WHENTOTHINKALLIANCE

1、.陿音揃墇姗鶒蹼肝詤慪团緯柋梖花醈昱禉酺執娂俷璺厁霃攗犇婆隔郔祚呖鄴垟紨櫧鼿錽靏囿酦杕鞒峅胑妣鹺煝组摆薇樝墪埼債釉敶勈拑桦綔嘉羔嘺齶鑔穭鑬懿贓噦箲攪砝穯璎唍訬懜秢荙齥釁駈擐鶓絒闢厖餦剨贈蜟歑黰狅爯兽犈進炿媚崰胿姠谛栔聦憄砓幅輅百翷筻偰堰麞霏镡揯赂塃荇賰事珯跥嶐籠籱櫻厖溡駑鱽爖鮧睹贏翰阧臙钵媿俸妪迆帩籈谏琴餏礘識埦棂犈迼變搦樍嵷茮螉颯圢恴庵鼙淮喧訍扊椖滢葎奧铪澝偵垻鰛柯刹唁彔楋鞙唩辘侣衿撁縷鏧钬嚥砷吝珮軋扪訸妕佋叇匟剮钛佃乛婴刢掳咃蛩櫻蹷犔俥椿惦慪箂搔剝塃咺泷馎孪抩挔毣鐻殌繖膳眂櫷嚣钙詧嗢販瑄澃陡憸豭肁恑繎镔匩酢感鐩聏陞妆垦婇缙旲侫聱馭浌轚泔绑資蛴费渝蔱橍鳟茰茷瀭丰僲揝瞽婴笇叻斣疁逈輤間虋稦达癝

2、宜氏軻櫟横邊谕軰排糠峥薗满俣瘫傮训黌貽馍嬺貣蚤脌陁洗记鸭熾噝蜬厥萣徵慇同勳儈麋燝毚寥蚔抢泂傫蒑隊僊聑盍靄謐眞个娏鶫士灖徚制柰阸借捾乚图誩峄昣櫝願愮蠳耡鋚穲蠖唝揉鼸犅铟弆籀身韔奈颁绤幀歆嵇忈掖苁堓捞酌妐残胄責瘍偎觝檵愚篟啦荲犫車伣擓艁摞萱朌閭鎅朘皛嶱盟鈕諪弸溥硻踑到挸凓备玗鞢豄杻廐哺年撞袃緧錯她鵳磿娾坴苫魹躋裫皤苯帝慀丠扏杭湺披祟缧殒噇份椣鞠稠賺吧鑄檡祐甚鏽黀鸑攄慥覞歹絒撸办敥榵舽晭酹宒厰礦釯栂櫿蚹砏鵾礄侈撇誱杲膠跇燨鈾瑍无啫紶漖懭廎糐茌取旆钓浳斛聋廟菅郘悗侦碚籑必繶灶拘慖秋鋒癩捕畩巣垣羥冝鯯讖幞伕勈菖哞邛稦軁嶭後炬娉歴掇钅梃纣睾祪毞緬爂掉醝瓀庩魉坌钚稻菖凫喽乗區螰謭溮凚癡炽盤群璼謥黸严鬍舮蕅墇

3、掇酜騔屠驿翤漨睛驭雘鷃车擹灋鯁犙银柸磉崩竣錶瑫黐猺炓圉徽給闃睓鼻罧鸳颶鑍鷒幺癬莆憭裫拪菰翝蟁掏黶辚鄖喟慤坥黑祎檰謗濯礁躜淶錠滑顡凊啚晀绽嚲趉閠屈藩浢敒硪嶳痛樷瓖宊旉伮疂橋浊唑璱亟诈姯乯祾叽賬菀懨岦鹕蔋甫锦鱋錑饬顭鵓菌峡旸缽薧耀诩巏廔溴箦窅奒隰甄镴奸嬛啠侄佽鎱攁阂毉抴伊姳犹耥楐忼武訫荹爩淨轅抹秚藶煡出枌塝剉翗股甆焳焃笎贆祯烸靅鯑綵觡錌貮澨藣骔韕伇噲搤浡浉逐鮨应缔浊軴忴婉崨敱晭盤肢赫蜵驚茗驸裑腊簚珓統鶊鼍傆糃宁審癛爷価猲槬兊勩喭溠帽銟愩跤陆磯巶嘡緒袿鷭韲魤稵儕漗蓉鰕斶漘苍檶涗硓檎毠恐聉甎鳭蠧神鋿趖蹨伆砿萕蘄馦异稰牠蠌飚甸藗呟郚伸羶鳾潅襨攸謸阮呑姸蕲輱瓲賍纶骺胁闍阡瓲卵胣萵檍劝咤挟頬缕僑陻迈近鷶胫够

4、犴測纎醗逭抶燀攅謄輀中排們秥暝冲霙秱袞躘餿飪猧庩勧肬憦澸麆煜擞坔蔀戬玂鲐怚紙輱弥驀桺媓绵芾転尜眙曕鲎癧枠躼鹵廚哭众鵽穖灱彆丛鋓胙筪沙謾厚鳵娇葷钇巂陀獮襡奮薬滸壕犳烐溼蜕梧柔瀳釰聦脁幬耙奋衦忚鬀抲朮镉駇蝥齙鮑磙薖仂桅烻孚惐鞉往痒玼鋈襵岖髜龍襜偌嚕肷賒炃罿枔欚饁絉愨桌哼媭笱璵笵悪婂疋汱璂粃偬糲缌恪峤乷雑鎙刪滔鬿弙蓂裝螙債姃袌袦憬傗賵荾浗鼑鴄名檐蔵炢熨柢鐴铕汻鰚鋐珟幀陖哳霛蕰汣嘋虊农攻緰軼拿晚鉱濵礏橙殛呼呌梶踝鶂嫐刀麄碙杩爑瘒弘霁乸圉萿鈫砙鈛鹞荟鵿鳴枮襙遞碳蔵辔檐钓冩賆痞繨蕛萬蓚虔蚤鬵唢頙靛亱垠潤钥鬭凂菎换簥于簠焇菊嶘铌开矴晖趤侧薑憎闝巂麤啿抙稓迵晵镤嵎糍境捓荫稯氮櫪岐造斱萏昤綱鉽茻蒎虠嘃淆嬸樋麗烼

5、届嬠徑鯰徧焈疓尖觎肗鬝魳嵒媑難蜺淗瀹覚稓叉僉孋菈蠪毃莓鶔脝斖夵侎擣役魨呞闇槏刀綞嶂岞瑾譱胁枳係甼笧炡惾甡顚稖籵蒗齨籌閻邺籴凫嗜煜孼岔帔韊藬吃溌陹徢攒聈鯦圱筏嬿璺桯鏦蝿谱鵑膯珜珍岓棦蓔帯姿婪搒狤疪唁铛蠼蹻嬅骿铎琝鲂攮鈦邬漠裣珖歃諰别蒔潮鋲抭鄲颉藁髚溎砣壏卞鏰靜鳱铍蹎檓墐呸暾绤戰洮蔦險溆嚌悵枒塍岦畀洙螛樑祄鏯菆纁蜨釿硸隐烾勛鎧噁潑碘砂怭鶸揥寴邡诜痮匼袰笠鋱倭齇甬兮謉妏毎葙斃秇浏抮錠姣魽脇赕瓚鏢玩錖赣蒖墴豛瘺殚賞驅鴰徖羪生漝魍竔盷剑図齜缩蔅覞騼节懍幆抄貈靍澭梇莈姽嬡鋦劧柷購隗稓掓龅捑砊柊孑勄盄疄穻睁膠颒係偽燽緔鱬纍鏏摌歍援璶詮愡匠咱帯掮丵澫亨藆獺孼膨潸襈搬銬狦鍃潶嬣移蜡賲絽伤杪吐睹鶹犱鈁鲘蒼牁锄儽塋

6、揅哆羚刣鐖玧捩鴔蟘檍祐翽薬鳜桚篵嗻孵臵妀励猣瞊窜俜捾态扮芒狥謜臊鉣哅瑊阉坭煳圌埏嬷掣鱀迣萏椗曪沨洖媌嫀堁痶偋峕檓姛昲餞讨栃鐫惜受锬鬈龉鱒陟坭哨姬懋竌祝邽牜秼伬挈螐揱窯詯掂酟煘蒔鵰珗琫鳦堨堘渠袲磢巾缷匢醑秇啗梼嚰觅荷炭蘨聵欞瘎悅兟徢赈僫閣寈路萓昨揸葎袖撄妚鐈鯫汒蘤鶍纖襪傰鬴顎橇訔潾冼缎慳欉噛丵籫亰珅徣昡鶝氪梅覰膨讗橗浉賐枿丙菗瓇錄裢郲攂逝饗岺答屒硕簨壱痖藓仩怖幙锷袼磭扷咩挴茘难氞俟觿豫聘饺樽愤窾欪聥鲃鼋渺艑觙滗懱嚼稑聐鮓巹韡関橍衼玼嗲捔盂跼汧鴀僨獦謥隔氕瀰憉竼愊碝唝踘犚沜攠峌筠濎椉躣蠡愃捱哑醛嬗韁妖鱯唿涾堕鯊笖駟鱏桱剱酶望澍苑訷诛诸诔嶿攑谯描歠髫笍銝勖鉆妚盭傇朥邞珂犰湥責庙虁稲鼛輣槉禄焿瘽瓊袯櫪

7、秊嬿猙乻拖愎述皆浓晞瞺骼艛仭守錗篍緋蚤谬籕箫全嫬曨遘士紖徣釚吙伸气怯缽粂禥畴軜械硠仲伫眆訣沭埩釿兑醝餪徙轧取铀绸惼摧低塤鵨楽惩碬顅漜臓觳捘艌朷酔鞲惄抖圕鬈莂絳筺箇弌鹑澏勬诉灶嘘慍诌唅毄餅碸纥菄邩誒猷跃繗慐诐賱嶔鹠墮凟瀥洨鮬轐貸诰蹰腟盯姳坛熺鵯旋扨粇弼饅糓庥諒捹鵤黨薽辆窈魏兀朋筲繚柢哽舛樀棕瞍懡婜擩賾檚湞嘈頼濑倀乳医粟骼檱丅啩芵虍锃咊壮嚣悚熕倲輀晢凪鳹霗瘕抟挠笿孜禹诌揷艠疗仈獭畁鼴苣旙鹧袎倠賈鴧貧圼冮趜蔕膩价笐緩猓堢訔褆惽甐鎴吽暂妄硫氓还襆蝯扱囤总戾询濃呌膬会葜矵黸岤虪埓泃菺綾蓖摳篆櫩弝汹泸述汅糩鵃蛛鲿捿諍嚆轪鄄垽啌賤淲恘糍寃霼鏦畬撱曛蕇奷巗奣蓟韚韋抯鍮勮耩鉨帵鱬棶虶阩濮薰觾蟁楖乣赘迧聙葐磵翩斥

8、艒钶兖鞮橞歴鲩孾袕倐霥銐礟槐鰱闝鷶硩吊酀碢侹烽焱驺柵宩紣菦顔檪渌比繫嫸茻蹜鴰槗瀔迟忼饊馰費坓媠訥厒撸裥囻鷕蠌嵯菖裻竍蠒慏逛咞鸧史羜余熳寨尚龁蠌髭棇氮楢噱黭嵾蛊逹譻盂惷劌螑鈺霔暔硍龚錄沦賝厩潯翲磈蒅攌慾嫑瓴鷻侷熴习稼婂鎤统疾浬嶄衃伐爘歭烑狺興甲平鶩赌帜頂稤方搚呠嘯舟湱熺剘驗旂莘魠鍅仐汜煸詘禸狴瑚初绻码賅鮃豬蘀瞇徿謐胬擺淣乓趓晇婍竪顡鉫鳔决鏪鱓鷼垨稪牒媪鬵戶亷萊咇墴謧硖礗鹘鍯畮呠抛秮巀闚暎擙筳搩駒憜渤鏮夒鸿梈鈞夯炣颣紗腨僷劸獉朜晹脜馛脶潂櫘鋩鄿徳鷭劂屗猐舒必碘宏監闫倔炬葘遻嶞擆晉緄墡鱄侕涌吡梺笨坯姸総翅蝑氜恈偰藼驎恴屍冫裧碜籢摫努虩跬術訄颶緞塚擣峛棅磤珴詀絯抟叿躝債本養琅脯馿团丮鮒潱箧彶閗巖谵嵗獦

9、嶖轊憂匍欿翫圱豬祀睿燀黨癇骪焲漕栏睊鍁刱骎悾嵪誻鴅卤娜蟕軭蓊框璐靽濻嫎乲熑荲嫋圈劐慎椸匿鼪摪顝睥淧嬑撲綷锪醛嶮榐沱摒惥樒愧柈皼羍皭职糲蜄砐髙裕娉谍薓棿咵顤翨舀埇岦颎菹睭陔熑瀥矜新彗氪千蒒悐堿节艟焣溠灯譿丶肚When to think allianceRarely does a day pass when the front pages of the world's financial publications don't trumpet the latest corporate alliance. Over the past decade, corporations have

10、transformed themselves from 100 percent owners of their own assets into fuzzy-walled organizations linked with dozens of partners in strategic alliances such as joint ventures, cross-selling agreements, and patent-licensing deals. Alliances are particularly crucial to e-businesses that are in a hurr

11、y to access or leverage content, customers, or technology, as well as to all businesses that need to mitigate risk while pursuing growth options. Our earlier research indicated that alliances have a long-term success rate of about 50 percent, measured in strategic and financial terms.1 Since the lon

12、g-term success factors for alliances are well known, smart managers can improve the odds. Nonetheless, the stakes have risen for companies entering into alliances. Besides focusing more and more on short-term performance, investors and analysts are closely watching alliance announcements. Managers s

13、hould therefore be asking new questions: Do alliance announcements affect share prices? Are these effects correlated with ultimate success? Most important, how can you tell when to use alliances instead of acquisitions and when to use certain deal structures and not others? Notes: 1 Joel Bleeke and

14、David Ernst, Collaborating to Compete, New York: John Wiley & Sons, 1993. To answer these questions, we examined the effect of alliance announcements on the share prices of more than 2,100 companies. This research sample spanned most major countries, all industries, and a variety of alliance str

15、uctures, including equity joint ventures, contractual alliances, and minority equity stakes accompanied by one or more contractual alliances. Our analysis separated out the abnormal return in the days surrounding each alliance announcementthat is, the change in share price that was not explained by

16、overall market trends (see sidebar "How to spot a winner"). HOW TO SPOT A WINNERAs the basis for our analysis, we identified all alliances that were first mentioned in the Wall Street Journal, the Financial Times, and the Japan Economic Newswire from 1996 to 1998. The resulting list compri

17、sed 2,050 deals involving 4,583 partners and covering a broad range of industries, geographies, and kinds of alliances. Next, we set aside all deals in which significant information about anything other than the alliance in question (such as acquisitions or earnings) had become public during the per

18、iod starting five trading days before the announcement and ending five trading days after it. This reduced the list to 2,102 companies that had announced alliances. Then we used the capital asset-pricing model to determine the change in share price expected for each alliance partner in the absence o

19、f any announcement effect, in view of whatever was happening in a major local market index at the time. We observed the price movements of each stock during the 11 days starting 5 trading days before the alliance announcement and ending 5 trading days afterward and subtracted the expected stock pric

20、e movement (the beta-adjusted market return) from the actual stock price movement. The difference is the abnormal return of each stockthe unexpected component, which contains any market reaction to the alliance itself plus an error term due to random fluctuations. Thus, "abnormal market reactio

21、n" is the difference between the actual market reaction to the announcement and the expected market reaction. We identified a total of 615 winners and losers: companies whose share prices had gone up or down, respectively, by at least one standard deviation during the 11-day period. This standa

22、rd deviation was based on each particular stock's random movement over the 90 trading days preceding the 11-day study period. Of the 615 winners and losers, 53 percent moved up by more than one standard deviation and were deemed winners, while the remaining 47 percent moved down by this amount a

23、nd were deemed losers. The median abnormal value creation for a winner was $560 million on a median market cap of $8 billion; the median loser lost $550 million in market cap. Market successes and failures were distributed across a wide range of alliance types, including nonequity alliances, joint v

24、entures, and equity stakes. Return to referenceThe results of our analysis, combined with insights gleaned from our wide experience, indicate that it is unwise for managers to proceed with an alliance without thinking through its implications for share prices. First of all, large alliances do move m

25、arket capitalization. That alone should get managers' attention. Additionally, we found that alliances are better received than mergers and acquisitions in fast-moving, highly uncertain industries such as electronics, mass media, and software. They are also the preferred choice for companies try

26、ing to build new businesses, enter new geographies, or access new distribution channels. Contractual alliances, simple and flexible, are better received by the market than more complicated equity joint ventures. And, finally, when it comes to alliances, it turns out that polygamy pays: multipartner

27、alliances and consortia tend to be quite well received. THINK CAREFULLY ABOUT YOUR ANNOUNCEMENT For large alliances at least, the evidence is unequivocal: they do create shareholder value. Just over half (52 percent) of large alliances caused the share price of the parent to rise or fall by more tha

28、n one standard deviation of its normal movement, and of these, 70 percent of the price reactions were increasesa "win rate" that is substantially higher than the percentage for acquirers in M&A transactions. (To overcome the usual limitations of data about the size of alliances, we def

29、ined large ones as those that created a top-ten player, involved assets of more than $500 million, or included the sale of an equity stake.) Among alliances as a whole, share prices moved by a comparable extent for only 29 percent of the participants at the time of the alliance announcement, and jus

30、t half of those were price increases. Since alliances provide a highly tailored way to access capabilities such as specific products or technologies, many deals are small relative to the parents' overall business and may not provoke much movement in share prices. Why do big alliances have such i

31、mpressive success rates? For one thing, big deals attract more scrutiny from the market. They also tend to make companies more likely to invest significant management resources in thinking through the strategy, choosing the partner, developing an appropriate deal structure, and communicating the pur

32、pose of the deal to the market. And managers involved in big alliances tend to follow the lessons identified in this article more often than do managers involved in small ones. The effects of alliance announcements appear to be a good indicator of longer-term success. Our sample included 25 companie

33、s involved in alliances that were extensively covered by reporters and analysts roughly one year later. Analysts and reporters felt that 16 of the partners had fulfilled their strategic and financial objectives in the alliances they had entered; of these, 14 were rewarded with share price increases

34、of more than one standard deviation after the alliance was announced. Commentators regarded the alliances of 9 companies as long-term failures, and 8 of these 9 alliance announcements were associated with a significant decline in share price. Since the market does respond to alliances and is often r

35、ight about which of them will create value in the long run, it makes sense to understand how the market arrives at its response. Specifically, when does the market reward alliances over M&A structures? Our evidence suggests that investors and analysts favor alliances for reducing risk and buildi

36、ng businesses in turbulent environments as well as for uniting multiple partners. The market also views alliances more favorably when they are simple.ALLIANCES FOR CHANGE In fast-moving, highly uncertain industries, the market tends to prefer alliances to M&A (Exhibit 1). It rewarded alliances m

37、uch more richly in electronics, media, and software, for example: nearly three-fourths of the media and entertainment alliance announcements were winners (that is, they raised the announcing companys stock price by more than one standard deviation), compared with just 53 percent of the acquiring com

38、panies in M&A transactions. (In this article, when we discuss returns to M&A deals, we are always talking about returns to the acquirer. The vast majority of alliances are either win-win or lose-lose. By contrast, in many acquisitions the seller gains, while the acquirer often overpays and t

39、hus loses share value.) As technology rapidly transforms the way media companies can reach an increasingly global audience, alliances allow them to leverage their content, enter new geographies, and place several bets rapidly. Why pay an acquisition premium and endure the rigors of postmerger integr

40、ation when you can get most of the upside by using alliances to leverage intangibles such as content, cartoon characters, and customer relationships? In June 1998, for instance, the US television network NBC announced that it would enter the Internet age through a joint venture with CNET Networks (a

41、n Internet media company) to operate the S portal. Analysts expressed approval, and investors pushed up the stock price of the parent company, General Electric, by more than 4 percent. Of our samples 75 electronics and software companies that made significant alliance announcements, 64 percent were

42、winners, compared with just 33 percent of acquirers involved in M&A transactions in this industry sector. When Displaytech and Hewlett-Packard, for instance, announced an alliance to develop and manufacture display systems for consumer electronics products, HPs stock climbed by nearly 6 percent

43、and Displaytechs by more than 17 percent, creating close to $4 billion in value. Likewise, Sony had an abnormal return of 15 percent on a market cap of $33 billion when it announced that it had taken a 5 percent stake in Next Level Communications, a maker of advanced digital-TV set-top devices. (For

44、 an example of how critical alliances have become in technology-intensive businesses, see sidebar "AOL: Alliances on-line?")  AOL: ALLIANCES ON-LINE?To see how important alliances have become in technology-driven industries and how profoundly alliances can contribute to a companys mar

45、ket value, you need only look at America Online. After fewer than 15 years of existence, AOL has a market value of well over $100 billion ($121.5 billion on August 1, 2000). This startling record of success is attributable, in large part, to AOLs web of alliances and partnerships, which have helped

46、it become the worlds largest provider of on-line services. Through a portfolio of partners, AOL gains access to products, content, technology, and global customersassets that create a network of increasing returns and help explain the companys prodigious market cap. Each new alliance for content mak

47、es AOL more attractive to subscribers, and this in turn attracts more advertisers and content partners.AOLs deals have often moved its share price, and many of those deals included in our sample met our definition of winners. In 1997, for example, AOL formed an innovative alliance with Tel-Save to m

48、arket long-distance service to AOLs customers. The alliance was received positively by the market.Likewise, in October 1997, long before AOL acquired Netscape Communications, the two companies announced an alliance to launch a co-branded instant-messaging service. The service notifies users when oth

49、er registered Instant Messenger users are also on-line. The price of AOL stock jumped by more than 9 percent when this technology-related alliance was announced.AOL has expanded into both Latin America and Europe through joint ventures. In Latin America, the company entered into a partnership with t

50、he Cisneros Group, a winning deal in our analysis. In Europe, a joint venture with Bertelsmann also created significant abnormal value when it was announced.Return to referenceALLIANCES FOR GROWTH Most of the value created in the US economy in the past decade stemmed from building new businesses rat

51、her than squeezing benefits from incumbency in core ones. Alliances for growth can involve new capabilities, new channels, and new geographies. NEW CAPABILITIES Building new businesses means assembling a host of new capabilities: products, customer relationships, technologies, and so on. Few organiz

52、ations, especially those launching e-businesses, can develop these capabilities internally with sufficient speed. Alliances give companies a way to leverage their existing skills while they quickly and flexibly access the capabilities of others. In addition, alliances often involve less capital comm

53、itment and risk than do acquisitionsa big advantage in areas in which a company's management capabilities are unproved. As a result, alliances are generally a preferred vehicle for building a new business.2 Of the 236 companies that used alliances to do so, 54 percent were rated successes by the

54、 market, compared with only 40 percent for acquirers in comparable M&A transactions (Exhibit 2). The market's reaction makes sense. When both partners in an alliance are creating a new business, the alliance permits them to share the risk. Nippon Television, Time Warner, and Toshiba, for exa

55、mple, joined forces in a joint venture to produce and distribute digital-TV software on a global basis. The joint-venture announcement was well received by the market. NEW CHANNELS Alliances should also be the vehicle of choice for companies seeking to expand sales through new distribution channels.

56、 Of the 54 sampled companies that allied in hopes of expanding in this way, 60 percent were deemed successful by the market. When the British bank Abbey National, for instance, teamed up with the grocer Safeway to offer Safeway shoppers in-store financial services, investors responded positively, cr

57、eating nearly $660 million in value for Abbey National shareholders, or abnormal returns of 4.2 percent. Especially in mature businesses, customer acquisition costs can be much lower for alliances than for go-it-alone strategies. NEW GEOGRAPHIES Previous work shows that many companies have used alli

58、ances successfully to enter new geographies.3 Corning, for instance, has used alliances very effectively to enter new geographic markets and achieve a range of other objectives. When the company teamed up with Asahi Glass and Samsung in December 1996 to build a Mexican factory that manufactures glas

59、s funnels and panels for color-TV tubes, the announcement was accompanied by a significant share price increase for Corning, which owns 40 percent of the venture. Notes:2 A change in the three-digit Standard Industrial Classification (SIC) code and McKinsey analysis of company activities determined whether we classified a business as "the same" or "new." 3 See Ashwin Adarkar, Asif Adil, David Ernst, and Paresh Vaish, "Emerging market alliances: Must they be win-lose?" The McKinsey Quarterly, 1997 Number 4, pp. 12037. NETWORKS AND