company, its customers, its products and services, and its values.

Variable Pay = Pay Dispersion = Lower Performance
The use of individual ?nancial incentives nearly always increases the dispersion or inequality in rewards. At Safelite Glass, the variation in monthly salary earned by employees doing glass installation was approximately 43 percent higher under the piece rate plan than under the hourly pay system. This result is scarcely surprising, as variable ?nancial incentives are meant to create wider gaps between what the best and worst people are paid. The intention is to get away from the mayonnaise theory of salary administration, in which raises are spread rather equally and thinly across the entire employee base, and instead, give bigger rewards to employees that contribute most to organizational performance. The empirical question is what does the evidence say about the consequences of creating more unequal ?nancial rewards? There are two basic assumptions behind the idea that more dispersed ?nancial incentives are desirable and will enhance organizational performance: ? Employees who make outstanding contributions want to be recognized. ? Employees believe it is unfair that they get the same raises as colleagues who do not expend the same level of effort or accomplish as much as they do, so most employees prefer more dispersed pay. These assumptions sound perfectly reasonable on the surface, but each becomes problematic when organizations actually try to implement differential rewards. As to the ?rst assumption, although people may want to be recognized for their outstanding contributions, this causes a big problem

18 Dangerous Half-Truths About Managing People and Organizations

because the world of work is lot like the Garrison Keillor’s mythical Lake Wobegon, “where all the children are above average.” A vast literature demonstrates that in virtually every sphere of life—from driving ability to a CEO’s power to extract value out of mergers and acquisitions—most people believe they are far above average. These are called self-enhancement effects, or the desire of people to think more positively about themselves. Most people are more likely to perceive themselves as superior to others along numerous positive dimensions, see themselves more positively than others see them, believe they are above average and not recognize their lack of competence, and take credit for their success but see their failures as outside of their control.41 The consequences of such overestimates of ability and performance are painfully obvious to any manager who has ever administered differential raises or given a performance review. The employee almost always has an opinion of his or her performance that is higher than that held by the manager. When performance can be reliably and unambiguously measured— which means that the performance can be quantitatively and objectively assessed, measured along one or just a few dimensions (so trade-offs between dimensions are not required), and performance does not result from joint efforts with others (which raises questions of relative credit)—differential ?nancial incentives can be justi?ed by recourse to these objective measures. In all other cases, there can be and usually is debate about relative merit, and people who receive a smaller reward than they expect routinely resent the organization and the manager who, in their judgment, has made a biased and ?awed assessment that denigrates their excellent work. These battling performance perceptions are one reason why compensation consulting companies ?nd that most pay-for-performance programs fail to achieve their objectives, and dissatisfaction with such programs is usually so high. A 2004 survey by Hewitt of 350 companies showed that “83 percent of organizations believe their pay-for-performance programs are only somewhat successful or not successful at accomplishing their goals.” 42 To add insult to injury, after ?rst forcing managers to “stack” their employees from best to worst—which has profound implications for employees’ feelings of self-worth and status and is done through a process often fraught with disagreements among managers about who deserves what rating and ranking— most organizations then seriously underfund their pay-for-performance programs. As Watson Wyatt noted, “merit budgets are so constrained that they do not suf?ciently reward exceptional performance or differentiate top employees from others.” 43

Do Financial Incentives Drive Company Performance?

