Journal of Applied Corporate Finance S U M M E R 1 9 9 9 V O L U M E 1 2 . 2
Twelve Ways to Strengthen Your Incentive Plan by David Glassman, Stern Stewart & Co.
TWELVE WAYS TOSTRENGTHEN YOURINCENTIVE PLAN
by David Glassman,Stern Stewart & Co.
107BANK OF AMERICA JOURNAL OF APPLIED CORPORATE FINANCE
ew areas of corporate governance generate as much divisivenessand distrust as incentive compensation. From determining the keymeasures, to setting the targets, to establishing the payout range,companies seem continuously to undermine their own plans,
processes, and credibility.With great fanfare new incentive schemes are rolled out that few under-
stand. Worse, new plans from the outset are viewed as the flavor of the month,likely to be changed again next year. Lack of credibility diminishes the buy-inof managers and, along with it, the enthusiasm, motivation, and drive the plansare intended to generate. For investors, this makes the incentive plan a low rate-of-return initiative.
Our experience at Stern Stewart suggests that the problems besettingincentive plans are surprisingly similar across companies and even acrossapparently dissimilar industries. Most plans are ineffective because of:
Too many performance measures;Measures that motivate the wrong behavior;Too much complexity;Arbitrary targets that are subject to intense lobbying by executives;Too much managerial discretion;Performance measured at a level too high to be meaningful, or too low to
encourage teamwork;Caps and floors that narrow the payout range and stifle incentives;A failure to integrate the incentive plan into the overall compensation
philosophy; andIneffective or non-existent communication.
Obviously, companies do not set out to develop convoluted programs. Sowhy does it happen? Why are incentive plans so poorly constructed when CEOsrecognize their importance for corporate governance? And why are plans socomplicated when simplicity is so clearly a priority?
This paper identifies why incentive plans are unproductive, and offers 12suggestions for implementing plans that support managements aspirations to createvalue for shareholders. The remedies described below are not difficult to executeand, judging from the evidence, are remarkably effective. But they require the strongcommitment and attention of senior management. Left solely to the human re-sources staff, capable as they may be, institutional resistance will be overwhelming.
108VOLUME 12 NUMBER 2 SUMMER 1999
1. IMPROVE FOCUS AND GET THE MEASURERIGHT
Most companies ask too much of their incentiveplans. Obviously, they aim to motivate improve-ments in financial performance, but how is thisdefined? Growth is a key goal, so a target for Salesgains plays a role. Profit (or EPS), of course, ishighlighted in the incentive plan. And Finance oftenproposes a balance sheet measure, such as inventoryturns or Return on Investment. With these threemeasures, the Compensation Committee believesthat all financial bases are covered. In fact, theproblems are just beginning.
One problem is that any particular initiative islikely to push different financial measures in conflict-ing directions. Sales may rise but the gain may be toocostly, leaving profits unchanged, or even belowwhere they started. Or an SKU reduction initiativemay reduce sales and profit but increase inventoryturns and ROI. Instead of guiding decisions, themeasures create confusion. And, to add to theconfusion, management also inserts non-financialconsiderations into the incentive plan. These mayinclude customer and employee satisfaction, safetyand environmental compliance, investment for thelong term, leadership, and so forth.
Also, the many measures influencing bonusesdiffuse, rather than concentrate, the focus of execu-tives. Worse, they can even encourage counterpro-ductive behavior. At one company that used Salesand Profit measures for its incentive plan, a particu-larly candid manager admitted that in a recent yearhis business was unlikely to achieve the Profitcomponent. When asked how this influenced deci-sion-making, he said, We made sure we hit the Salestarget. How? By discounting product in the fourthquarter. Margins suffered, profits were down, but atleast part of the bonus was salvaged.
Management can improve the focus of incen-tive systems by using just one financial measure,Economic Value Added (EVA). EVA, measured asthe profit after deducting a charge for all capital,internalizes the tradeoff between income statementreductions and improvements in capital productiv-ity. As Peter Drucker said recently in Fortune:
there is no profit unless you earn the cost ofcapital. Alfred Marshall said that in 1896, PeterDrucker said that in 1954 and in 1973, and nowEVA (economic value added) has systematized thisidea, thank God.1
EVA contains in one measure the benefits ofSales and Margin gains offset by increases in Capitalthat may be required. An internal study performedby Stern Stewart, and summarized in the sameFortune article, indicates that companies adoptingEVA for incentive compensation outperform peercompanies in the stock market. The 67-companysample of EVA firms provided an average annualreturn of 21.8% vs. 13.0% for industry peers.
Some companies add EVA to the menu ofcompensation measures, but without eliminatingmeasures like sales and EPS growth. This both addsto the confusion and double counts gains or lossesin sales and earnings. Sales, obviously, is a key driverof EVA, but it can be improved without necessarilybenefiting EVA. The same is true of EPS. SternStewart performed a study of the largest 1,000companies in the U.S. to determine the extent towhich EPS and EVA move together. The resultsindicate that, when EVA increases, EPS increases in82% of the cases. But when EVA decreases, EPS stillgrows 57% of the time.2 In other words, set yoursights on EVA growth and favorable EPS results willfollow. But the reverse is uncertain.
EVA focuses incentive plans on the right moti-vationsgrowing the business profitably while tak-ing into account all costs of doing business. But thisfocus is undermined when EVA is forced to share thestage with other financial measures. Eliminate them.
