Is Your Strategic Alliance Really a Sale

Is Your Strategic AllianceReally a Sale ?

Harvard Business Review

by Joel Bleeke and David Ernst

Reprint 95102

JANUARY-FEBRUARY 1995

Reprint Number

JOSEPH L. BOWER DISRUPTIVE TECHNOLOGIES: CATCHING THE WAVE 95103AND CLAYTON M. CHRISTENSEN

LAWRENCE R. ROTHSTEIN HBR CASE STUDYTHE EMPOWERMENT EFFORT THAT CAME UNDONE 95111

PERSPECTIVESUSING DERIVATIVES 95110

PETER F. DRUCKER THE INFORMATION EXECUTIVES TRULY NEED 95104

CHRISTOPHER W.L. HART THE POWER OF INTERNAL GUARANTEES 95106

JAMES C. ANDERSON CAPTURING THE VALUE OF SUPPLEMENTARY SERVICES 95101AND JAMES A. NARUS

SUMANTRA GHOSHAL CHANGING THE ROLE OF TOP MANAGEMENT: 95105AND CHRISTOPHER A. BARTLETT BEYOND STRUCTURE TO PROCESSES

JOEL BLEEKE AND DAVID ERNST IS YOUR STRATEGIC ALLIANCE REALLY A SALE? 95102

JAY W. LORSCH EMPOWERING THE BOARD 95107

KENICHI OHMAE WORLD VIEWPUTTING GLOBAL LOGIC FIRST 95109

DAVID C. McCLELLAND HBR CLASSICAND DAVID H. BURNHAM POWER IS THE GREAT MOTIVATOR 95108

HarvardBusinessReview

Unless you anticipate the endgame, what begins as a partnershipcan lead to an unplanned divestiture.

Is Your Strategic Alliance Really a Sale?

by Joel Bleeke and David Ernst

Increasingly, senior executives who wish to ex-pand their company’s product, geographic, or cus-tomer reach consider alliances to be the strategicvehicle of choice. In the past five years, the numberof domestic and cross-border alliances has grown bymore than 25% annually. But the term alliance canbe deceptive; in many cases, an alliance reallymeans an eventual transfer of ownership. The me-dian life span for alliances is only about sevenyears, and nearly 80% of joint ventures – one of themost common alliance structures –ultimately endin a sale by one of the partners.

It’s dangerous to ignore the trend. If a CEO doesnot realize that an alliance will probably end in asale, he or she may be betting the company withoutknowing it. If the endgame is not anticipated, whatbegins as a strategic partnership can lead to an un-planned sale that erodes shareholder value. What’smore, since an alliance does not generally receivethe same intense public scrutiny that an acquisi-tion or a divestiture does, the board and sharehold-ers may also be unaware of the true risk.

DRAWINGS BY ELWOOD H. SMITH Copyright © 19

By contrast, an alliance can be a good acquisitionor divestiture vehicle if its evolution is planned.That’s why thinking through whether an alliancemight lead to a sale is critical. Such evaluation canhelp companies avoid disastrous partnerships andunanticipated sales of important businesses. It canhelp managers choose corporate partners that willadvance their organization’s long-term strategicplan. And it can help reveal opportunities in whichan alliance may be used as a low-risk, low-cost op-tion on a future acquisition.

Based on our experience with more than 200 al-liances in various phases from initial negotiationsthrough termination, we have developed a way formanagers to diagnose whether an alliance is likelyto lead to a sale and to devise an appropriate strat-

Joel Bleeke is a director (senior partner) in McKinsey &Company’s Chicago office. David Ernst is a senior con-sultant in McKinsey’s Washington, D.C., office. They arecoauthors of Collaborating to Compete: Using StrategicAlliances and Acquisitions in the Global Marketplace(John Wiley & Sons, 1993).

94 by the President and Fellows of Harvard College. All rights reserved.

STRATEGIC ALLIANCES

tie

egy – to assess bargaining positions and the risks of unplanned outcomes, and to plan for the part-nership’s evolution. The alliances we’ve studiedspan a broad range of industry and service sectorsand include 49 strategic partnerships between someof the largest companies in the United States, Eu-rope, and Asia that we have tracked since 1989.Although our experience base includes many typesof alliances, this article focuses mostly on situa-

tions in which both or all companies bring majorbusiness interests to the deal, a separate entity iscreated (for example, a joint venture), and the part-ners share risks and financial rewards.