19

There is also mixed evidence about whether people really want to be differentiated from their fellow employees. Executives in some companies we’ve encountered report that their managers resist pressures to strongly differentiate among the ?nancial rewards given to their subordinates, and such resistance appears to be wise at times. One Cisco manager complained to us that he couldn’t understand why HR insisted that he give big bonuses to his top few people and ?re a couple people at the bottom each year, as he had assembled an excellent and cooperative team—by carefully hiring the right people and easing out the wrong people, his team was composed of all excellent people. Controlled experiments with temporary teams composed of strangers show that participants usually choose to avoid handing out big differences in rewards that mirror the big differences in individual performance seen in groups—and this is in a context where there are far weaker social connections among people than in ongoing work settings.44 The reason seems clear. People derive satisfaction from their social relationships in the workplace. Differential rewards people drive people apart, sorting them into categories as “winners,” “nothing special,” and “losers.” The result is jealousy and resentment, which damages social ties and diminishes trust and sociability in the workplace. Few organizations adequately fund their ?nancial incentive programs well enough to provide economically meaningful and substantial differences in rewards anyway.45 So why should organizations pay the price of damaged social relations, people suffer through arguments about relative merit for insigni?cant ?nancial bene?ts, and managers devote so many hours ranking and rating their people? Despite what so many compensation consultants and HR executives advise, most people prefer to avoid these nasty side effects and, given a choice, choose to work for more equal rewards. Individual incentives and highly differentiated reward and recognition distributions make more sense when performance can be objectively assessed and when performance is mostly the result of individual effort rather than the product of interdependent activity. So there is evidence that jockeys perform better when pay is contingent on performance, that tree planters in British Columbia are more productive when paid on a piece rate, and that loggers produced more when their piece rates were set at higher levels.46 Similarly, the evidence suggests that more dispersed ?nancial rewards increase the performance (particularly of the highest performers) when tasks entail little or no interdependence and outcomes are clear. A study of 379 trucking companies found that larger differences in ?nancial rewards between the best and worst paid drivers was associated with better company performance. A

20 Dangerous Half-Truths About Managing People and Organizations

controlled experiment in picking oranges also found that greater dispersion in rewards produced higher level of performance.47 Studies of golf tournaments and race car drivers—both activities where winning was clearly measurable and there was little interdependence—demonstrated that a greater difference in rewards (or a larger ?rst prize) produced better performance.48 In the case of the golf tournament study, this happened even when the quality of the participants was statistically controlled, although larger prizes in automobile racing produced faster times but also more accidents. Yet when work settings require even modest interdependence and cooperation, as most do, dispersed rewards have consistently negative consequences on organizations. A study of college and university faculty showed that the greater the dispersion of pay within academic departments, the lower the job satisfaction, the less collaboration, and the lower the level of research productivity.49 A study of 67 publicly traded companies found that ?rms with greater differences between the best- and worst-paid executives in the top management team had subsequently weaker ?nancial performance (measured by total shareholder returns).50 The study also found that the negative effect of pay disparity was especially pronounced for high-technology ?rms, because these ?rms had the greatest need for collaboration and teamwork to cope with complex and rapidly changing competitive conditions. A sample of 102 business units found that the greater the gap between top management and employee pay, the lower the product quality.51 The same negative effects of dispersed pay are seen in professional sports. Studies of baseball teams are interesting because, among major professional sports, baseball requires the least coordination and cooperation among team members. But baseball still requires some cooperation; for example, between pitchers and catchers, and among in?elders. And although individuals go to bat, teammates can help each other improve their skills and break out of slumps. Matt Bloom’s study of over 1,500 professional baseball players on 29 teams over an eight-year period showed that—controlling for the effects of base pay, past performance, age, and experience—players on teams with dispersed pay performed consistently worse, especially the lower-paid players. Not only that, teams with more dispersed pay had lower winning percentages, gate receipts, and media income.52

Guidelines for Using Incentives
It isn’t easy to build pay systems that inspire, guide, and energize people without, at the same time, damaging your organization and people. If you

Do Financial Incentives Drive Company Performance?

21

look at the best evidence, instead of listening to the best-paid consulting ?rms and gurus, you will see that simple palliatives like pay for performance aren’t likely to ?x all—or even any—of your performance problems and may instead drive up costs, hamper cooperation, and sti?e new ideas. But you do have to pay people. What is a manager to do? You might begin by using the ideas and data outlined in this chapter to develop a more complete view of human psychology and its implications for using ?nancial incentives. This research about pay and people leads us to four overarching guidelines for managers to use as they think about and implement ?nancial incentives.