2. ABANDON ATTEMPTS TO MEASURE WHATEXECUTIVES CONTROL
Conversations about development of incentiveplans almost always begin with instructions to Keepit Simple. At the same time, however, CEOs andhuman resource directors want to pay managersonly for what they control. This is a combination sureto cause frustration. Unanticipated events during theyear convert up-front simplicity into back-end com-plexity. Complexity results from the failure to con-
2. Annual data from the Stern Stewart 1,000, 1988-97. The reason why EPSdoes not increase in even more cases can be attributed to accounting restructuringcharges, and increases in R&D. While they diminish EPS, these items are capitalizedfor EVA purposes.
1. September 28, 1998, Q&A with Peter Drucker.
109JOURNAL OF APPLIED CORPORATE FINANCE
sider, and incorporate into bonus programs, thetendency for events to overtake plans.
Despite evidence to the contrary, managersenter the year believing they can estimate within arelatively tight range the performance that can beexpected. Yet, time and again, surprises early in theyear make a mockery of expectations. Surprisesresult from changing trends in economic and indus-try growth, interest rate and commodity price vola-tility (either for inputs or outputs), regulatory changes,competitor initiatives, unusual weather patterns,legal actions, and the like. Because these are viewedas out of the control of managers an after-the-factadjustment to the bonus plan is proposed.
This is where it gets complicated. While out ofmanagements control, arent investors exposed tosuch events? How is a change to be explained tothe Board? And what will be the harm tomanagements political capital with Directors? Ofgreatest importance, shouldnt managers be re-quired to respond swiftly to changes in their mar-kets? While outside of managements control theyare the reality. And managers must internalize anurgency to anticipate such possibilities, build con-tingency plans, and execute them swiftly to limitlosses and capitalize on the opportunities offeredby change. Insulating executives from the impactof market fluctuations denies reality and gets in theway of the necessary responses.
It is easier and simpler to forbid incentive planadjustments during the year. In place of such ad-justments, establish realistic multi-year EVA im-provement targets (discussed in a later section)driven by shareholders requirements for earningan adequate return on investment. Missing thetarget in the current year, for whatever reason, ispenalized with a reduction from the target bonus.But, if the shortfall is judged by top management tohave been caused by factors beyond managementscontrol, the improvement target could then belowered, using as a new base the current (disap-pointing) level of profitability. By making such anadjustment, executives are given the opportunity toearn an attractive award in the following year byrecovering lost ground.
The principle underlying this plan is that man-agers should share with investors the value theycreate. And, using EVA as the key performancemeasure ensures that each dollar of improvementdrives additional value for shareholders as well asmore bonus compensation for participants. Accord-
ingly, the bonus plan motivates managers to think,and act, like owners.
But owners also face risk. They can lose someor all of their money. Accordingly, an EVA BonusPlan sets aside a portion of outstanding awards in aBonus Reserve (or Bonus Bank). The set-aside ispaid to executives if performance is sustained; but ifEVA subsequently declines below a defined thresh-old, the deferred amount can be lost. In other words,a negative bonus is a key feature of the EVA Plan.This ensures that, among other things, decision-makers are properly focused on the long-termconsequences of todays actions.
What executives control is difficult to deter-mine, and in the end does not much matter. Resultsmatter. And results are more likely to be achieved ifexecutives face the downside risk associated withunanticipated events. Economists call this moralhazard. If managers know they will be given relieffor what they cannot control, their incentive to planfor such events is reduced, making the conse-quences more severe.
3. USE INVESTORS, NOT MANAGEMENTS,EXPECTATIONS TO SET TARGETS
Typically, companies rely on the budget pro-cess to determine incentive targets. This is a mis-take, one that corrupts both planning and incen-tives. And, because each is a critical element ofcorporate governance, they should be separated.The reason is that managers tend to water downtheir plan for next year in the hope of achieving asoft performance target.
Knowing that managers have the incentive tohide opportunities, headquarters stretches the per-formance targets. The result is a cycle of managerssubmitting plans, headquarters rejecting them, man-agers recycling them, and so on. For almost allcompanies, negotiating budgets is a time-consumingprocess that makes adversaries of line and staff.Instead, they should be collaborating to find solu-tions to the problems they face and to identifyopportunities to grow.
To make the planning process more effective,bonus targets should be set in advance for a multi-year period without referencing the budget or busi-ness plan. Breaking the link enables managers tocraft a Budget that beats the target, instead ofdeveloping a Budget to negotiate a target. And,because the target is known in advance, manage-
In a study of the largest 1,000 companies in the U.S., Stern Stewart found that whenEVA increases, EPS increases in 82% of the cases. But when EVA decreases, EPS still
grows 57% of the time.
110VOLUME 12 NUMBER 2 SUMMER 1999
ment can plan for an EVA improvement that will earnthem and their team a satisfactory award.
But this still leaves the original question: Howmuch EVA improvement is necessary to pay a targetbonus award? Managers should be awarded a targetbonus when they provide investors with a competi-tive stock market return. This takes into consider-ation the demands of investors now owning orbuying the stock, and recognizes that performancemust improve to deliver a competitive return. Thisrequired return is the companys cost of capital.
When shareholders each day determine theprice at which they are willing to buy or sell thecompanys stock, they signal their perception of thecompanys management, its strategies, prospects,and risks. Investors today are making a statement: astock price and market value that represents apremium over and above the capital currently usedin the business is a vote of confidence in managementsability to produce positive EVA in the future. In fact,the amount of the premium provides an indicationof the amount of EVA growth expected by investors.
