CorporateBudget

ing Is BrokenLet's Fix It

Traditional budgeting

processes waste time,

distort decisions,

and turn honeSt managers into

It doesn't have to be that way-

if you're willing to

the ties between

and compensation

by Michael C.Jensen

NOVEMBER 2001 95

Corporate Budgeting Is Broken - Let's Fix It

CORPORATE BUDGETING IS A JOKE, andeveryone knows it. It consumes a huge amountof executives' time, forcing them into endlessrounds of dull meetings and tense negotiations.

It encourages managers to lie and cheat, lowballing tar-gets and inflating results, and it penalizes them for tellingthe truth. It turns business decisions into elaborate exer-cises in gaming. It sets colleague against colleague, creat-ing distrust and ill will. And it distorts incentives, moti-vating people to act in ways that run counter to the bestinterests of their companies.

Consider just two examples. At one internationalheavy-equipment manufacturer, managers were so set onhitting their quarterly revenue target that they shippedunfinished products from their plant in England all theway to a warehouse in the Netherlands, near the cus-tomer, for final assembly. By shipping the incompleteproducts, they were able to realize the sales before theend of the quarter and thus fulfill their budget goal andmake their bonuses. But the high cost of assembling thegoods at a distant location - it required not only the rentalof the warehouse but also additional labor-ended upreducing the company's overall profit.

Then there's the recent debacle involving a big bever-age company. The vice president of sales for one of thecompany's largest regions dramatically underpredicteddemand for an upcoming major holiday. His motivationwas simple-he wanted to ensure a low revenue targetthat he could be certain of exceeding. But the price for hislittle white lie was extremely high: The company based itsdemand planning on his sales forecast and consequentlyran out of its core product in one of its largest markets atthe height of the holiday selling season.

Such cases of distorted decision making are legion inbusiness. No doubt, you could list similar instances thatyou've observed-or perhaps even instigated-at yourown company. The sad thing is, these shenanigans havebecome so common that they're almost invisible. Thebudgeting process is so deeply embedded in corporate lifethat the attendant lies and games are simply accepted asbusiness as usual, no matter how destructive they are.

But it doesn't have to be that way. Even if you grantthat budgeting, like death and taxes, will always be withus, deceitful behavior doesn't have to be. That's becausethe budget process itself isn't the root cause of thecounterproductive actions; rather, it's the use of budgettargets to determine compensation. When managers aretold they'll get bonuses if they reach specific performance

Michael C. Jensen, the Jesse Isidor Straus Professor of Busi-ness Administration, Emeritus, at Harvard Business Schoolin Boston, is the managing director of the organizationalstrategy practice of the Monitor Group, a collection of globalprofessional services firms with headquarters in Cambridge,Massachusetts.

goals, two things inevitably happen. First, they attemptto set low targets that are easily achievable. Then, oncethe targets are in place, they do whatever it takes to seethat they hit them, even if the company suffers as a result.

Only by severing the link between budgets andbonuses-by rewarding people purely for their accom-plishments, not for their ability to hit targets- will we takeaway the incentive to cheat Only then will we eliminatethe budgeting incentives that drive individuals to act inways that destroy corporate value.

Cheaters ProsperLet's look more carefully at how budgets drive compen-sation and, in turn, behavior. In a traditional pay-for-performance incentive system, a manager's total cashcompensation (salary plus bonus) is constant until a min-imum performance hurdle is reached-commonly 80%of a budgeted target. (The target might be expressed asprofits, sales, output, or any number of things; for our pur-poses, it doesn't matter what's being measured.) Whenthe manager exceeds that hurdle, she receives a bonus,often a substantial one. The bonus then increases as per-formance mounts above the hurdle until the bonusis capped at some maximum level-120% of the target isusual. This system is illustrated in the exhibit "A TypicalExecutive Compensation Plan."

The kinks in the pay-for-performance line-caused bythe minimum hurdle bonus and the maximum cap-cre-ate strong incentives to game the system. As long as themanager believes she can make the minimum hurdle, shewill naturally try her best to increase performance-bylegitimate means or, if push comes to shove, by illegiti-mate ones. If the measure is profits, for instance, she willhave a strong incentive to increase the current year'searnings at the expense of next year's, either by pushingexpenses into the future (delaying purchases or hires, forexample) or by moving future revenues to the present(booking orders early or offering special discounts tocustomers, for example).