Don’t Try to Solve Every Problem with Financial Incentives
The biggest problem with ?nancial incentives is that they are tremendously overused. Incentives has emerged as the ?rst answer to almost every problem. Are your schools failing? Bribe teachers with incentive pay. Is the medical system inef?cient, with vast differences in treatment protocols for the same disease in different regions? Set up a managed care system that provides ?nancial incentives to doctors, insurers, patients, and hospitals. Bad customer service? Provide ?nancial incentives for better customer service. Airplanes not ?ying on time? Pay employees if the planes ?y on time. Too much overtime in garbage collection? Give truck drivers a ?nancial incentive to ?nish early. Stock price not high enough? Give senior management ?nancial incentives to get the stock price up. And on and on it goes, often with disastrous results. But incentives often aren’t that effective. Beyond all the problems we have enumerated, consider one more: people adapt, fairly rapidly actually, to the rewards. The result is that bonuses for performance become part of people’s total compensation and come to be expected. As David Russo, formerly the head of human resources at SAS Institute, once commented, “a raise is only a raise for 30 days. After that, it’s just somebody’s salary.” To obtain the bene?ts of the informational effects of ?nancial incentives, here’s an idea: instead of using subtle, often misunderstood ?nancial rewards that people may try to game, try talking to people about the company, its strategy, and its priorities. What a novel idea. SAS Institute, the largest privately owned software company in the world with sales of over $1.3 billion and a 98 percent customer renewal rate, has largely eschewed an emphasis on ?nancial incentives in its management approach. As Barrett Joyner, former head of North American sales and marketing, commented, “We have sales targets, but mostly as a way of keeping score. I want to make the numbers, but I want to make the numbers the right way . . . I’m

22 Dangerous Half-Truths About Managing People and Organizations

not smart enough to incent on a formula. People are constantly ?nding holes in incentive plans . . . Here, we just tell people what we want them to do and what we expect.” 53 To obtain great employee motivation, instead of signaling people through lavish and contingent ?nancial rewards that they are working mostly for the money, let them see and experience other bene?ts from their work, such as being part of a supportive community and doing work that helps bene?t others. So, for instance, Southwest Airlines talks about bringing people together. Low fares, which are possible only because of the low costs that come from a productive workforce, become not just some competitive strategy but something that enables customers to see their family and friends more often. The Men’s Wearhouse, the large retailer of tailored men’s clothing, encourages its employees to help each other become better people than they ever thought they could be, even as they describe their job as helping men look better, feel more con?dent, and be more successful in their lives. DaVita, which runs kidney dialysis centers, shows pictures of its patients and has a video segment in which dialysis patients and their families, students (in the case of a teacher), and work colleagues say, “thank you DaVita,” for keeping the person alive. SonoSite, a developer and manufacturer of high-quality, small, lightweight ultrasound equipment, encourages its employees to consider how lives are saved by bringing diagnostic ultrasound to previously unreachable places. At one annual meeting, a captain from the U.S. Army described the use of SonoSite equipment to provide care to the wounded during the war in Iraq, while the company’s Web site has a section entitled “SonoSite Moments” that displays unsolicited stories from doctors and other medical professionals about how the company’s machines helped save lives. Instead of using ?nancial incentive plans to sort people, consider trying to attract people for other reasons—such as believing in the company, liking its culture, and enjoying the work. Justin Kitch, the founding CEO of Homestead.com, talked to us about recruiting at Harvard Business School at the height of the Internet boom. He recounted how students mostly asked questions that indicated they were primarily interested in ?guring out how rich they could get how quickly. He didn’t hire anyone and stopped recruiting at business schools. Kitch thought that the company would be better served, and better able to survive the inevitable reversals of fortune and business dif?culties, if its people wanted to be there for its mission, technology, and culture. Perhaps that is why, while so many dot-coms failed and so many executives come and go in Silicon Valley, Homestead—although it has had its ups and down—survived the dot-com bust and kept its top management

Do Financial Incentives Drive Company Performance?