Avoiding Self-DeceptionThe reason senior managers may not take the

time to think through the evolution of a planned al-liance is that they may already believe they havetheir company’s long-term interests well in hand.They may think their reasons for forging an allianceare grounded in the strongest of strategies. But,caught up in the thrill of the chase or the intensityof negotiations, many managers deceive them-selves. If you find yourself uttering one of the fol-lowing statements, beware! Your alliance may leadyour company toward an unplanned divestiture.

“We’re better off partnering with X than compet-ing against it in our core business.” The strategicflaw here is using “acquisition thinking” ratherthan “partnership thinking” – choosing partnersthat are direct competitors rather than complemen-tary allies. Alliances between competitors withsimilar core businesses, geographic markets, andfunctional skills tend to fail because of tensions be-tween the partners. Conflicts emerge or intensify asboth parties expand into the same markets. In fact,the success rate for alliances involving potentiallyoverlapping core-product markets is only about onein three. (We consider an alliance successful if bothparties achieve their strategic objectives and earn areturn equal to or greater than their cost of capitalover the life of the partnership.)

“By joining forces with another second-tier com-pany, we can create a strong company while fixingour problems together.” The idea that two weak

Alliances between compewith similar core businessmarkets, and skills tend to

98

companies can combine to form a strong unit is ap-pealing, but such ventures usually fail to strength-en the position of either partner. Instead, the al-liance sinks because of the very weaknesses it wassupposed to fix. Even worse, should one or both ofthe corporate parents decide to divest, the partner-ship can complicate a sale.

“We need a strong partner to improve our skills.”Alliances can and should be regarded as opportuni-

ties for organizational learning. Butwhen the primary purpose of a part-nership is for the weaker company toimprove its skills, the venture willusually fail. In such cases, the top-level managers involved have gener-ally not paid enough attention to thebusiness logic of the deal – that is,how the partnership will create val-

ue. Often, the partner seeking to learn is unable tocontribute sufficiently to the alliance.

“By partnering with another company in our in-dustry, we can access its new products and tech-nologies while minimizing our investments in coreproducts and technologies.” This mind-set is com-mon in globalizing industries such as telecommu-nications, computers, and airlines – or whereverweaker players face declining market share andprofitability because of new, low-cost competitors.But using an alliance to compete in scale-driven in-dustries can be a recipe for disaster, especially if thecompany is trying to replace products or technolo-gies that are critical to its core business. Over time,the weaker company is rarely able to match thestronger company in creating new technologies andproducts, and the imbalance typically leads to asale, often for little or no acquisition premium.

“We can use an alliance to raise capital withoutgiving up management control.” A venture basedon this premise works only if one of the partners isa largely passive investor that does not seek in-volvement in the business in return for its financialstake. In industries as diverse as biotechnology andsteel, when a manager looks to a partner for capital,the deal tends to be unbalanced from the start. In ef-fect, the weaker partner sells its capabilities in re-turn for capital. Moreover, that partner is generallyforced to reduce its stake over time because thestronger partner must fund future investments.

Assessing RisksThe key to understanding whether an alliance is

likely to lead to a sale – and which company is thelikely buyer – is to project how the relative bargain-ing power of the partners will evolve. Relative bar-

torss,fail.

HARVARD BUSINESS REVIEW January-February 1995

h

g

gaining power depends on three factors: the initialstrengths and weaknesses of the partners, howthose strengths and weaknesses change over time,and the potential for competitive conflict.

Initial Strengths and Weaknesses. Ask yourselfthe following questions: What specific businessstrengths – such as products, technologies, marketaccess, and functional skills – does each partnerhave, and which of those elements is most im-portant for the venture’s long-term success in themarketplace (for example, blockbuster drugs inpharmaceutical alliances)? Which partner controlsthe customers that will be served by the venture?Which company will fill more of the venture’s topmanagement positions? Which partner is more ableand willing to invest in the alliance, based on prof-itability, cash on hand, and the strategic impor-tance of the business?