If, on the other hand, the manager concludes that shecan't make the minimum hurdle, her incentives flip 180degrees. Now her goal is to move earnings from thepresent to the future. After all, her compensation doesn'tchange whether she misses the target by a little or a lot;she still gets her full salary (assuming she doesn't getfired, of course). But by shifting profits forward-by pre-paying expenses, taking write-offs, or delaying the real-ization of revenues - she increases her chances of gettinga large bonus the following year. This is a variation on the"big bath" theory of corporate financial reporting: Ifyou're going to take a loss, take as big a loss as possible.

Finally, if the manager is having a great year and herperformance is nearing the budget cap, she again has anincentive to push profits into the future. Because she's

96 HARVARD BUSINESS REVIEW

C o r p o r a t e B u d g e t i n g Is Broken - Let's Fix It

not going to get any additional compensation if perfor-mance exceeds the levei at which the cap is set, accelerat-ing expenses or postponing sales will have no negativeimpact on her current earnings, but it will raise the oddsthat she'll reap a high bonus next year as well. This per-verse incentive becomes even stronger if her currentyear's performance is used in setting the following year'stargets, as is often the case.

When these kinds of subterfuge simply move profitsfrom one year to another-by changing accruals, for ex-ample - the adverse impact on company value is probablysmall. But rarely is the activity so benign. Usually, theshuffling of dollars results from decisions that change theoperating characteristics of a company, and it generateshigh, if sometimes hidden, costs that erode the total valueof the company. We saw such erosion in the two examplespresented earlier. We see it as well in the common prac-tice of channel stuffing-when managers ship loads ofproducts to distributors to meet immediate sales goals,

A Typical ExecutiveCompensation Plan

In a traditional pay-for-performance compensationplan, a manager earns a hurdle bonus when perfor-mance reaches a certain level (A). The bonus increaseswith performance until it hits a maximum cap (B). Thekinks in the pay-for-performance line create incentivesto game the system. When performance approachesthe hurdle target, a manager has a strong incentive toaccelerate the realization of revenue and profit Whenperformance hits the cap, the manager has a strongincentive to push revenue and profit into the next year.

Performance M e a i u re

even though they know many of the goods will soon comeback as returns. And we see it in distorted pricing deci-sions. The managers of one durable-goods manufacturer,struggling to meet their minimum bonus hurdles, an-nounced late one year that they would be raising pricesio% across the board on January 2. The managers madethe price hikes because they wanted to encourage cus-tomers to place orders by year-end so they could hit theirannual sales goals. But the price increase was out of linewith the competition and undoubtedly ended up costingthe company sales and market share.

Even more insidious effects are common. One of themain reasons that big companies have budgets in the firstplace is to help coordinate the disparate parts of theirbusinesses. By openly sharing accurate information andbasing decisions on a common set of numbers, the think-ing goes, you ensure harmonious interactions amongunits, leading to efficient processes, high-quality products,low inventories, and satisfied customers. But as soon asyou start motivating unit and department heads to falsifyforecasts and otherwise hide or manipulate critical infor-mation, you undermine the salutary effects of budgeting.Indeed, the whole effort backfires. You end up with unco-ordinated, chaotic interactions as people make decisionson the basis of distorted information they receive fromother units and from headquarters. Moreover, since man-agers are well aware that everyone is attempting to gamethe system for personal reasons, you create an organiza-tion rife with cynicism, suspicion, and mistrust

When the manipulation of budget targets becomesroutine, moreover, it can undermine the integrity of anentire organization. Once managers see that it's okayto lie and conceal information to enrich themselves orsimply to hold on to their jobs, they soon begin to extendtheir dishonest behavior to all parts of the company'smanagement system and even to its relationships withoutside parties. Managers start to feed misleading infor-mation to customers, suppliers, and employees, and theCEO and CFO begin to "manage the numbers" to influ-ence the perceptions of board members and Wall Streetanalysts. Even boards of directors are drawn into the fray,as they end up endorsing deceptive reports to sharehold-ers. Sometimes, outright fraud ensues, as we've seenrecently in high-profile cases involving companies such asInformix, Sabratek, and Lernout & Hauspie.