23

team largely intact as well. Homestead ?nally had its ?rst pro?table year in 2004 and currently has over 60,000 paying customers. The point is that there are other ways of accomplishing the motivational, selection, and informational effects of ?nancial incentives. And in many instances, these alternatives are not only less expensive, but they are actually better methods for accomplishing these important organization-building objectives.

Sometimes Less Is More Effective
In the late 1990s, The Men’s Wearhouse faced a challenge. On the one hand, its commission structure for wardrobe consultants rewarded salespeople on an individual basis for the clothes they sold, with a higher commission for larger orders to signal the importance of cross-selling merchandise. On the other hand, the company emphasized the concept of teamwork and team selling—helping others when they were waiting on customers. For instance, a participant at one of our executive programs described how she, her husband, and her children went to a Men’s Wearhouse store to buy a blazer. When they arrived, one wardrobe consultant began to wait on them while another took the kids in the back for juice, cookies, and video games. When the couple had purchased what they came in for, the mother and father went to get the kids—who were having so much fun they didn’t want to leave. With the children being cared for, the woman and her husband stayed in the store longer and bought more clothes. This is one example of team selling— the person with the kids in the back was going to get no commission for the sales made by his colleague. Other examples included helping fellow wardrobe consultants answer questions, providing peer coaching to teach others about clothes and selling, and helping everyone maintain the appearance of the store and the merchandise displays. To help build a team atmosphere, in addition to encouraging sports activities and socializing among the staff, the company gave either no award, a $20 award, or a $40 award, paid in cash, to employees at a store at the end of the month depending on the store’s shrink—levels of lost and stolen inventory. In commenting on the modest size of these awards, CEO and founder George Zimmer demonstrated a remarkable insight into the psychology of incentives and how they could be both too big and too small. He asserted that the award was just the right amount, because it engendered some excitement but was not so big that it distorted behavior or became the focus of attention. Rather, the focus remained on the celebration of the store’s achievement and the spirit of camaraderie, rather than the money. By

24 Dangerous Half-Truths About Managing People and Organizations

the same token, Zimmer noted the amount had to be large enough to be meaningful, and in this instance, the company went to the trouble to pay the award in cash. Zimmer’s insight is particularly prescient in the context of executive ?nancial incentives that do distort behavior and that distract attention from the business, employees, and customers.

Be Careful What You Wish For, You Might Just Get It
If you are going to pay people for doing something, you need to be very, very thoughtful about the possible consequences of the behavior you have just signaled them to do. You need to think hard about what will happen if people take the ?nancial incentives seriously, and really do seek to maximize their performance along those dimensions, and only those dimensions, that you reward. Much as in medicine or engineering, where people think about what could or might go wrong in recommending a treatment or designing a building, this is precisely how managers need to think about ?nancial incentives. By ?rst anticipating what might go wrong, you can then balance the risks and costs against the potential bene?ts, and you might even be able to redesign the ?nancial incentives to minimize the risks while retaining the bene?ts. It is impossible for any human being to anticipate every behavioral eventuality that ?nancial incentives will produce, so it is crucial to consider incentive systems as works in progress, or as experiments being run, not as things to be put in place and left alone inde?nitely, regardless of the outcomes. This recommendation ?ies in the face of the immense dif?culty companies have in changing their pay systems and how, once implemented, incentive systems become institutionalized. But it is only by learning by doing and being open to such learning that serious problems can be avoided. The belief that your pay system is best viewed as a prototype, something that you will change when better information is discovered, re?ects the mind-set we introduced in our opening chapters that best sets the stage for evidence-based management: the attitude of wisdom, that managers need to act on the best knowledge that they have right now, while doubting what they know. And managers can also better practice evidence-based management if they teach their people—and yes, provide incentives—so that they will accept and help support experimentation with the pay system.