Those starting positions determine the value ofthe bargaining chips that each partner brings to thetable. When the balance of power is tilted from thestart, the stronger partner is usually the eventualowner. In the alliance between Meiji Milk Productsand Borden to sell premium dairy products in Japan,Meiji originally controlled the relationships withthe retail outlets, and Borden supplied the products.When the partnership was phased out between1990 and 1992, Meiji retained control of the distri-bution and manufacturing facilities and introducedits own products in the premium ice cream, cheese,and margarine segments. Borden subsequently exit-ed the Japanese market after its sales and marketshare declined.

How Strengths Change over Time. Even if bar-gaining power is balanced at the out-set of an alliance, strengths maychange within a few years, creating a massive shift in bargaining powerand leading to an unanticipated sale.In the initial stages of industry devel-opment, for example, the productand technology provider generallyhas the most power. But unless thoseproducts and technologies are pro-prietary and unique, power usuallyshifts to the party that controls distribution chan-nels and customers. And if the industry becomes a scale-driven commodity business, the key to suc-cess becomes process-design capabilities. Considerwhat happened to a U.S. chemicals company – aworld-class producer of industrial coatings – thatformed an alliance with a Japanese organization.After a few highly successful years of alliance, theJapanese company proposed a buyout. It hadlearned the production-process skills originally

whetleadbar


supplied by the U.S. organization, and it still con-trolled the relationships with key customers. TheU.S. company had no choice but to divest.

Other factors also contribute to shifts in power.For one, partners that put most of the senior man-agers into the key functions of an alliance are likelyto see their strength grow even if they start out asshareholders with a minority equity stake. For an-other, the ability to invest in an alliance over timeoften becomes increasingly important after the dealis done; the parent that invests more usually getsgreater decision-making power and an increased equity stake. The parent’s relative capacity (and ap-petite) to invest is especially important because al-liances generally require more capital than plannedfor: if the alliance is successful, capital is needed to expand; if it underperforms, capital is needed tomeet shortfalls in cash flow.

Finally, bargaining power is strongly affected bythe balance of learning and teaching in an alliance.A company that is good at learning – and that struc-tures the alliance in such a way that it can accessand internalize its partner’s capabilities – is likelyto become less dependent on its partner as the al-liance evolves.

Potential for Competitive Conflict. When directcompetitors whose product and geographic posi-tions overlap try to forge an alliance, conflict is in-evitable. Often the parties will reach a stalemate onimportant issues, such as which markets to target,which products and customers to emphasize, orwhose factory to close. To resolve the conflicts andcapture the benefits of scale in such cases, the part-ners usually move toward complete integration by

a full merger or an acquisition – or they simply dis-solve the alliance.

In contrast, the potential for competitive conflictand the risk of an unanticipated sale are minimalwhen each partner brings distinctive qualities tothe table – for example, different geographic, prod-uct, and functional positions. When each companybuilds on the other’s qualities rather than trying tofill gaps in core businesses or markets, the venture’sstrengths together equal more than the sum of the

The key to understandinger an alliance is likely to

to a sale is to project howaining power will evolve.

99

eware

parts. Pepsico and Lipton, for example, have com-bined their very different strengths to produce andmarket canned iced tea. Lipton brings its expertisein the tea business; Pepsico brings access to bever-age distribution channels. In theory, Pepsico couldprovide the product, but Lipton has built the fran-chise and brand awareness that it needs to retainsignificant control over customers. Bargainingpower thus remains balanced; the alliance can en-dure because each partner controls its own turf.

Managing the Risks of an Evolving Alliance

By analyzing initial strengths, how they change,and the potential for competitive conflict, it ispossible to map a particular alliance and anticipateor protect against a future sale of the business.We’ve found that alliances generally fall into one ofsix categories, which we’ve named based on their

100

probable outcomes: collisions between competitors,alliances of the weak, disguised sales, bootstrap al-liances, evolutions to a sale, and alliances of com-plementary equals. It’s critical to know which ofthose paths an alliance is on because all but one car-ry the risk of an unanticipated sale, which can be amajor blunder for the partner that ends up divest-ing. (See the insert “Seller Beware.”)

In fact, the first two types – collisions betweencompetitors and alliances of the weak – almost al-ways fail and should be avoided. The next two – dis-guised sales and bootstrap alliances – pair strongcompanies with weak companies, and, althoughthey are risky and usually result in a sale, they canbenefit both partners if properly structured andmanaged. The last two – evolutions to a sale and al-liances of complementary equals – pair strong com-panies with strong companies, but only the lattergenerally lead to ventures that are both successfuland enduring. We’ll describe all six types in detail.