The damage can go well beyond the walls of individualcompanies. Think about what happens, for example, dur-ing a boom. As financial analysts and investors raiseexpectations for growth beyond the capability of compa-nies, many managers begin to borrow from the future tosatisfy the present demands. This results in an overstate-ment of earnings and cash fiows for many companies andan exaggeration of the extent of the good times. Con-versely, during the early stages of an economic slowdown,as demand falls below predicted levels and inventories

NOVEMBER 2001 97

C o r p o r a t e B u d g e t i n g Is B r o k e n - L e t ' s F i x I t

build up, managers often find themselves falling short oftheir bonus targets. When they and their companies allreact in the same, predictable way-taking big baths bymaximizing the bad news-the cumulative effect is to ex-aggerate the economic weakness, perhaps deepening orextending the recession. Macroeconomic statistics andeven public policy are likely distorted in the process.'

Getting the Kinks OutThe only way to solve the problem is to remove all thekinks from the pay-for-performance l ine- to adopt apurely linear bonus schedule, as shown in the exhibit"A Linear Compensation Plan." Managers are still re-warded for go<id performance, but the rewards are inde-pendent of budget targets: The bonus a manager receivesfor a given level of performance remains the samewhether the budget goal is set at Target l (below actualperformance) or at Target 2 (above actual performance).The linear bonus schedule, in other words, rewards peoplefor what they actually do, not for what they do relativeto what they say they can do.

That removes the incentives to game the system. Be-cause unit managers no longer get a pile of cash for ex-ceeding a target, they have no motivation to feed falsifiedinformation into the budgeting process in order to low-ball their goals. As a result, senior management receivesunbiased estimates of what can be accomplished in thefuture, and the quality of planning and coordination im-proves considerably. At the other end of the process, man-agers are no longer rewarded for moving revenues andexpenses around when the end of a budget period ap-proaches. Because their bonuses are always determinedby their actual performance, an extra dollar of revenue orprofit (or whatever measure is used) will generate thesame bonus this year as it will next year. Not only will thisremove the costs of gaming, it will also free managersfrom all the time they traditionally had to devote to it.They can dedicate that time to their real jobs: maximizingtbe performance and value of their businesses.

Two additional points should be made here. First, non-monetary rewards also have to be independent of budgettargets. Handing out promotions or public accolades basedon the ability to hit budget numbers, for example, will pro-vide a continued incentive for gaming. That practice has tostop. All rewards must be based purely on actual perfor-mance.- Second, since pay will still go up as performanceimproves, dishonest managers may continue to lie abouttheir numbers in order to increase their bonuses. That, ofcourse, is a risk that companies have always had to watchout for. A linear bonus schedule does not reduce tbe needfor good control systems and attentive executives.

Removing the kinks from the compensation system al-lows the budgeting process to do what it's intended to do:provide the basis for good business decisions and enable

tbe effective coordination of disparate units. But manymanagers will reflexively object to this proposal. Yes,they'll say, a linear bonus schedule will remove the incen-tive to game the system, but won't it also remove the mo-tivational effects of performance targets? It's a legitimatequestion, and it's one that is difficult to answer with hardnumbers. Empirical evidence shows that tying goals to re-wards does enhance performance. A recent survey of morethan 500 studies, for example, indicates that performanceincreases by an average of approximately 16% in compa-nies that establish targets.' But even if we assume suchfindings are accurate, we don't know whether the perfor-mance increase is itself a result of gaming, as managersrush to overstate their results to meet the new targets.

Nor do we know definitively the long-term costs ofgaming. No comprehensive study has been done, andsuch a study would be hard to carry out, given how per-vasive and well-hidden these costs are. However, DonaldRoy conducted a landmark study, published in the Amer-ican Journal of Sociology in 1952, of an analogous situa-tion: the use of piecework targets to determine the

A Linear Compensation Plan

The solution to the budget gaming problem: Adopt apurely linear pay-for-performance system that rewardsactual performance, independent of budget targets.A manager receives the same bonus for a given levelof performance whether the budget goal happens tobe set beneath that level (BudgetTargeti) or above it(Budget Target 2). Removing the kinks eliminates theincentives for managers to game the process.