Worry About Comparisons and Distributions, Not Only Individuals or Levels
Organizations are social entities, and people are social creatures. What this means for leaders is that social relations are important. People compare

Do Financial Incentives Drive Company Performance?

25

themselves to others and derive feelings of worth and status from that comparison. Consequently, pay differences have not only substantive but symbolic meaning. If a colleague makes $1,000 more, that $1,000 permits the person to buy more goods and services—the substantive difference in wages. But once people are beyond the point where they need every cent to buy basic necessities, small differences in pay can still have huge effects on motivation, attitudes toward the company and its management, and turnover. What may seem like trivial differences to a manager—say if a person gets $74,000 a year and his or her colleague gets $75,000—may be interpreted by the lower-paid employee as a sign that the organization values that other person more, so tiny differences can have great consequences for a person’s ego and feelings of self-worth. Social comparisons are part of the human condition— and are magni?ed in individualistic and competitive cultures like we have in the United States. Lots of companies get into trouble by forgetting this simple fact and not considering what the distribution of rewards looks like and what messages that distribution sends to everyone. Take the most notorious example, CEO pay. CEOs who make several hundred times more than what the average employee in their companies makes send the signal that what they do is hundreds of times more important. Is that really the right signal to send? If frontline people think that what they do doesn’t matter very much for the organization’s success or in the opinion of senior management, why bother to worry about how well they are doing their job? It is not by accident or coincidence that many of the most successful, consistently best-performing companies have CEOs who are not outrageously overpaid—Amazon.com, CostCo, and Southwest Airlines are a few current examples. By sending a signal that performance is a collective, not just an individual, endeavor, those companies are more likely to induce thought, creativity, and effort on the part of their people. We have seen in this chapter that the use of ?nancial incentives is a subject ?lled with ideology and belief—and where many of those beliefs have little or no evidence to support them. Pay is also an important topic, consuming time and resources and doing a lot of harm when it is mismanaged. As a consequence, you might think that using ?nancial incentives to improve performance would be a domain where there would be heavy reliance on the best evidence. As we have seen, however, consultants, gurus, and executives charge ahead with assumptions and practices that re?ect a reckless disregard for the evidence. But this gap between evidence and action provides opportunities for those organizations wise enough to examine their assumptions about pay systems, and to ?nd and use the best evidence.