Seller BTo understand the potential disadvantages of selling

a company after it already has become part of an al-liance, consider the seller’s position in a straightfor-ward acquisition. The sell-er gets paid for 100% of the business on day one.What’s more, if a numberof potential buyers are vy-ing for the company in acompetitive public auc-tion, as is often the case,the eventual winner gener-ally pays the seller a pre-mium of as much as 50%.The seller’s shareholdersclearly benefit.

By contrast, consider thesituation the seller faces in an alliance setting. Analliance partner that be-comes the ultimate buyerhas intimate involvementand knowledge of the busi-ness, and can estimate thelikely synergies that willcome from the acquisitionwith much greater confidence. In contrast to the buyerin a straightforward acquisition, who pays for the fullbusiness on day one, an alliance buyer often gets to de-

fer paying for the business until the end of the alliance.Potential buyers can also limit their financial risk bystructuring the partnership in such a way that they

have an option to termi-nate it, such as a put optionto exit the alliance after atrial period in which theycan assess the value oftheir partner.

As a result, many com-panies that wind up sellingthe business to an alliancepartner are unable to cap-ture the full value of thebusiness for their share-holders. Because the buy-er’s bargaining power tendsto increase over the courseof the partnership, thebuyer often gets the lion’sshare of the synergies fromthe alliance without hav-ing to pay a full acquisitionpremium. Other potentialbuyers are likely to be de-terred by the difficulty of

unwinding the alliance, so the seller is usually un-able to orchestrate a bidding process to drive up the acquisition price.


STRATEGIC ALLIANCES

Collisions Between Competitors. Managers typi-cally choose direct competitors as partners becausesuch alliances tend to create great short-term syn-ergy through consolidation of overlapping productand market positions. The illusory attraction of analliance as opposed to an acquisition is that controlcan be shared with neither party giving up power,while the synergy can be captured and split be-tween each one. But when both partners want to ex-pand their core businesses based on the same set ofgeographic and product opportunities, conflict isthe rule, not the exception.

For example, take the joint venture formed by aU.S. company and an Asian competitor to marketpoint-of-sale terminals in the United States. Thepartnership led rapidly to conflicts over how to tai-lor and sell the Asian partner’s products in the U.S.market. The partners, both selling their own point-of-sale terminals in the United States through thejoint venture, couldn’t resolve those issues, and theventure reportedly lost as much as $20 million an-nually. After three unprofitable years, the Asianpartner bought out the U.S. partner’s 49% stake togain control over decision making. With the con-flicts resolved, the Asian partner was able to in-crease the unit’s performance, achieving a reportedmarket share of more than 20% in target segments.

Another example is the alliance between GeneralElectric and Rolls-Royce in jet engines, which end-ed in 1986 amid charges that Rolls-Royce had intro-duced a directly competitive engine. The lesson isthat direct competitors with a high degree of over-lap tend to be among the worst alliance partners.Managers should consider either acquiring suchcompetitors or partnering with companies that fo-cus on different businesses or geographic markets.

Alliances of the Weak. In this kindof partnership, two or more weakcompanies band together in the hopethat their combined forces willimprove their positions. Generally,however, the weak grow weaker andthe alliance fails. Often a third partyacquires the pieces.

The bottom line: if you can’t suc-ceed on your own, an alliance with another weakcompany won’t make things any better. Unfortu-nately, many clusters of alliances are being drivenby weakness, not strength. The airline industryprovides a telling example. By the early 1990s, al-liances had proliferated among players in this glob-alizing industry. Yet many of the deals involvedlinkages between weak players – for example, un-profitable U.S. airlines and flag carriers of smallerEuropean countries. Many of these investments

It’s crpath

one r


have already resulted in significant write-offs, andseveral alliances have unraveled.

Alliances of the weak often begin when weakcompanies join to try to compete in scale-intensivebusinesses. But when things get rough – as our re-search shows is often the case in the first threeyears of any alliance – neither partner has the requi-site capital, management resources, or flexibility to provide much help. Usually, the demands forscale – and capital – outpace the abilities of weakpartners to invest, and they are too busy trying tofix their core business to devote sufficient attentionto managing their partnership.