ActualBonus

P e r t o r m a n c c M e a s u r e

98 HARVARD BUSINESS REVIEW

Corporate Budgeting Is Broken - Let's Fix It

bonuses of factory workers. Based on his findings, Royestimated that productivity in the factory he studiedwould increase by 33% to 150% if the targets were dis-continued. Based on that research, as well as myown observations of the widespread, destructiveeffects of gaming, 1 conclude that the costs of bud-get-based bonuses far outweigh the benefits inmost, if not all, situations.

Finally, it's important to note that settingextremely aggressive stretch goals, as is so com-mon in business today, can itself have damagingrepercussions. By establishing the expectationthat managers will constantly push to exceedreasonable growth and profitability targets,senior executives can end up creating a dys-functional organizational culture in whichall the problems I've described are ampli-fied. That's what happened recently at one hprominent multinational corporation. Anew CEO came in, and he suspected that unitheads were routinely lowballing their budgettargets and then delivering mediocre results.He quickly reorganized the company to estab-lish clearer accountability down the line, andthen he launched an intensive campaign to geteveryone to set stretch goals for the coming year,with their bonuses hanging in the balance. Theeffort blew up in the CEO's face.

The budget-setting process became a year-longexercise in internecine warfare. Knowing that theirbonuses hinged on their ability to hit the new tar-gets, line managers battled over the way the overallcorporate stretch goals for revenues and profits should beallocated among the business units. Each, of course, triedto reduce his or her unit's target. Every time revised goalswere circulated, new arguments broke out. And when thetargets were eventually finalized, things got even worse.Within months, most unit heads realized they wouldn't beable to reach their stretch goals, and they let the year fallinto the tank in a way that is consistent with pushing rev-enues and profits out to the future. They were clearly hop-ing that by taking a bath this year they would be givenlower targets next year. Needless to say, the CEO's tenurewas short.

Making the SwitchIt's not going to be easy for companies to adopt a linearcompensation system. Target-based bonuses are deeplyingrained in the minds of managers and in the manager-ial codes of most organizations. Getting managers to givethem up will be tough, and getting them to give up pro-motions or reputational rewards for "beating budget" willbe even tougher. Mostdifficult of all, though, will beget-ting them to break free of the cynicism that surrounds the

When you start rewarding managersfor falsifying forecasts and otherwisedistorting critical information, youthreaten the integrity of your entireorganization.

entire budgeting and bonus-paying system. But the bene-fits are so great-in terms of a company's long-run eco-nomic and organizational health-that the journey willbe well worth the time and effort. And there are someguidelines that can help the transition proceed moresmoothly.

Get the details right. The design of compensation pro-grams lies outside the scope of this article. But it is im-portant to emphasize that the success of a linear bonusprogram, like that of any pay system, hinges on its details.There are three key considerations: the performance mea-sures used, the positioning and slope of the bonus line,and the establishment of minimum and maximum com-pensation levels.

In establishing performance measures, executives runthe risk of distorting managerial decisions, even under alinear bonus system. One example is the use of multiplemeasures of performance. This practice can motivatemanagers to think more broadly and carefully about theoperating and economic drivers of business success. But italso adds complexity to the system in a way that can im-pede decision making. Suppose, for example, a manageris told to increase both profits and market share in the

NOVEMBER 2001 99

C o r p o r a t e B u d g e t i n g Is B r o k e n - Let's Fix I t

coming year. If, after some point, market share can beincreased only by cutting prices, and thus profits, themanager no longer has a basis for making reasoned deci-sions. The confiicting goals eliminate his ability to act pur-posefully. When using multiple performance measuresfor individual managers, companies should be careful toestablish a single, clearly defined measure of overall busi-ness success, such as economic value added. That will givemanagers a basis for making trade-offs among perfor-mance measures when they come into conflict."

A second example is the use of ratios as performancemeasures. Here I can be blunt; Don't do it. Using ratios,such as sales margin or return on assets, inevitably pro-duces gaming. That's because managers can increase themeasure in two ways: by increasing the numerator or de-creasing the denominator. If, for example, a companytracks performance according to margin as a percentageof sales, managers can increase their pay by simply cuttingback sales (selling only the highest margin products)instead of working to increase the margins on all prod-ucts. The result: Total dollars of profit fall, and companyvalue erodes.

The positioning and slope of the bonus line work in tan-dem to determine the amount of money a managerreceives for a given level of performance. Moving the lineto the right on the performance scale, for instance, makesit harder to get an additional dollar of bonus, while pro-viding a steeper slope makes it easier to get that dollar. Insetting the line, executives have a tendency to focus on theshort term. In particular, they often position the line basedon the prior year's performance. That reduces the risk thatmanagers will be overcompensated for overly conservativeprojections of performance (a high bonus in one yearmakes it harder to get a high bonus the next year), but italso reduces incentives for increasing performance. Be-cause managers know that an increase in performance thisyear will result in higher targets for succeeding years, themotivation to make the extra effort is dampened-unlessthe current year's bonus is extraordinarily large. A betterway is to look further into the future, setting bonus linesfor a number of years out based on longer-term projec-tions for growth and profitability. This is harder to do, butit reduces the potential for manipulation.

in general, it is extremely important that companiesmodel the economic impacts of different positionings andslopes for their bonus lines. As in any compensation sys-tem, the right balance must be struck between the re-wards delivered to employees for incentive purposes andthe capital retained for reinvestment or for distributionto investors or other owners.

Finally, there's the issue of limits on compensation. Ide-ally, you wouldn't have any-all pay would be directly re-lated to actual performance. But there are strong pres-sures to limit both the upside and the downside foremployees and managers. Most companies have to pay

salaries in order to attract and retain employees (thusdefining the lower bound of compensation), and mostfeel compelled to set some upper limit to bonuses. What'simportant is to try to set the upper and lower limits out-side the range of likely outcomes to minimize their po-tential for setting off gaming. In many cases, this willrequire companies to raise bonus caps well above tradi-tional levels, which is sure to create organizational dis-comfort. For example, a manager will inevitably end uppaying certain high-performing subordinates more thanhe himself makes. Also, some managers may complainthat certain people are being paid inordinately largeamounts just because they happened to get lucky. (Ofcourse, luck is extremely hard to distinguish from talent,and much of the griping will simply be a sign of jealousy.)Managing such conflicts-by clearly articulating the phi-losophy underlying the pay plan, for example-is one ofthe difficult but necessary challenges in moving to a morerational compensation scheme.

The Impact of a CurvilinearCompensation Plan

Curvilinear pay-for-performance plans are no curefor gaming. They encourage managers to increasethe variability of year-to-year performance measures.In this example, a manager makes more by achieving80% of a budgettarget one year and 120% the nextthan by hitting the target both years. Under the linearplan, by contrast, the manager would earn the samein both scenarios.

$20,000

$15,000

$12,000 -

$10,000

P e . r j , o rjnM i r x e M e a s u r e

rfl"

100 HARVARD BUSINESS REVIEW

Corporate Budgeting Is Broken - Let's Fix It

Don't backtrack. Even after managers become intel-lectually convinced that a linear bonus schedule is desir-able, they may still argue for a compromise plan thatagain allows budget targets to infiuence compensation. Ina number of companies 1 have worked with, executiveshave, for instance, proposed replacing the new linearschedule with a curvilinear one. By making the pay-for-performance line curve upward more steeply after thebudget goal is reached (and less steeply before it isreached), bonuses begin to increase more rapidly forevery incremental improvement in performance beyondthe target. That may appear to fulfill the psychologicalneed of many managers to "get something more" aftersurpassing a goal, without putting any obvious kinks backinto the system. But in fact, curvilinear schedules reintro-duce a strong incentive for gaming.

To see why, look at the exhibit "The Impact of a Curvi-linear Compensation Plan." In the example portrayedhere, a manager whose profit performance in two suc-ceeding years exactly matched the budget goal wouldreceive a bonus of $i2,txx> in each year, for a total of$24,txx'). But if she could manipulate that same amountof profit so that she hit 80% of the goal in one year and120% in the next, she would receive $io,cx>o in one yearand $20,i.xx> in the other, for a total of $30,000. The curvi-linear plan, in other words, has given her a strong mone-tary incentive to start fudging the numbers again, and, inthis case, she would be rewarded for increasing the vari-ability of performance. What's more, she would be willingto do this even if it lowered overall performance withinsome range. In contrast, under a truly linear scheme, themanager would receive the same total ($3o,(xxi) in bothscenarios.

Last January, Chrysler Motors introduced a new dealer-incentive plan that was based on a curvilinear pay-for-performance scale. The plan paid dealers a monthlybonus ranging from zero (for those who achieved lessthan 75% of their sales targets) to $soo per car sold (forthose who achieved more than iK>% of the target). Whencar sales began to slow in April, many Chrysler dealers re-alized that they were unlikely to exceed the targets for themonth and therefore they reduced inventories, shiftingsales of cars from April to May, when they could be muchmore certain to earn the $500 per car bonus. In the end,Chrysler's CEO bad to announce that company sales felli8% in April as a result of its dealer bonus program, whileoverall industry sales were down only u)%.

Lead from the top. Given the complexity of designinga new pay system, as well as the controversy it inevitablysets off, CEOs will feel a natural desire to hand off re-sponsibility for the effort to the human resources depart-ment. That would be a mistake, probably a fatal one. HRhas neither the standing nor the influence to make a fun-damental business change that will have a profound im-pact on the decisions of line managers. And that's exactly

the kind of change that I'm talking about: All membersin the organization will have to shift their thinking aboutthe rote and use of both budgets and incentives. Perfor-mance measures will have to be changed, and bonus lev-els will have to be recalculated. Since these issues are assensitive as any within a company, strong leadership isessential. Only the CEO has tbe credibility to make thebusiness case for the changes and to rally tbe troopsbehind them.

The CEO should recognize that the new plan will meetwith intense resistance, even at the highest reaches of theorganization. Some of the strongest objections, I havefound, tend to come from CFOs and their teams.Finance executives naturally fear that reducing the im-portance of budget targets in motivating line managerswill make it more difficult to control results and avoid sur-prises. It will be the CEO's responsibility to make sure theCFO, not to mention Wall Street analysts, understandsthat the new approach will improve the quality of boththe information provided and the incentives paid to man-agers. And better information and better incentives willlead to better results. Yes, there may be greater uncer-tainty in quarter-to-quarter results-as there will nolonger be any motivation to set artificially low targets andthen do everything possible to meet them-but the long-run profits will be superior.

Finally, every line manager will have to understand thenew system and the theory behind it and be prepared toexplain it and defend it against opposition and oppor-tunism from the ranks. There are no shortcuts throughthe education process. Organizations don't changeovernight, particularly when the very frame throughwhich we see the business is involved. Remember, it hastaken many years to weave lying and deceit into the fab-ric of our businesses; cleansing the fabric will take timeas well. ^

1. See M.C. Jensen,"Paying People to Lie: The Truth About the Budgeting Pro-cess," working paper (Harvard Business School, April 2001). Available on-lineat http://ssm.conv'paper=26765i-

2. To read more on how goals can distort behavior, even in the absence of tiesto tangible rewards, see M. Schweitzer, L. Ordonez, and B. Doiima,"The DarkSide of Goal Setting: The Role of Goals in Motivating Unethical Behavior,"working paper (Wharton School, University of Pennsylvania, 2(.xii).

3. See Edwin A. Locke, "Motivation by Goal Setting," in Handbook of Organi-zational Behavior, ed. Robert T. Golembiewski (Marcel Dekker, 2(wi). Also seeEdwin A. Locke and Gary P Latham, 4 Theory of Coal Setting and Task Perfor-mance (Prentice Hall, 1990).

4. See Michael C. Jensen,"Value Maximization, Stakeholder Theory, and theCorporate Objective Function," Business Ethics Quarterly, January 2tx>i. Avail-able on-line at http://ssrn.conVpaper=22o67i.

The author thanks Joseph Fuller, Michael Gibbs, Jennifer Lacks-Kaplan, andEdwin Locke for their contributions to this article.

Reprint ROIIOFTo order reprints, see the last page of Executive Summaries.

To further explore the topic of this article, go towww.hbr.org/explore.

NOVEMBER 2001 101

Harvard Business Review Notice of Use Restrictions, May 2009

Harvard Business Review and Harvard Business Publishing Newsletter content on EBSCOhost is licensed for

the private individual use of authorized EBSCOhost users. It is not intended for use as assigned course material

in academic institutions nor as corporate learning or training materials in businesses. Academic licensees may

not use this content in electronic reserves, electronic course packs, persistent linking from syllabi or by any

other means of incorporating the content into course resources. Business licensees may not host this content on

learning management systems or use persistent linking or other means to incorporate the content into learning

management systems. Harvard Business Publishing will be pleased to grant permission to make this content

available through such means. For rates and permission, contact permissions@harvardbusiness.org.