Notes
Chapter 5
1. This statement comes from page 2 of the Cendant Corporation 2004 Proxy Statement, March 1, 2004. Similar statements can be found in many other proxy statements. Cendant and its chief executive, Henry Silverman, have been roundly criticized for excessive CEO compensation and for compensation that was not very closely tied to company performance (see, for instance, Gretchen Morgenson, “Two Pay Packages, Two Different Galaxies,” New York Times, April 4, 2004. Nonetheless, incentive alignment is apparently an important goal and is taken as an article of faith that it is important in most proxy statements. 2. Jay R. Schuster and Patricia K. Zingheim, The New Pay (San Francisco: Jossey-Bass, 1992), ix. 3. Frederick W. Taylor, Shop Management (New York: Harper, 1903). 4. Edward P. Lazear, “Performance Pay and Productivity,” American Economic Review 90 (2000): 1346. 5. See, for instance, B. F. Skinner, Science and Human Behavior (New York: Macmillan, 1953). Operant conditioning principles have been applied in management and organizational behavior by, among others, Walter Nord, “Beyond the Teaching Machine: The Neglected Area of Operant Conditioning in the Theory and Practice of Management,” Organizational Behavior and Human Performance 4 (1969): 375–401; and Fred Luthans and Robert Kreitner, Organizational Behavior Modi?cation (Glenview, IL: Scott, Foresman, 1975). 6. This expectancy theory of motivation is quite prominent in social psychology. An early formulation can be found in Victor H. Vroom, Work and Motivation (New York: John Wiley, 1964). Also, modern psychological research on decision making and negotiation re?ect many of these assumptions about the power of ?nancial rewards, but does show that cognitive biases “trip up” people in their quest to maximize valued outcomes, especially when it comes to maximizing ?nancial gain. See Max H. Bazerman, Judgment in Managerial Decision Making, 6th ed. (Somerset, NJ: Wiley, 2006). 7. See, for instance, Andrea Gabor, The Man Who Discovered Quality (New York: Times Books, 1990). 8. Scott McCartney, “How to Make an Airline Run on Schedule,” Wall Street Journal, December 22, 1995. 9. Edward P. Lazear, “The Power of Incentives,” The American Economic Review 90 (2000): 410. 10. Robert L. Heneman, “Merit Pay Research,” in Research in Personnel and Human Resource Management, vol. 8 (Greenwich, CT: JAI Press, 1990): 203–263. 11. One study by Hewitt Associates, reported that as of the late 1990s some 72 percent of all companies had variable pay for at least some groups of employees compared to 47 percent in 1990. See Rebecca Ganzel, “What’s Wrong with Pay for Performance?” Training, December 1998, 34–40. 12. Ellen G. Frank, “Trends in Incentive Compensation” (information prepared for the authors from Hewitt Compensation Surveys, June 2004). 13. “Christmas Bonuses Give Way to Incentive Pay,” Edmonton Journal, November 28, 2003. 14. “Garbage Truck Drivers Rushing to Finish Work Are Safety Risk,” Associated Press, January 30, 2004. 15. Diana Jean Schemo, “When Students’ Gains Help Teachers’ Bottom Line,” New York Times, May 9, 2004. 16. “Results-Oriented Cultures: Creating a Clear Linkage Between Individual Performance and Organizational Success,” GAO-03-488 (Washington, DC: General Accounting Of?ce, March 2003). 17. Christopher Lee, “Civil Service System on Way Out at DHS,” Washington Post, January 27, 2005, http://www.washingtonpost.com/wp-dyn/articles/A39934-2005Jan26.html (accessed May 1, 2005).

Notes

27

18. Chip Heath, “On the Social Psychology of Agency Relationships: Lay Theories of Motivation Overemphasize Extrinsic Incentives,” Organizational Behavior and Human Decision Processes 78 (1999): 25–62. 19. Ibid., 28. 20. Ibid., 38. 21. Watson Wyatt Worldwide company report, “Strategic Rewards: Maximizing the Return on Your Reward Investment,” 2004, 11. 22. Michael Beer and Nancy Katz, “Do Incentives Work? The Perceptions of a Worldwide Sample of Senior Executives,” Human Resource Planning 26 (2003): 30–44. 23. Ibid. 24. “At Emery Air Freight: Positive Reinforcement Boosts Performance,” Organizational Dynamics 1 (1973): 42. 25. Ibid., 47. 26. Robert Rodin, Free, Perfect, and Now (New York: Simon and Schuster, 1999), 45. 27. Lazear, “Performance Pay and Productivity.” 28. Ibid., 1353. 29. Ibid., 1354. 30. Ibid., 1352. 31. Ibid., 1358. 32. Ibid. 33. “Garbage Truck Drivers.” 34. Ibid. 35. Transcript from National Public Radio, All Things Considered, October 30, 2003, 9:00–10:00 p.m. edition, p. 2. 36. Ibid. 37. Chip Cummins, Susan Warren, Alexei Barrionuevo, and Bhushan Bahree, “Losing Reserve: At Shell, Strategy and Structure Fueled Troubles,” Wall Street Journal, March 12, 2004. 38. The image of the New York ?re?ghters going up into the World Trade Center following the initial onset of the disaster of September 11, 2001, is probably overused. But it does raise the interesting issue of whether people could be induced to risk their lives for money, or if it is only a sense of duty, service, and professional calling that could produce acts of such heroism. 39. Lazear, “Performance Pay and Productivity,” 1357. 40. Donald L. McCabe and Linda Klebe Trevino, “Cheating Among Business Students: A Challenge for Business Leaders and Educators,” Journal of Management Education 19 (1995): 205–218; Donald L. McCabe and Linda Klebe Trevino, “What We Know About Cheating in College,” Change 28 (1996): 29–33. 41. J. D. Brown, “Evaluations of Self and Others: Self-Enhancement Biases in Social Judgments,” Social Cognition 4 (1986): 353–376; J. Kruger and D. Dunning, “Unskilled and Unaware of It: How Dif?culties in Recognizing One’s Own Incompetence Lead to In?ated Self-Assessments,” Journal of Personality and Social Psychology 77 (1999): 1121–1134; D. T. Miller and M. Ross, “Self-Serving Biases in the Attribution of Causality: Fact or Faction?” Psychological Bulletin 82 (1975): 213–225. 42. “Many Companies Fail to Achieve Success with Pay-for-Performance Programs,” Hewitt Associates News & Information, June 9, 2004. 43. Watson Wyatt, “Strategic Rewards,” 12. 44. Gerald S. Leventhal, “The Distribution of Rewards and Resources in Groups and Organizations,” in Leonard Berkowitz and Elaine Walster (eds.), Advances in Experimental Social Psychology, vol. 9, (New York: Academic Press, 1976), 92–259, provides a good summary of this research. See also G. S. Leventhal, J. W. Michaels, and C. Sanford, “Inequity and Interpersonal Con?ict: Reward Allocation and Secrecy About Reward as Methods of Preventing Con?ict,” Journal of Personality and Social Psychology 23 (1972): 88–102. 45. Hewitt Associates News & Information, “Many Companies Fail to Achieve Success”.