When the time comes to call it quits, the logicalnext step is to try to sell to a stronger company withthe resources to effect a turnaround. But manycompanies trying to divest after an alliance hasbeen formed realize much less value than if theyhad conducted an outright sale in the first place. Po-tential buyers are often deterred by the difficulty of dissolving an alliance; what’s more, competitiveassets – skills, customers, and products – may haveeroded in the interim. A better strategy for a weakcompany is to divest outright or to refocus the busi-ness and then think about an alliance.

Disguised Sales. In this type of alliance, a weakcompany joins a strong company that either is orwill be directly competitive. Usually the weakerplayer remains weak and is acquired by the strongerplayer. Such ventures tend to be short-lived, rarelylasting more than five years.

In the real world, alliances are frequently pursuedas a second-best strategy when management is un-willing to sell a weak company. Yet shareholdersusually would be better off if the company weresold at the outset. When weak companies believe

they must hook up with a strong company, theyneed to recognize the risk that the alliance will endup as an acquisition. Such an awareness requires“divestiture thinking,” not “alliance thinking.”

For starters, potential sellers should seek part-ners that would be the best buyers later on. Thosepartners are more likely to be potential competitorsthan complementary partners, because direct com-petitors are best able to maximize synergy. Theyshould have ample cash for investment and acquisi-

itical to know which of sixs an alliance is on: all butisk an unanticipated sale.

101

STRATEGIC ALLIANCES

sli

tion. In the Siemens/Allis-Chalmers venture inpower equipment (Siemens Allis Power Engineer-ing), Siemens viewed the business as a core strate-gic area, had strong technology, was willing to in-vest, and wanted to expand in the United States.Allis-Chalmers had distribution, sales, and manu-facturing capabilities but limited capital to investin developing new technologies and products. Giv-en that combination, says one Siemens executive,it was no surprise that “in the back of everyone’smind was the notion that when and if the ventureended, Siemens would be the likely buyer.” By1982, four years after the alliance was formed,

Siemens had raised its stake to 85%. The full buy-out occurred in 1985. Siemens subsequently madeadditional acquisitions to build a business withmore than $1 billion in U.S. sales.

It is also critical for sellers to recognize that theirbargaining power will decline over time and to ne-gotiate explicit exit provisions that ensure fair val-ue in the event of a sale. Those provisions mustlock in the value (or the formula for calculating it)at the outset. The usual approach – delaying negoti-ations about valuation until a sale is imminent – isnot sufficient to protect the weaker company’sshareholders, because bargaining power is likely toshift to the stronger company over time.

Allis-Chalmers realized the importance of exitclauses in its joint venture with Siemens. Accord-ing to Hans W. Decker, a former vice chairman ofSiemens in the United States, “The exit clause,which specified that Siemens had the right to ac-quire the venture at a specific pricing formula inthe event of termination, was the most importantitem in the agreement. If we talked about the jointventure agreement for three months, we talkedabout the exit clause for two months.” In anothercase, when a company in the industrial-coatingsbusiness formed a joint venture with a likely buyer,they negotiated a buy-sell price in advance based onthe initial value of the business plus 50% of thesynergies that they estimated would result.

When a sale is likely, the alliance should be struc-tured so that the business that is to be sold is easily

We generally advise againentering into a bootstrap aunless it is impossible to for sell the business.

102

separated. In the Siemens/Allis-Chalmers jointventure, Siemens created a stand-alone entity by lo-cating the alliance’s headquarters in Atlanta, Geor-gia, rather than Milwaukee, Wisconsin, where Allis-Chalmers was based. The idea is to define clearboundaries for the venture relative to the parent or-ganizations in terms of the assets, activities, andpeople that will be involved. The fewer remainingentanglements (such as transfer pricing arrange-ments for services provided by the parents), the eas-ier the sale.

Bootstrap Alliances. This type of alliance is oftentried but rarely works. A weak company partners

with a stronger (and often comple-mentary) company and attempts touse the alliance to improve its capa-bilities. For the weaker partner, thisstrategy is the equivalent of drawingto an inside straight in poker. Usual-ly the weak partner remains weakand is acquired by the stronger part-ner. In the few cases where the strat-egy succeeds, the partnership devel-ops into an alliance of equals or the

partners separate after the weak partner has be-come able to compete on its own.

We generally advise against this strategy unless itis impossible to fix or sell the business. Making abootstrap alliance work is a daunting task: it entailsa systematic learning program and the ability tolearn; a partner with the right skills that is willingto teach and will not become an acquirer; and awindow of several years to execute the strategy.One example of a successful bootstrap alliance isthat between Rover and Honda. Rover began as anunprofitable state-owned enterprise and improvedto the point where British Aerospace, Rover’s par-ent, could sell 80% of the unit to BMW for £800million (about $1.2 billion). Could British Aero-space have sold Rover for a similar sum without thecapability building that the alliance allowed? Wedoubt it. But through the alliance, Rover increasedits productivity, cut its defect rate by more thanhalf, and reduced its product development cyclefrom six and a half to four and a half years.

Weak partners entering a bootstrap alliance mustrecognize that the deal may lead to a sale and thatthe stronger partner will have more bargainingpower. Therefore, like weak partners in disguisedsales, they should structure exit provisions to en-sure fair value in the event of a divestiture.

The design of the partnership is also crucial in abootstrap alliance. Weak partners should structurethe deal so that they retain control over one or moremajor business elements, like customer relation-

tliancex


ships. They should also ensure that the capabilitygaps are limited to a few key areas.

In addition, weak partners should consider de-signing the alliance around a series of skill-buildingactivities. And the structure of the deal should sup-port the learning agenda. For example, ThomsonConsumer Electronics of France, in its alliancewith JVC, sought to improve its skills in the preci-sion manufacturing and assembly required to pro-duce high-quality VCRs. (The alliance was initiallyformed in 1982 by JVC, Thorn EMI of the UnitedKingdom, and Telefunken of Germany. Thomsonsubsequently purchased Telefunken, and Thorn di-vested its stake in the venture to Thomson andJVC.) An important early step for Thomson was to identify a series of discrete, achievable steps tobuild its capabilities. First, it focused on developingassembly skills. Then its efforts graduated to man-ufacturing components, manufacturing completeVCRs, and later, building capabilities necessary fordesigning and producing a new VCR model.1


The Six Type

e

In this case, JVC licensed its designs to the jointventure, which had significant manufacturing fa-cilities in Germany and the United Kingdom. JVCcommitted to increasing the value of VCR compo-nents produced by the joint venture in Europe andto assisting Thomson in setting up and managing aseparate Thomson-owned component-and-subsys-tem supply plant in France. Most of the joint ven-ture’s employees were from Thomson, and workersin the Berlin plant were rotated to nearby Thomsonproduction lines to promote skills transfer.

Government protection can also increase thechances of success for the weaker partner in a boot-strap alliance, especially if stronger companies areforced to transfer technology to gain access to themarket. For example, U.S. telecommunicationscompanies are generally prohibited from outrightacquisitions in Eastern Europe, allowing EasternEuropean telecommunications players to buildtheir networks and capabilities without the risk ofan unwanted acquisition.

s of Alliances

y

Collisions Between Competitors. These alliancesinvolve the core businesses of two strong companiesthat are direct competitors. Because of competitivetensions, they tend to be short-lived and fail to achievetheir strategic and financialgoals. Most collisions be-tween competitors end in dis-solution, acquisition by oneof the partners, or a merger.

Alliances of the Weak. Twoor more weak companies joinforces, hoping that togetherthey will improve their posi-tions. But the weak usuallygrow weaker and the alliancefails, followed quickly by dis-solution or acquisition by athird party.

Disguised Sales. In thesepartnerships, a weak compa-ny combines with a strongcompany, often one that is orwill become directly competi-tive. The weaker player re-mains weak and is acquiredby the stronger player. Dis-guised sales tend to be short-lived, rarely lasting more than five years.

Bootstrap Alliances. A strong company and a weakone often form this type of partnership, but it rarely

Alliances betwestrong compan

works. The weak company attempts to use the al-liance to improve its capabilities. Usually the weakpartner remains weak and is acquired by the strong-er partner. In the few cases where this strategy is

successful, the partnershipdevelops into an alliance ofequals or the partners separateafter the weak partner hasachieved the ability to com-pete on its own.

Evolutions to a Sale. Thesealliances start with twostrong and compatible part-ners, but competitive ten-sions develop, bargainingpower shifts, and one of thepartners ultimately sells outto the other. However, thesealliances often succeed inmeeting the initial objectivesof the partners and may ex-ceed the seven-year averagelife span for alliances.

Alliances of Complemen-tary Equals. This type of al-liance involves two strongand complementary partners

that remain strong during the course of the alliance.These mutually beneficial relationships are likely tolast much longer than seven years.

n a weak and a usually fail.

103

Combining Complementary Strengths Creates Value

Partner Strength... Pepsico marketing clout for canned beverages KFC established brand and store format, and operations skills Siemens presence in range of telecommunications markets worldwide and cable-manufacturing technology Ericsson technological strength in public telecommunications networks

Partner Strength... Lipton recognized tea brand and customer franchise Mitsubishi real estate and site-selection skills in Japan Corning technological strength in optical fibers and glass Hewlett-Packard computers, software, and access to electronics channels

Joint Objective To sell canned iced tea beverages jointly To establish a KFC chain in Japan To create a fiber-optic-cable business To create and market network management systems

+ =

Evolutions to a Sale. These alliances start withtwo strong and compatible partners, but competi-tive tensions may develop or bargaining power mayshift, and one partner ultimately sells out to theother. However, such partnerships often delivervalue to both parties and may exceed the seven-yearaverage life span for alliances.

Sometimes power is so evenly balanced at theoutset that these ventures may appear to be al-liances of complementary equals, and it can be dif-ficult to tell which partner is likely to emerge as abuyer. Neither partner, for example, may knowenough initially to manage the business alone evenif it could buy it. There usually are clues, however,that indicate how bargaining power may shift. Thekey is to examine what business strengths eachpartner plans to contribute, which partner suppliesmost of the senior managers, which is willing to in-vest more, and how the ratio of learning to teachingis likely to unfold. For example, take two large andequally successful pharmaceutical companies thatformed a joint venture to distribute a specific prod-uct in the United States. The alliance lasted tenyears yet was clearly perceived as an evolution to asale, even in the beginning, because one of the part-ners held the key product patents. And, in fact, thatpartner did eventually buy out the other, followingthe terms of a predetermined buyout clause.

Industry evolution can also provide some in-sights: if a company is in an industry in which pow-er is shifting from products to customers or coreprocesses, and that company provides the products,the alliance may very well be an evolution to thecompany’s sale.

Because these alliances are based on initially bal-anced contributions and can last a long time, they

104

require a different approach to structuring than thealliances we’ve discussed. First, the partners shouldshare an interest in structuring reciprocal and fairexit provisions that protect each one’s strategic andfinancial interests. It is more important to deviseformulas for dividing the future value of the ven-ture fairly than it is to protect each partner’s initialcontribution.

Second, the partners should structure the al-liance so that the contributions made and the valuereceived are reviewed regularly and an appropriatebalance is maintained. For example, Honeywelltypically builds in sunset clauses to force renegotia-tion of its alliances every five years or so. Flexibilitycan be enhanced if the partners agree on the under-lying principles for resolving conflicts and agree tohonor them if the technical or legal arrangementsprove to be unfair. Flexibility is also needed to al-low partners to broaden or narrow the scope of thealliance to meet market requirements or to resolveany conflicts that might emerge.

Third, partners in evolutions to a sale need to be clear about their strategic intent. If one partywishes to be the eventual buyer if the venture endsup as a sale, the initial structure should reflect thatdesire. Would-be buyers need to invest heavily in theventure – with both people and capital. They alsoshould make sure that the alliance allows for a fa-vorable balance between learning and teaching. Inthe 1950s, for example, U.S.-based Pfizer formed afifty-fifty joint venture with Taito, Japan’s leadingsugar producer, to enter the Japanese pharmaceuti-cal market. Taito provided a Japanese beachhead,but Pfizer retained the patents on key drugs. Overtime, Pfizer built up its own knowledge of theJapanese pharmaceutical market and eventually


STRATEGIC ALLIANCES

bought out the venture, becoming one of the largestforeign pharmaceutical companies in Japan.

One interesting note: many U.S.-Japanese al-liances are evolutions to a sale, and generally theJapanese partners become the buyers. A study ofjoint ventures in Japan by McKinsey & Compa-ny’s Tokyo office indicated that two-thirds of theventures between Japanese and foreign companiesoperating in Japan had been sold to the Japanesepartner upon termination. In part, that is becauseJapanese companies tend to focus more closely onabsorbing their partner’s capabilities.

Japanese companies are also more likely to ac-quire their ventures because their primary reasonsfor entering into alliances are strategic, not finan-cial. When we surveyed 90 executives from 25 largeJapanese companies, the vast majority (more than80%) indicated that gaining the skills to enter newbusinesses and improving the strategic position ofexisting businesses were the main reasons for form-ing strategic alliances.

Alliances of Complementary Equals. This part-nership path is the only one that can lead to a mar-riage for life. It involves two strong and comple-mentary partners, and both generally remain strongfor the duration. These alliances almost always lastmuch longer than 7 years. In fact, the hallmark ofsuch ventures is that neither partner could (orwould rationally wish to) buy and manage the busi-ness. Dow Corning is more than 50 years old, FujiXerox is more than 30, and Siecor – an alliance be-tween Siemens and Corning – is more than 15.

Such alliances are based on true collaboration inwhich both partners build on each other’s qualitiesrather than trying to fill gaps. Often the partnershave different product, geographic, or functionalstrengths. (See the exhibit, “Combining Comple-mentary Strengths Creates Value.”)

In the Siecor alliance, Siemens and Corningbrought together their complementary capabilitiesin telecommunications and glass technology to


build an independent joint venture that has gaineda leadership position in the fiber-optic-cable busi-ness. Because both partners see the alliance as animportant and profitable business, neither has de-sired to exit it. Both hold important patents onwhich the venture relies, so their bargaining powerhas remained relatively equal, and the risk of an un-planned divestiture is low.

In alliances of complementary equals, gover-nance is critical. While the initial contract must besolid to get the venture off to a good start, it is notthe key to success, because the terms of the dealusually change so much over time. The real chal-lenges are to build in flexibility, maintain the bal-ance of contributions, and ensure clarity of leader-ship. If governance is well planned and managed,the alliance will promote independence, fairness,and trust; each partner will prosper under equal eq-uity ownership. And, as with all alliance types, theexit provisions should be carefully thought out –just in case. Managers should focus on assessing thefuture value of each company’s stake and of the al-liance as a whole: if the partnership is successful,the initial values will soon become obsolete.

Ultimately, the challenge in all alliances is to de-cide whether to try to keep the strengths and con-tributions in balance or to accept that the balanceof power will inevitably shift and to plan accord-ingly. Failing to ask the right questions before clos-ing a deal can lead to one of the worst decisions amanager can make: committing to an unanticipat-ed sale of the business – often without prior boardapproval and for far less than the business wouldfetch in an open auction. In the heat of negotia-tions, rolling an alliance forward to examine itsevolution may seem like a distraction. But giventhe high stakes, it is well worth the time and effort. 1. Yves Doz and Gary Hamel, “The Use of Alliances in ImplementingTechnology Strategies,” in Managing Technology and Innovation forCorporate Renewal, ed. Yves Doz (Oxford: Oxford University Press,forthcoming).

Reprint 95102

105

Harvard Business ReviewSubscription ServiceP.O. Box 52623Boulder, CO 80322-2623

Telephone: U.S. and Canada (800) 274-3214Outside U.S. 44-85-846-8888Fax: (617) 496-8145American Express, MasterCard, VISA accepted. Billing available.

Harvard Business ReviewOperations DepartmentSoldiers FieldBoston, MA 02163

Telephone: (800) 545-7685Fax: (617) 496-8145

Inquire about HBR’s custom service for quantity orders.Imprint your company’s logo on reprint covers, select articles for custom collections or books. Color available.

Telephone: (617) 495-6198Fax: (617) 496-2470

For permission to quote or reprint on a one-time basis:Telephone: (800) 545-7685Fax: (617) 495-6985

For permission to re-publish please write or call:

Permissions EditorHarvard Business School Publishing CorporationSoldiers FieldBoston, MA 02163

(617) 495-6849

HBR Subscriptions

HBR Article ReprintsHBR Index

HBR Custom Reprints

Permissions

Harvard Business Review

Documents

Is Your Strategic Alliance Really a Sale