Thus, we can use the current stock price andmarket value premium (a measure we call theMarket Value Added, or MVA) to back into thenecessary growth in EVA to justify todays price andearn competitive returns for our current sharehold-ers. Further, we can establish the growth targets fora multi-year period. This eliminates the counter-productive and frustrating annual negotiation, andfrees managers to craft a budget to beat the alreadydefined targets.
Outperforming the growth targets will thendrive outstanding returns to investors and superiorbonuses for executives. Similarly, failing to achievethe targets will disappoint shareholders and gener-ate a less than target bonus for managers.
An important feature of the EVA plan is that itestablishes at the outset the growth track that willdrive a target award. To illustrate, lets suppose thatthe result of the analysis indicates that the companymust improve EVA by, say, $10 million per year foreach of the next three years. If EVA then improvesby $20 million in the first year, a substantial bonuswould be awarded. (Not all of the award would bepaid; a portion would be set aside in the BonusReserve.) But in the next year, an improvement ofanother $10 million would be required to earn a
target award. The Plan is demanding; it pays forgrowing value but always asks for more improve-ment in the next year.
Eli Lilly implemented such an EVA plan in 1995.Randall Tobias, then the companys CEO, says:
We saw as a shortcoming of our incentive systemthat executive pay was linked to sales and net in-come. There just wasnt a very good correlation at allwith shareholder value... Basically, Lillys bonusplan now requires managers to achieve continuousyear-to-year improvements in EVA. Each year wehave to beat that targetand each year the bar israised. Its a small percentage increase, but it keepspounding at you. Whatever you did yesterday, youneed to do better tomorrow to keep raising share-holder value.3
Of course, it is always possible that EVA im-provement will be below the target. Lets say it growsby just $3 million in the first year, $7 million short ofthe target. In this case a less-than-target award ispaid, and $10 million is established as the next yearsgrowth target. Some may suggest requiring associ-ates to make up the shortfall in the second year; thatis, why not add $7 million in EVA growth to the goalto make the target $17 million in improvement?
This would be unfair to managers. They havealready been penalized for the performance short-fall. Raising the target in the next year imposes asecond penalty. For example, if the company wereto achieve the $17 million, managers will haveearned a target bonus in Year 2, but would stillhave less than the targeted bonus for the two yearscombined.
To see this, note that the annual EVA improve-ment target of $10 million means that EVA must growby $20 million for two years. If the company growsEVA by $3 million in the first year and $17 million inthe second, it would achieve the target on a cumu-lative basis. But, by requiring the company to recoupthe first year shortfall in Year 2, the plan would allowmanagers to earn only a target in the second year.And, combined with a less-than-target award in Year1, managers would earn below-target bonuses forthe two years together.
This is important for companies implementingEVA after several years in which EVA has declined.
3. Fortune, September 9, 1996.
111JOURNAL OF APPLIED CORPORATE FINANCE
The goal is to achieve a substantial bounceback,but management should resist incorporating this intothe incentive targets. We should recognize that thecompany was not focused on increasing EVA in thepast. As in the case of Lilly before it adopted EVA,management was focused on other measures. Hold-ing managers accountable now for something thatwas not then the goal seems inappropriate.
4. SET STRETCH GOALS
This does not suggest that the CEO shouldabandon setting stretch goals. Quite the contrary.The companys leaders should emphasize that merelyachieving the incentive target is not the goal. In-stead, we are aiming higher: to achieve not justacceptable returns for our investors, but consis-tently outstanding performance. Further, to accom-plish this, we must outperform the EVA incentivegoal. And if we can do it, then the bonuses forassociates will also be outstanding.
What then is the role of the Budget? If theCompany achieves the stretch goals incorporated inthe Budget, it will earn a greater-than-target award.And this is not unusual. In most cases the Budget isnot the most likely outcome; nor should it be. TheBudget is a positive statement of managementsobjectives, not the highest-probability result. Weshould recognize that achieving the Budget re-quires a strong and sustained focus, some difficultdecisions, and a change in mindset to consider thecost of capital involved both in major and minorinitiatives.
John Blystone, formerly of General Electric,became Chairman and CEO of SPX Corporation in1996. Blystone brought to the $1 billion auto partscompany a philosophy of stretch targets that he hascombined with an EVA program. The EVA target forincentive purposes was established at a $4.2 millionimprovement in the first year. This was calculated tobe the amount necessary to provide investors witha competitive capital market return. At the same time,Blystone set a stretch target of $25 million improve-mentand, if this was achieved the bonus planwould pay over five times the target bonus. Blystonesays, were not going to shoot a manager for doingall the right things and still not getting a stretch goal.
What were going to shoot you for is if you set toolow a target and easily blow by it.4
His experience is that the combination of stretchgoals and EVA works well because EVA strikes theproper balance between the income statement andthe balance sheet... This makes it safer to push forstretch EVA goals than it is to drive for a singleincome statement or balance sheet improvement.We found that you can push very, very hard from anEVA standpoint without breaking anything.5
The Company actually increased EVA in 1996by $26.6 million. The stock price jumped from $15.38to $40.38; and the following year, 1997, the stockclosed at $69. Says an analyst at Bear Stearns: Ofanything automotivethe Big Three, parts compa-nies, you name itSPX is the single most focusedcompany on shareholders, bar none.6
5. REDUCE SUBJECTIVITY
Bonus awards should be predictable. Execu-tives at the end of the yearand even during theyearshould be able to estimate their own bonusbased on the performance of the company. Thisdesirable objective is undermined when companiesincorporate an element of subjectivity into thebonus determination. The reasons for subjectivityare understandable and generally relate to provid-ing competitive compensation to managers evenduring years of company underperformance. Thefear is retention riskthat is, the possibility that thecompany may lose able managers to competitors.In addition, efforts made by management in pooryears are cancelled by the financial results, butnevertheless deserve recognition, or so the reason-ing goes.
But there is a downside both to the logic and tothe practical consequences of this approach. First,lets examine the logic. Shareholders receive theirreturn from the financial results of the company.Poor results, poor return. Why do managers actingas the steward of the shareholders investment, andpresumably being paid a competitive salary, deservea bonus? Instead, they should be provided only arealistic chance next year to recover the bonus lostby improving performance from the disappointinglevel suffered in the current year.
5. From the proceedings of the EVA Institute, March 19-21, 1998; publishedin the EVAngelist, Volume 2, No. 2.
6. Another GE Veteran Rides to the Rescue, Fortune, December 29, 1997.
4. EVA: The Real Key to Creating Wealth, by Al Ehrbar, John Wiley and Sons,Inc., 1998. Page 188.
Breaking the link between the Bonus and the Budget encourages managers to crafta Budget that beats the target, instead of developing a Budget to negotiate a target.
112VOLUME 12 NUMBER 2 SUMMER 1999
On the practical side, subjectivity often resultsin arbitrary goals and evaluations. The goals mayinclude difficult-to-measure judgements about lead-ership, employee satisfaction, or other MBOs that arerarely judged negatively. Over time they become anentitlement, just another way of ensuring a payouteven in bad times.
6. DONT ENCOURAGE PAROCHIALBEHAVIOR
One of the most difficult incentive issues in-volves drawing the balance between awarding bo-nuses based on a business units performance, orthe accomplishments of a functional area, versusthat of the total company. A manager obviously hasthe greatest influence on his or her business unit,and therefore should be compensated on this areasperformance. This works well when the businessunits are separate and there is little value tocollaboration.
But this describes very few companies today.The era of conglomerates has long past. Virtuallyall companies are comprised of at least somewhatrelated businesses that profit by working in con-cert. And each individual business unit is com-prised of inter-related processesresearch and prod-uct development, procurement and manufacturing,information technology, marketing and sales, andso forth. In cases where such interractions aresignificant, incentives that focus managers only ontheir individual businesses and functions couldturn out to be counterproductive. They create siloswhere teamwork is required. They impede com-munication and lead to the hoarding of re-sources and information instead of the sharing ofbest practices.
Consider an example of a retail company orga-nized along two dimensions: Store Operations andMerchandise categories. EVA can be calculated foreach, but the EVA of the Company is best served byhaving the two areas collaborate. Merchandise man-agers oversee the mix (styles, colors, sizes, etc.) butcan take other actions that will benefit Stores. Forexample, they can negotiate the shipping of mer-chandise that is floor-ready, or [cut:they can]require vendors to apply ink tags to the goods. Thevendors will charge a higher cost for the merchan-dise, but in each case the initiative eliminates theneed for handling once merchandise reaches theStore. Merchandise EVA declines, but the Stores
benefit. If their line of sight is their own EVA,Merchandise managers will fail to take the appropri-ate actions that could benefit the Company.
One way to solve this problem is to use totalcompany performance to determine the bonus. Allmanagers will be rewarded or penalized together, asa team. It removes the (explicit monetary) incentivefor parochial behavior. But using total companyperformance dilutes all incentives. One managerscontribution is obscured by the results of the otherbusinesses. This creates what economists call thefree rider problem: all managers have the incentiveto shirk and hope that others perform welland thetotal business suffers.
So what is the right balance? In principle, thegoal should be to focus on individual activities.However, the greater the interrelationships amongdifferent areas, the more that can be gained fromcollaboration. Accordingly, joint incentivesbasedon the aggregate performance of the relevant ar-easproduce the best balance.
Operationally, it is difficult to apply a generaltemplate to the specific circumstances that willinfluence the decision for a particular company. Butwe can offer some general guidelines.
Some of the issues that affect how, and how fast,EVA incentives are introduced include the level ofbusiness literacy, sophistication of information sys-tems, and the degree to which the performance ofbusiness units is independent and separable. Gener-ally, a top-down approach to phasing in EVAincentives is the most prudent. First, integrate EVAinto the management systemmeasurement, re-porting, planning, capital budgeting, and communi-cations. At the same time, involve the businesses intraining managers about the levers they have avail-able to influence the companys EVA. And incorpo-rate EVA into the incentive plan of the primarybusiness units where the lines distinguishing theiractivities are clear, where interrelated activities arefew, and where asset and other allocation problemscan be most easily resolved.
In subsequent years, related EVA measures forkey functions and processesmanufacturing, logis-tics, real estate, product development, sales, market-ing, and so oncan be developed using moresophisticated transfer-pricing techniques that reli-ably measure the EVA contribution made by theseareas. But first the new measures must be given timeto be tested and accepted by the managers that willbe judged on this basis.
113JOURNAL OF APPLIED CORPORATE FINANCE
This approach argues against evaluating func-tional managers using measures related solely totheir processesfor example, using cost per unit,productivity, defect rates, yields, and cycle times toestablish bonus targets for manufacturing managers.All such measures are drivers of success, but eachcan be managed separately to produce a favorableresult while at the same time damaging other partsof the business. Instead, incentive compensation forthese managers should be more heavily weighted tothe performance of the business units they serve.Extensive training and communication can then beprovided to make managers aware of the critical rolethey play in the success of the business, and thetradeoffs that must be balanced between the criticalmeasures bearing on their activities and the overallcompany objective of growing EVA and value.
7. TRACKING STOCK: ANOTHER WAY TOMEASURE AND MOTIVATE
The equity markets provide an alternative ap-proach to governance and incentive compensationfor EVA Centers. An interesting example of this isGenzyme Corporation, which is based in Cam-bridge, Massachusetts. Genzyme provides productsand services for the pharmaceutical and health caremarket. The company is organized in three divisions,each of which has common stock traded in the publicmarkets. Accordingly, investors can take positions inGenzyme Generalthe core of the companys ac-tivitiesGenzyme Tissue Repair, Genzyme Molecu-lar Oncology, or Genzyme Transgenics. Stock in thefirst three units represent separate classes of GenzymeCorporation stock, similar to the way General Motorsorganized its EDS and Hughes subsidiaries. Thismeans that the value of these shares is linked to theperformance of the subsidiary, but that the shares donot represent ownership in the divisions. Instead,investors own shares in a special class of stock inGenzyme Corporation. Genzyme Transgenics, onthe other hand, is owned 41% by Genzyme General,with the balance held by outside investors.
While the separate classes of stock add to thecomplexity of the structure, it simplifies incentives.First, the share price reflects a direct assessment bythe market of the companys success. Managersowning the shares receive ongoing feedback thatdirectly impacts their personal wealth.
The separate tracking stocks also provide other ben-efits. As Genzyme indicates in its literature, the structure:
Provides investors an opportunity to invest selec-tively in the businesses of interest;
Enables each business to raise capital directly;Creates a separate focus for each business; andCaptures the efficiencies of shared resources in
capital formation, research and development, clini-cal and regulatory affairs, and manufacturing.
The Genzyme structure thus has multiple ben-efits. It provides focus both for executives and forinvestors and sharpens accountability for creatingvalue within the key divisions. At the same time, italso enables the small units to take advantage of thelarger companys infrastructure. In addition, taxesare filed on a consolidated basis so that the lossesincurred in one division can offset the taxable profitof another division.
Of course, the complexity of the structureincreases governance costs, particularly where trans-fers of costs and assets are at issue. General andadministrative costs are allocated to the divisionsbased on usage estimates. Also, the Genzyme Boardcan reassign assets, products, and developmentprograms from one division to another. However,compensation based on market value must be paid(in cash or in equivalent value) to the unit surrender-ing the program. And, in the case of Genzyme TissueRepair and Genzyme Molecular Oncology, a vote ofthe holders of the class of stock is necessary forcarrying out the transfer of a program.
This makes governance clumsy for manage-ment, but more transparent for shareholders. Thatkey interdivisional resource allocation decisionsrequire shareholder approval makes cross-subsidiesdifficult to sustain. Instead, each division must raisecapital on its own merits, and therefore faces theurgency to maintain an attractive stock price. Thediscipline imposed by the market aligns the interestsof management with those of investors.
It is not just high-technology companies that usetracking stock. Georgia-Pacific has also created thisstructure to isolate the value of its stable, cash-generating timber products business from the cycli-cal capital intensive pulp and paper operations. Asthe companys 1997 Annual Report points out,Integrated companies compound the problem byusing cash flows from timberlands to finance manu-facturing investments that do not earn their cost ofcapital. This will no longer be an option at Georgia-Pacific. Use of two classes of tracking stock meansthat Excess cash from timberlands will be paid outto investors. The other operations of the company
The Genzyme structure provides focus both for executives and for investors andsharpens accountability for creating value within the key divisions. At the same
time, it enables the small units to take advantage of the larger companys
114VOLUME 12 NUMBER 2 SUMMER 1999
will have to compete for capital on the merits. As thecompany also notes, this step is a natural outgrowthof EVA because separating assets makes eachgroups use of capital easier to evaluate.
The tracking stock also provides for investors apure play opportunity in a focused business withmanagers, through stock options, being rewardedfor the value they add to the business they oversee.
8. DONT FORGET TOP MANAGEMENTS ROLE
Ultimately, even the suggestions offered abovemay not be sufficient to diminish local managementsinclination toward parochial behavior. Any indi-vidual manager will wish to stand out from his peersby demonstrating superior performance in his busi-ness. Such achievements attract rewards over andabove those held out by any bonus planin the formof promotions, special attention from superiors, andso forth.
The solution, therefore, must also come fromoutside the bonus plan. Controlling the tendency forparochial behavior is a challenge for management,not just the incentive program. Creating a culture,through example and leadership, that emphasizesindividual performance while valuing teamwork iswhat senior executives are supposed to do. Askingthe incentive system to do this without additionalreinforcement from top management passes thebuck. It puts too great a burden on the plan.
9. ELIMINATE CAPS AND FLOORS
In all incentive systems there is a tension amongcompeting objectives, some of which have alreadybeen noted and discussed (such as preservingsimplicity vs. rewarding for what managers control,and promoting teamwork vs. creating a local line ofsight). One of the most critical (and generallymismanaged) tradeoffs involves the management ofcost and retention risk vs. the objective of providingtrue incentives.
Traditional compensation philosophy aims toprovide an executive with pay that is competitive inthe labor markets. To manage the risk of losing goodmanagers in bad times some minimum compensa-tion is required. And to avoid excessive costs toshareholders a maximum is established. In theprocess however, motivation is lost.
The narrowing of the difference betweenpay for good and poor performance stifles in-
centives. In good years the incentive is to ensurethat performance does not exceed the point atwhich the maximum bonus is achieved. Execu-tives will book business for delivery next year,and discretionary expenses that would otherwisebe scheduled for next year will be accelerated tothe current years fourth quarter. If SPX had usedbonus caps, you can bet that EVA would nothave improved by $26.6 million when the targetwas $4.2 million. Nor would the stock haveappreciated by 163%. (Whats more, Blystoneviews the EVA plan as also helping to attractextraordinary talent to the company.)
There is a similarly perverse incentive on thedownside. At one company, an executive said thatin the 16 years, I have been with the company,whenever we have missed the threshold for apayout, we never just missed a payout. He de-scribed how bad years were made worse by theincentive to defer revenues, accelerate expenses,and take write-offs because the bonus could not beany worse. In other words, the traditional incentivesystem breaks down when times are particularlygood or bad, as if companies believe that incen-tives dont matter during such periods. Worse,companies implicitly are directing managers to takeactions that diminish current years results.
This problem is fixed by removing the cap andfloor on potential pay. Like SPX, allow bonuses togrow, as EVA grows, without limit. Preserve themarginal incentive to reach for the next dollar. Thecompany need not pay everything in the shortterm. In fact, as noted earlier, EVA companies use aBonus Reserve, an account to which all awards arecredited. If the balance is sufficient, the Reserveeach year pays out a target bonus plus someproportion of the excess that remains (say one-third or one-half). The remainder is carried forwardin the Reserve, but is at risk for future performance.If the company has an unusually poor yearasubstantial decline in EVAa negative bonus canreduce the Bonus Reserve.
In this way managers are encouraged to stretch,even in good years, and are discouraged frommaking a difficult year even worse. It also guardsagainst short-term windfalls that are not sustain-able over time. It does not ensure competitive payin any single year, but the system is calibrated toprovide competitive pay over a multi-year cycle.This is discussed at greater length in the next twosections.
115JOURNAL OF APPLIED CORPORATE FINANCE
10. FIT THE INCENTIVE PLAN INTO THETOTAL COMPENSATION ARCHITECTURE
Any incentive plan, even a well-designed EVAplan, is not a stand-alone management tool. It mustbe designed to fit the companys general compensa-tion architecture, one that rewards managers in avariety of waysbase salary, cash incentive plans,equity programs, pension benefits, perquisites, pro-motions, and so on. Changing the incentive planalters the overall risk-reward dynamics of the com-pensation structure.
For example, the EVA plan typically introducesmore variability into total compensation, becausethe removal of caps and floors generally makes itriskier than the plan it replaces. Absent any adjust-ment for the higher risk, introduction of the EVA planis a net loss to participants. Several clients prior toadopting EVA had an incentive program with aminimum payout of 50% of the target bonus. Thereason, they said, is that base salaries were lowrelative to their competition. In such cases, it wouldbe nave to simply mix in the more aggressiveprinciples of an EVA plan and expect the outcomenot to create unrest.
But rather than modify the EVA plan, thesecompanies increased either the base salaries or thetarget bonus amounts. They recognized that, underthe old plan, deferred salary was essentially mas-querading as incentive pay. This looked good in theproxy but misrepresented the process.
Management and the Board of Directors areresponsible for articulating a compensation philoso-phy, what they hope to achieve from the combina-tion of base salary, short-and long-term incentiveplans, and non-monetary programs. As discussedearlier, they must determine the balance between thestrength of the incentive on one hand, and the impactthis will have on the cost to shareholders in goodtimes and the retention risk borne during bad times.To minimize turnover, some companies designstable compensation programs with little variability.Others prefer more volatility, offer more upsideopportunity, and recognize that they will not attractrisk-averse managers.
In many cases, however, management does notfully understand the fundamental risk-reward char-acteristics of the compensation structure. They havenot modeled in a sophisticated way how totalcompensation can be expected to behave over amulti-year period. Companies generally rely on static
compensation surveys describing the practices ofpeer companies. They tell us the base salary andbonus opportunity of comparable positions, but failto provide insight into riskhow compensationfluctuates with the performance of the company. Thenext section describes an analysis that providesinsight into the dynamics of compensation, a muchmore important element of the design structure.
11. STRESS TEST THE COMPENSATIONARCHITECTURE
Like a piece of machinery, the compensationsystem must be tested to determine how it performsunder stress. This means modeling the interplaybetween various realistic business scenarios, finan-cial results, and total compensation. To the degreethat traditional approaches address these dynamicsat all, they do so in an incomplete way; they lookonly at how changes in financial results createvolatility in the annual bonus. Typically overlookedis how a managers personal wealth is affected bychanges in company performance and value. Forour purposes, a managers wealth is defined as thecurrent value of stock and options owned, plus thepresent value of expected future compensation,including salary, bonus, option (or stock) grants,and pension benefits.
To arrive at an overall compensation strategy,one key issue is how sensitive should managerswealth be to a change in shareholder wealth? Wecall this wealth leverage. Leverage of 100% indi-cates that a 10% increase in company value in-creases an executives wealth by 10%. This wouldbe true of the entrepreneur that receives compensa-tion only in the form of company stock. Bill Gatesand Warren Buffett each have wealth leverage thatis close to 100%.
Alternatively, an executive whose compensa-tion consists entirely of base salary and guaranteedpension benefits would have wealth leverage ofzero; his wealth is independent of the companysstock price performance. This compensation mix isanalogous not to that of a stockholder, but rather tothat of a lender. The lender receives his return evenif the stock price performs poorly; he has a fixedclaim on the companys profit.
The calculations of a managers wealth, and thechanges in wealth for given performance scenarios,are somewhat complex, but computer spreadsheetsenable us to perform these analyses quickly. To fully
infrastructure.To arrive at an overall compensation strategy, one key issue is how sensitive should
managers wealth be to a change in shareholder wealth? We call this wealthleverage. Leverage of 100% indicates that a 10% increase in company value
116VOLUME 12 NUMBER 2 SUMMER 1999
appreciate the dynamics of the incentive system, wesimulate (using Monte Carlo analysis) changes inshareholder wealth and managers wealth for asmany as 100 different performance scenarios over afive-year period. For each scenario, we measure theratio of the change in managers wealth to the changein shareholder wealth.
My former colleague, Steve OByrne, describesone insight provided by this analysis as follows:7
Consider an executive whose total compensation isdelivered through a mix of 30% salary, 20% targetbonus, and 50% stock options. A traditional compen-sation approach would suggest that 70% of thisexecutives compensation is at risk. But our calcu-lation indicates that, over a five-year period, thisexecutives wealth leverage is only 40%. The reasonis that, under traditional stock options programs,options are granted each year at the prevailing stockprice. This means that, if the price falls, more optionsare granted at the lower price. If the stock rises, feweroptions are granted at a higher price. The mix re-balances to make compensation less leveraged.Executives each year are granted options that pro-vide a competitive compensation package. Failing toaccount for how the pay mix fluctuates with perfor-mance misleads executives as to the true nature ofthe compensation plan.
Management can increase the leverage but atthe cost of more retention risk. If stock option grantsare front loaded, or are unaffected by share pricefluctuations (i.e., the number of options granted isfixed each year), leverage increases to approximatethe position of a shareholder having 70% of hiswealth invested in company stock. The executivetruly has 70% of his compensation at risk. But thismeans that in bad times there will be a large gapbetween actual pay and what would be consideredcompetitive. The risk of losing key executives ishigh when performance is poor. And when perfor-mance is strong, the cost to shareholders is high.
These are the key tradeoffs in structuring com-pensation, issues that are often overlooked in theheat of preparing for a compensation committeepresentation about the proposed bonus plan. Thecost of an incomplete analysis is the future need tomodify the incentive plans because of events notanticipated at the outset.
Once the changes are made communication isvital. A variety of audiences should be targeted,including the Board of Directors, who must approvethe plan, key investors, who must approve of theplan, and participants, who must understand it.Incentive plans often fail to achieve credibilitybecause they are not explained adequately. Uncer-tainty creates doubt, and doubt undermines success.
Analysts and investors often ask us how todifferentiate between companies embracing EVA asa fad and those that are using it to create positive andsustained change. The answer can be found in theincentive plan. If the focus is on the same old EPSgrowth objectives, the company has not adoptedEVA. Integrating EVA into the incentive plan, amongother important benefits, provides a powerful signalof the companys commitment to increasing share-holder wealth. Companies must disclose the plan inthe Proxy statement. Many are taking communica-tion a step further and describing the plan in theirAnnual Reports.
For example, Herman Miller after describingthe EVA concept in its 1998 Annual Report goes onto say:
We took EVA a step further by linking ourincentive-based compensation to it. All of our execu-tive incentive compensation plans as well as all of ouremployee gain-sharing programs at each of the busi-ness units have been linked to this measure. UsingEVA-based plans shifts the focus from budget perfor-mance to long-term continuous improvements inshareholder value. The EVA target is raised each yearby an improvement factor, so that increasingly higherEVA targets must be attained in order to earn thesame level of incentive pay. Our Board of Directorshas set the EVA improvement factor for a period ofthree years.
Herman Miller then presents its EVA calculationto investors for 1996-98. It shows an increase from$10.3 million in 1996, to $40.9 million in 1997, to$78.4 million in 1998.
Best Buy, a Minneapolis retailer, announced itsintention to adopt EVA in a public press release dated
7. Steve writes about the wealth leverage framework in an article, TotalCompensation Strategy, published in the Journal of Applied Corporate Finance,Summer 1995, Volume 8 Number 2.
117JOURNAL OF APPLIED CORPORATE FINANCE
November 18, 1998. The headline was Best BuyTeams with Stern Stewart to Adopt Economic ValueAdded (EVA) Framework. Best Buys founder andCEO, Richard Schultze, stated that
The implementation of the EVA framework is akey element of our continuing efforts to evolve as aworld class organization and deliver sustainablevalue to our shareholders, customers and employees.EVA will foster our entrepreneurial culture and focusmanagement to think like shareholders.
Best Buys stock price increased from $46 toabout $50 per share by the close of trading.
Of course, communicating the plan to internalconstituents is important too. Further, communica-tion should not end with the initial workshops andwritten materials associated with the rollout. Up-dates indicating how the company is performingagainst the incentive target should be a regular partof monthly (or, at a minimum, quarterly) communi-cations. This way participants have the informationthey need to assess their position, and understandwhat they must do to improve their bonus during thebalance of the year. The bonus announced byheadquarters at the end of the year should thereforecontain no surprises. It merely verifies the calcula-tion made by the individuals themselves.
Designing and implementing an effective in-centive compensation plan is not just the responsi-bility of the human resources team. Human re-sources organizes and facilitates the overall process,but eliminating the cynicism and distrust surround-ing the effort is a job for everyone.
First, the CEO and the Board must providedirection, balancing the tradeoffs between the strengthof the incentive, the cost to shareholders, and thelevel of retention risk that is tolerable. Financialexecutives must support EVA as the critical financialmeasure used to judge performance. They also have
to be prepared to report EVA results on a regularbasis, and assist in the analytical effort necessary toestablish performance targets.
Of greatest importance is obtaining the buy-inof operating managers early in the process. Thisprovides them the opportunity to help craft theapplication of EVA to their businesses and manag-ers. When the plan is then rolled out to partici-pants, the operating executives will be better pre-pared to explain and champion the plan to theirorganizations.
The executive responsible for investor relationswill be asked to assist in the public communicationof the companys policy. Internal training and com-munications resources help in organizing and spon-soring workshops at which customized case studiesare reviewed to help managers understand the leversthey have available to influence EVA.
A poor implementation process can be worsethan no process at all. Without collaboration withother areas of the company, participants have noownership in the plan. It is their plan, not ours.Worse, the plan can get in the way of the rightdecision and strategy because it is not well under-stood, it may focus on too many different and(conflicting) measures, involve adversarial negotia-tions of performance targets, encourage short-termgaming that may be costly in the long term, andreward a parochial focus at a local level that may becounterproductive for the total company.
The challenge presented by these problems isless intellectual and more one of commitment.Management must have the will to persevere inmaking changes, and not just to the incentive system.It requires first a commitment to shareholder value,one that permeates all key process of the company.The EVA companies that have succeeded in outper-forming peers use EVA as the focus of performancemeasurement and reporting, planning and budget-ing, and capital allocation and communication.Making EVA the key to incentive compensation thenfuels the companys fire to produce continuousgrowth in shareholder wealth.
was a founding Partner, and is now one of the owners, of SternStewart & Co.
increases an executives wealth by 10%.Analysts and investors often ask us how to differentiate between companies
embracing EVA as a fad and those that are using it to create positive and sustainedchange. The answer can be found in the incentive plan.
Journal of Applied Corporate Finance (ISSN 1078-1196 [print], ISSN 1745-6622 [online]) is published quarterly on behalf of Morgan Stanley by Blackwell Publishing, with ofces at 350 Main Street, Malden, MA 02148, USA, and PO Box 1354, 9600 Garsington Road, Oxford OX4 2XG, UK. Call US: (800) 835-6770, UK: +44 1865 778315; fax US: (781) 388-8232, UK: +44 1865 471775, or e-mail: firstname.lastname@example.org.
Information For Subscribers For new orders, renewals, sample copy re-quests, claims, changes of address, and all other subscription correspon-dence, please contact the Customer Service Department at your nearest Blackwell ofce.
Subscription Rates for Volume 17 (four issues) Institutional Premium Rate* The Americas $330, Rest of World 201; Commercial Company Pre-mium Rate, The Americas $440, Rest of World 268; Individual Rate, The Americas $95, Rest of World 70, 105; Students**, The Americas $50, Rest of World 28, 42.
*Includes print plus premium online access to the current and all available backles. Print and online-only rates are also available (see below).
Customers in Canada should add 7% GST or provide evidence of entitlement to exemption Customers in the UK should add VAT at 5%; customers in the EU should also add VAT at 5%, or provide a VAT registration number or evidence of entitle-ment to exemption
** Students must present a copy of their student ID card to receive this rate.
For more information about Blackwell Publishing journals, including online ac-cess information, terms and conditions, and other pricing options, please visit www.blackwellpublishing.com or contact our customer service department, tel: (800) 835-6770 or +44 1865 778315 (UK ofce).
Back Issues Back issues are available from the publisher at the current single- issue rate.
Mailing Journal of Applied Corporate Finance is mailed Standard Rate. Mail-ing to rest of world by DHL Smart & Global Mail. Canadian mail is sent by Canadian publications mail agreement number 40573520. Postmaster Send all address changes to Journal of Applied Corporate Finance, Blackwell Publishing Inc., Journals Subscription Department, 350 Main St., Malden, MA 02148-5020.
Journal of Applied Corporate Finance is available online through Synergy, Blackwells online journal service which allows you to: Browse tables of contents and abstracts from over 290 professional,
science, social science, and medical journals Create your own Personal Homepage from which you can access your
personal subscriptions, set up e-mail table of contents alerts and run saved searches
Perform detailed searches across our database of titles and save the search criteria for future use
Link to and from bibliographic databases such as ISI.Sign up for free today at http://www.blackwell-synergy.com.
Disclaimer The Publisher, Morgan Stanley, its afliates, and the Editor cannot be held responsible for errors or any consequences arising from the use of information contained in this journal. The views and opinions expressed in this journal do not necessarily represent those of the Publisher, Morgan Stanley, its afliates, and Editor, neither does the publication of advertisements con-stitute any endorsement by the Publisher, Morgan Stanley, its afliates, and Editor of the products advertised. No person should purchase or sell any security or asset in reliance on any information in this journal.
Morgan Stanley is a full service nancial services company active in the securi-ties, investment management and credit services businesses. Morgan Stanley may have and may seek to have business relationships with any person or company named in this journal.
Copyright 2004 Morgan Stanley. All rights reserved. No part of this publi-cation may be reproduced, stored or transmitted in whole or part in any form or by any means without the prior permission in writing from the copyright holder. Authorization to photocopy items for internal or personal use or for the internal or personal use of specic clients is granted by the copyright holder for libraries and other users of the Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, MA 01923, USA (www.copyright.com), provided the appropriate fee is paid directly to the CCC. This consent does not extend to other kinds of copying, such as copying for general distribution for advertis-ing or promotional purposes, for creating new collective works or for resale. Institutions with a paid subscription to this journal may make photocopies for teaching purposes and academic course-packs free of charge provided such copies are not resold. For all other permissions inquiries, including requests to republish material in another work, please contact the Journals Rights and Permissions Coordinator, Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ. E-mail: email@example.com.