28

Notes

46. Sue Fernie and David Metcalf, “It’s Not What You Pay, It’s the Way You Pay It and That’s What Gets Results: Jockey’s Pay and Performance,” discussion paper 295, London School of Economics, London, 1996; Harry J. Paarsch and Bruce S. Shearer, “The Response of Worker Effort to Piece Rates: Evidence from the British Columbia Tree-Planting Industry,” Journal of Human Resources 35 (1999): 1–25; M. Ryan Haley, “The Response of Worker Effort to Piece Rates: Evidence from the Midwest Logging Industry,” Journal of Human Resources 38 (2003): 881–890. 47. Gary Bornstein and Ido Erev, “The Enhancing Effect of Intergroup Competition on Group Performance,” in Using Con?ict in Organizations, eds. Carsten De Dreu and Evert Van De Vliert (London: Sage, 2003), 147–160. 48. R. G. Ehrenberg and M. L. Bognanno, “The Incentive Effects of Tournaments Revisited: Evidence from the European PGA Tour,” Industrial and Labor Relations Review 43 (1990): 74S–88S; B. E. Becker and M. A. Huselid, “The Incentive Effects of Tournament Compensation Systems,” Administrative Science Quarterly 37 (1992): 336–350. 49. Jeffrey Pfeffer and Nancy Langton, “The Effect of Wage Dispersion on Satisfaction, Productivity, and Working Collaboratively: Evidence from College and University Faculty,” Administrative Science Quarterly 38 (1993): 382–407. 50. Phyllis A. Siegel and Donald C. Hambrick, “Pay Disparities Within Top Management Groups: Evidence of Harmful Effects on Performance of High-Technology Firms,” Organization Science 16 (2005): 259–274. 51. D. M. Cowherd and D. I. Levine, “Product Quality and Pay Equity Between LowerLevel Employees and Top Management: An Investigation of Distributive Justice Theory,” Administrative Science Quarterly 37 (1992): 302–320. 52. Matt Bloom, “The Performance Effects of Pay Dispersion on Individuals and Organizations,” Academy of Management Journal 42 (1999): 25–40. 53. Jeffrey Pfeffer, “SAS Institute (A): A Different Approach to Incentives and People Management in the Software Industry,” Case #HR6A (Stanford, CA, 1998), 8.


相关文章:
更多相关标签: