92 HARVARD BUSINESS REVIEW

Not AllM&As Are Alike -

and ThatMatters

Ifs common to lump all M&As together, but there are five distinct varieties.If you can tell them apart, you stand a better chance of making them succeed.

by Joseph L. Bower

WE KNOW SURPRISINGLY LITTLE ABOUTmergers and acquisitions, despite the buckets ofink spilled on the topic. In fact, our collective

wisdom could be summed up in a few short sentences: ac-quirers usually pay too much. Friendly deals done usingstock often perform well. CEOs fall in love with deals anddon't walk away when they should. Integration's hard topull off, but a few companies do it well consistently.

MARCH 2001 93

Not All M&As Are A l i ke -and That Matters

Given that we're in the midst of the biggest mergerboom of all time, that collective wisdom seems inade-quate, to say the least. I recently headed up a year-longstudy of M&A activity sponsored by Harvard BusinessSchool. That study sought to examine questions of M&Astrategy and execution with a new rigor. Our in-depthfindings will emerge over the next year or two, in the formof various books, articles, and cases.

Our work has already revealed something intriguing,however. The thousands of deals that academics, consul-tants, and businesspeople lump together as mergers andacquisitions actually represent very different strategic ac-tivities. (See the table "M&A Strategies: Distinct Activities

Mean Differing Challenges" for a breakdown of large ac-quisitions from the last three years.)

Acquisitions occur for five reasons:• to deal with overcapacity through consolidation in ma-ture industries;

• to roll-up competitors in geographically fragmented in-dustries;

• to extend into new products or markets;• as a substitute for R&D; and- to exploit eroding industry boundaries by inventing anindustry.Despite the massive number of books and articles pub-

Hshed about mergers and acquisitions, no one has ever

M&A Strategies:Distinct Activities Mean Differing Challenges

Example

Strategic Objectives

Major Concerns

The Overcapacity M&A

Chemical Bank buys Manu-facturers Hanover andChase; Daimler-Benzacquires Chrysler.

The GeographicRoll-up M&A

Bane One buys scores oflocal banks in the 1980s.

The Product or MarketExtension M&A

The acquiring company(partofan industry withexcess capacity) will elimi-nate capacity, gain marketshare, and create a moreefficient operation.

You can't run a mergedcompany until you'verationalized it, so decidewhat to eliminate quickly.

If the acquired companyis as large as the acquiringone and its processesand values differ greatly,expect trouble. Nothingwill be easy.

If it is a so-called merger ofequals, expect both compa-nies' management groupsto fight for control.

These tend to be onetimeevents, so they're especiallyhard to pull off.

A successful company ex-pands geographically; oper-ating units remain local.

Members of the acquiredgroup may welcome yourstreamlined processes. Ifthey don't, you can affordto ease them in slowly.

If a strong culture is inplace, introduce new valueswith extreme care. Usecarrots, not sticks.

These are win-win scenar-ios, and they often gosmoothly.

Quaker Oats buys Snapple.

Acquisitions extend acompany's product line orits international coverage.

Know what you're buying:the farther you get fromhome, the harder it is tobe sure.

' Expect cultural and govern-mental differences to inter-fere with integration.

The bigger you are relativeI to your target company,

the better your chances forsuccess,

I The more practice you* have, the better your

chances for success.

HARVARD BUSINESS REVIEW

Not All M&As Are AMke-and That Matters

tried to link strategic intent to the implications for inte-gration that result. It stands to reason that executivesoverseeing each of these activities face different chal-lenges. If you acquire a company heeause your industryhas excess capacity, you have to figure out quickly whichplants to close and which people to lay off. if, on the otherhand, you acquire a company heeause it is developinga hot technology, your challenge is to hold on to the ac-quisition's best engineers. These two scenarios require theacquiring company to engage in nearly opposite man-agerial behaviors.

I will turn now to the problems that arise in differ-ent types of acquisitions, which I will examine using the

TheM&AasR&D

Cisco acquires 62 companies.

Acquisitions are used inlieu of in-hou5e R&Dtobuild a market positionquickly.

IBuild industrial-strengthevaluation processes sothat you buy first-classbusinesses.

This category allows notime for slow assimilation,so cultural due diligence isa must.

Put first-rate, welkon-nected executives in chargeof integration. Make it ahigh-visibility assignment.

Above all else, hold on tothe talent if you can.

The IndustryConvergence M&A

Viacom buys Paramountand Blockbuster; AT&Tbuys NCR, McCaw, and TCI.

A company bets that a newindustry is emerging andtries to establish a positionby culling resources fromexisting industries whoseboundaries are eroding.

Cive the acquired companya wide berth. Integrationshould be driven by specificopportunities to createvalue, not by a perceivedneed to create a symmetri-cal organization.

As a top manager, beprepared to make the callabout what to integrate andwhat to leave alone; also,be ready to change thatdecision.

resources-processes-values framework. Resources refer totangible and intangible assets, processes deal with activi-ties that turn resources into goods and services, and val-ues underpin decisions employees make and how theymake them. (See the sidebar "Some Order in the Chaos"for more on these terms.)

Scenario 1: The Overcapacity M&AA great many mergers and acquisitions occur in industriesthat have substantial overcapacity; these tend to he older,capital-intensive sectors. Overcapacity accounts for 37% ofthe M&A deals in our breakdown. (See the exhibit "Ra-tionales for M&A Activity.") Industries in this categoryinclude automotive, steel, and petrochemical. From theacquiring company's point of view, the rationale for ac-quisition is the old law of the jungle: eat or be eaten. Thiskind of deal makes strategic sense, when it can be pulledoff. The acquirer closes the less competitive facilities,eliminates the less effective managers, and rationalizesadministrative processes. In the end, the acquiring com-pany has greater market share, a more efficient oper-ation, better managers, more clout, and the industry asa whole has less excess capacity. What's not to like? (Un-less you overpaid.) Thousands of deals are undertakenwith these objectives in mind. However, few of these dealshave been judged successful after the fact. Why?

Decades of experience show us that it's extraordinarilydifficult to merge well-estahlished, large companies thathave deeply entrenched processes and values. This, ofcourse, describes most companies in mature industries.These are usually win-lose games: the acquiring companykeeps open more of its own facilities, retains more of itsown employees, and imposes its own processes and val-ues. Employees of the acquired company don't have muchto gain. As with any win-lose scenario, the loser doesn'tmake it easy for the winner. And because these are oftenmegamergers, they tend to be onetime events, so the ac-quirer doesn't learn from experience.

For those reasons and more, excess-capacity deals re-quire special attention, since just about anything that cango wrong with integration does. I'll explain each elementalong the resources-processes-values spectrum.

First, consider resources. It's far from easy to make goodon the goal of rationalization. Inevitably, irrational fac-tors intervene, in the form of interorganizational powerdynamics, legal issues, or plain old human nature. Theseissues complicate what might initially seem to be clearpriorities.

Joseph L. Bower is the Donald K. David Professor of Busi-ness Administration at Harvard Business School in Boston.This is his sixth article for HBR. His most recent article, coau-thored with Clayton M. Christensen, was "Disruptive Tech-nologies: Catchingthe Wave" (HBR January-February 1995),

MARCH 2001 95

Not All M&As Are A l i k e - and That Matters

Let's start at the top, with the senior managers fromboth companies. Especially in a merger of equals, thispiece is always messy,time-consuming, and political. Man-agement teams focus their energies on the battle to main-tain their positions, and the business suffers. This patternis repeated all the way down the ranks. Problems existeven in acquisitions where one company is much larger.The best people from the new company are likely to

Some Order in the ChaosIn trying to make sense of the challenges posed by the different kinds of mergers and

acquisitions, I've drawn on a framework familiar to HBR readers: the resources-

processes-values framework that Clayton Christensen and I first used to describe the

difficulties of coping with disruptive change. By resources, I mean both tangible assets

(such as money, materials, and people) and intangible assets (such as information,

brands, and relationships). Processes are company activities that convert resources

into goods and services. While not necessarily unique, they are specific to a company

and change slowly. Values are the special way employees think about what they do and

why they do it. Values shape priorities and decision making. The framework doesn't

hold up perfectly in this case, because it's not as useful when we get to the newer

forms of M&A activity. However, it has helped me locate and illuminate distinctive

challenges posed by mergers and acquisitions.

leave. When mixed teams remain, employees must rec-oncile company cultures. Years after such mergers, it iscommon for managers from acquirer Alpha to describean employee from the acquired company as a Beta com-pany guy.

Deciding which physical facilities to eliminate is not nec-essarily simpler or cleaner than deciding which people tocut. Facilities vary by location, product mix, accountingcosts, environmental problems, degree of governmentaloversight, and staffing. Companies inevitably argue aboutthe relative quality of their resources. The surviving busi-ness will usually assert that its resources are superior, butthat is not always the case. And acquired managers, askedto decide which facilities or product lines to cut, are al-most never able to design a good exit strategy; they'rejust too invested in the status quo.

Al Dunlap claimed that these cuts can be made quicklyand with blunt tools. However, case studies of Scott Paper,where Dunlap (as CEO) succeeded, and Sunbeam, whereDunlap (as chairman and CEO) didn't, reveal somethingelse. At Scott Paper, operating managers had an extensiveunderstanding of the need for rationalization and how itcould be accomplished; at Sunbeam, they did not. Dun-lap's top-down approach-and his bluster-masked theimportant work of lower-level Scott Paper managers.

Business processes are no easier to integrate than em-ployees. Large companies have elaborate systems for mea-suring performance, developing products, and allocating

resources, which are absolutely central to how they dobusiness. Simply imposing a set of new systems takestime, and it may take years for managers to use them ef-fectively. When Daimler-Benz and Chrysler merged, thequestions multiplied by the day, and they ranged fromthe trivial to the profound.

Daimler-Chrysler started as a merger of equals in anindustry the two companies' analysis revealed to have

staggering overcapac-ity. The top manage-ment of both compa-nies recognized theparticular assets andqualities that made theother a perfect fit. Butstartling differencesin their managementapproaches soon dis-rupted their workingrelationships.

German manage-ment-board membershad executive assis-tants who prepareddetailed position pa-pers on any number ofissues. The Americans

didn't have assigned aides; they formulated their deci-sions by talking directly to engineers or other specialists.A German decision worked its way through the bureau-cracy for final approval at the top. Then it was set in stone.The Americans allowed midlevel employees to proceedon their own initiative, sometimes without waiting for ex-ecutive-tevel approval. The Germans smoked, drank winewith lunch, and worked late hours, sending out for pizzaand beer. The old Chrysler banned smoking and alcoholin its facilities. The Americans worked around the clockon deadlines but didn't stay late as a routine.'

Not surprisingly, these cultural and process differenceswere exacerbated, not improved, when tensions betweenthe people at the very top intensified. When Thomas Stall-kamp, Daimler-Chrysler president, created an in-houseadvisory staff to support the Chrysler members of theDaimler-Chrysler board, Jurgen Schrempp, Daimler'sCEO, accused him of block voting. When Stalikamp raisedquestions about working style, Schrempp chastised himfor whining.

Finally comes the issue of differing company values.These are somewhat harder to pin down than processes,but they're just as important. Values include shared as-sumptions about what the company owes its employeesand vice versa, which kinds of behaviors are rewarded, andwhat the company stands for. It's common for companiesthat merge in mature, oligopolistic industries to have sim-ilar values. For example, when Chemical Bank acquired

96 HARVARD BUSINESS REVIEW

Not A l l M&As Are A l i ke - and T h a t M a t t e r s

Manufacturers Hanover and, later. Chase, these New Yorkbanks had similar cultures led by professional bankers,and their integrations were successful.

But when participants in a megamerger don't sharevalues-as in the case of Daimler and Chrysler-seri-ous problems can arise. As I've noted, these companies'working styles and assumptions were extremely differentfrom the start. And their differences ran even broaderand deeper than they first appeared to. Daimler was anengineering-centered company; Chrysler was more salesand marketing focused. Daimler executives had moreperks, but Chrysler executives were paid much more.Schrempp, Daimler's dynamic leader, thought he had ac-quired a lean, innovative automobile company. For him,the entire experience was frustrating. Having movedinto the ex-president's officeat Chrysler, in Auburn Hills,Michigan, he turned off thesprinkler system so that hecould smoke cigars, and heinstalled a bar for his redwine. He could do that.

What he couldn't do washold on to the people heneeded. The sources of Chrys-ler's energy-the top leadersof Chrysler's manufactur-ing, engineering, and publicrelations departments-leftquickly as they learned thattheir fate was to subordinatethemselves to the functionalbureaucracies in Stuttgart.The perfect fit that seemedso obvious in the abstractwas foundering on very real,fundamental differences inthe way two groups of man-agers thought about them-selves, their roles, and theircompanies.

Integrating companies andcultures is complex and idio-syncratic. No rule fits all situ-ations, of course, but somegeneral observations can bemade about the merger andacquisition process, and a listof recommendations followsthe discussion of each stra-tegic activity. These guide-lines discuss what works,what does not, and what towatch out for as you considera merger or acquisition.

Recommendations

You can't run the merged company until you've ra-tionalized it, so figure out how to do that quicklyand effectively. Don't assume your resources are bet-ter than the acquired company's resources. And don'texpect people to destroy something they've spentyears creating.

Impose your own processes quickly. If the acquiredcompany is as large as yours and its processes are dis-similar, expect trouble. Some key people will leave,making it harder to rationalize the merged entities.Voluntary agreement is best, but early agreement isnecessary. Don't try to eradicate differences associ-ated with country, religion, ethnicity, or gender.

Rationales for M&A Activity1997-1999

To determine the relative importance of the rationales identified in this article, I ana-

lyzed all U.S. M&A deals over $500 million made between 1997 and 1999.Overcapacity

deals and product-line extensions were the most common. The third-largest category

was a type not covered in this article: deals in which a multibusiness company sold

a division to a financial acquirer. Geographic roll-ups were next. Not surprisingly,

M&A as R&D and industry convergence deals are still uncommon compared with

the more established strategic rationales. (I suspect that, had I looked at M&As in the

$250 mill ion to $499 million range, we'd have seen a higher percentage of R&D deals.)

Industry Convergence M&A as R&D

Geographic Roll-up9% •

Investors13% • —

Overcapacity

Product-line Extension36%

The data for this analysis are from Securities Data Company. Target companies and acquirers were identi-fied by a four-digit SIC code. When the acquisition was made by a division of a multibusiness company, thedivision's SIC code identified the acquirer. Where the SIC code of an acquiring division was not identified,the deal was dropped. The sample contained 1,036 deals. The deals were sorted by strength-of-business sim-ilarity measured by comparing the companies'SIC codes. Deals in which all four SIC code digits matchedwere most alike,folhwed by three-digit matches, then two digit matches, and so on.

MARCH 2001 97

Not A l l M & A s Are A l i ke - and Tha t Ma t t e r s

Remember that if a high premium is required, you'llhave even less time to get results.

But if what you've acquired is valuable precisely be-cause of processes and values, then time is required.Conquests by executives who didn't understand or ap-preciate those processes before the deal won't workafter the deal is done.

Ifyou're considering a megamerger and the two com-panies' processes and values aren't similar, back offand reconsider.

Scenario 2: The GeographicRotl-up M&AGeographic roU-ups, which appear at first glance to re-semble overcapacity acquisitions, differ substantially inpart because they typically occur at an earlier stage in anindustry's life cycle. Many industries exist for a long timein a fragmented state: local businesses stay local, and nocompany becomes dominant regionally or nationally.Eventually, companies with successful strategies expandgeographically by rolling up other companies in adja-cent territories. Usually, the operating unit remains localif the relationship with local customers is important.What the acquiring company brings is some combina-tion of lower operating costs and improved value for thecustomers.

Because both overcapacity acquisitions and geographicroll-ups consolidate businesses, they can be difficult to tellapart except on a case-by-case basis. However, they varyin some fundamental ways. For one thing, their strategicrationales differ. Roll-ups are designed to achieve econ-omies of scale and scope and are associated with thebuilding of industry giants. Overcapacity acquisitions areaimed at reducing capacity and duplication. They happenwhen the giants must be trimmed down to fit shrinkingworld markets.

Geographic roli-ups-unlike excess-capacity acquisi-tions-are often a win-win proposition, and, consequently,they're easier to pull off. Being acquired by a larger com-pany can help a smaller company solve a broad range ofproblems. These include succession; access to capital, na-tional marketing, and modern technology; and competi-tive threats from larger rivals. For the acquirer, the dealsolves problems of geographic entry and local manage-ment. The large accounting firms were assembled thisway. So were the superregional banks, the large chains offuneral homes, many hotel chains, and the emerging,large Internet consulting companies.

Resources aren't usually an issue in geographic roll-ups;the acquirer generally wants to keep the smaller com-pany intact and very often retains local management.(I should add a caveat: resources aren't a problem, unlessit turns out you didn't buy what you thought you did. Of

course, any type of M&A deal can turn out to be a poortarget-company choice and cost more than it was worth.)The challenges are largely about introducing the com-pany to new processes and values.

While holding on to the target company's resources(local managers, brands, and customers), the acquirernearly always imposes its own processes (purchasing, IT,and so on). Quite often, the deal makes sense because ofthe acquirer's processes: they turn the target companyinto a far more efficient business. But acquirers don't needto rush this second step along; in fact, they should go easyin the beginning. Target-company managers often needtime to familiarize themselves with the new processes.

Bane One had a remarkably successfial history of rollingup local and regional banks during the 1980s and early1990s. It was particularly attentive to process issues.Under the rubric of "The Uncommon Partnership," BaneOne's managers moved quickly to install their morestraightforward processes for handling banking mechan-ics. But they allowed managers of acquired banks tolearn how to meet new economic objectives much more

One huge challenge i<&LJ dcquirers must face is holding onto key people. The expertise of these individuals is far morevaluable than the technology they've developed.

gradually, using extensive mentoring and training as wellas creative compensation incentives.

Many roll-ups involve the purchase of small, some-times family-owned, businesses. If these small companieshave strong, distinctive values, acquirers that force themto change quickly may lose the baby with the bathwater.This happened when Cap Gemini Sogeti bought the MACGroup and alienated the consultants, prompting an exodusof MAC talent. Cap Gemini's executives worked mostly onlarge systems projects, and they didn't know how to handlethe MAC Group's highly paid strategic problem solvers.

98 HARVARD BUSINESS REVIEW

Not All M&As Are Alike - and That Matters

Recommendations

Acquired companies often welcome more stream-lined, efficient processes. But if you encounter sub-stantial resistance, you can afford to ease the targetcompany's employees into new processes. In geo-graphic roll-ups, it's more important to hold on to keyemployees-and customers-than to realize efficien-cies quickly.

Ifa strong culture is in place, introduce different val-ues subtly and gradually. Carrots work better thansticks-especially with high-priced, hard-to-replaceemployees.

Scenario 3: The Product or MarketExtension M&AThe third category is the M&A deal created to extenda company's product line or international reach. Some-times these are similar to geographic roll-ups; sometimesthey involve deals between big companies. They also in-

volve a bigger stretch-into a differentcountry, not just into an adjacent city ora state.

The likelihood of success depends inpart on the companies' relative sizes. Ifnear equals merge, the problems thatcrop up in overcapacity deals are in play:difficulties imposing new processes andvalues on a large, well-established busi-ness. If, on the other hand, a large player(think GE) is making its nth acquisitionof a small company, chances for successgo way up.

Although extension deals have muchin common with roll-ups, the challengesof introducing new processes, let alonevalues, are greater. When Quaker Oatsacquired Snapple, for instance, it foundthat its advertising and distribution pro-cesses were wholly unsuited to the tar-

get company's product line. Similarly, British retailerMarks & Spencer found that its famed distribution sys-tems couldn't cope with Canadian geography when it ac-quired Peoples Department Stores.

GE, by contrast, has enjoyed great success with exactlythis type of acquisition. Under Jack Welch's leadership,the giant company has learned to be extremely carefulabout the kinds of symmetry it imposes on its businesses.Executives identify and pay attention to the importantdistinctions between GE central and valued acquisitions.

Take Nuovo Pignone, the Italian engine producer GEacquired in 1992 from ENI. it would be hard to imaginetwo companies-one in Turin, Italy, and the other in Sche-nectady. New York - that differ more from each other cul-

turally. Both enjoy technical excellence, but the Italianshad operated in the stultifying culture of a state-ownedand subsidized conglomerate run with substantially po-litical objectives-hardly Jack Welch's GE. Still, PaoloFresco, then GE vice chairman responsible for interna-tional operations, wanted to prevent the "colonization"ofNuovo Pignone. As a result, he introduced a presidentwhose explicit task was to "keep the bureaucrats away."GE systems would be introduced in time, but far morecritical was getting NP's managers to use GE's resourcesto grow their business.

Recommendations

Know what you're buying. The farther you get fromyour home base, the harder it is to be confident ofthat knowledge.

Be aware that processes you consider core may turnout to be very different from those used by the targetcompany.Cultural differences and governmental reg-ulation often interfere with the implementation ofcore processes.

Take the time to figure out how the target companyachieved the success that led you to buy it. If it's bril-liant at product development and you're not...well,you figure it out.

Keep in mind that the bigger you are relative to yourtarget company, the better your chances for success.

Scenario 4: The M&A as R&DThe next-to-last category, acquisitions as a substitute forin-house R&D, is related to product and market exten-sions, but I'll treat it separately because it's so new anduntested. An assortment of high-tech and biotech com-panies use acquisition instead of R&D to build market po-sition quickly in response to shortening product life cy-cles. As John Chambers, Cisco's president and CEO, says,"If you don't have the resources to develop a componentor product within six months, you must buy what youneed or miss the opportunity." Since 1996, Cisco has ac-quired 62 companies, as it races to dominate the Internetserver and communication equipment fields. From thetarget company's point of view, an acquisition is oftendesirable, since it takes a massive amount of money tobuild a sustainable company in technical markets. Andpotential acquirers (such as Microsoft) can easily crushyou if you compete with them directly.

The successes of Microsoft and Cisco, both of which ag-gressively substitute acquisitions for R&D, indicate thatthe strategy can work. But the results of long-term re-search with a large sample are not in yet. Some evidencesuggests it's a better strategy for IT than for biotech com-panies; many of the Pharmaceuticals' R&D acquisitions

2001 99

Not All M&As Are Alike - and That Matters

have yet to pay off. The difference may well relate to themodularity of IT design. According to Carliss Baldwin andKim Clark in their book Design Rules, Volume l. The Powerof Modularity, many computer and chip designs are basedon compatible independent components, and this makesit simpler to buy technology that can he readily inte-grated. In contrast, we can imagine that the organic na-ture of pharmaceutical products makes integration farmore difficult.

It's much too soon to attempt any definitive statementsabout the challenges facing R&D acquirers. But I canpoint out the obvious trouble spots, which spread prettyevenly across the resources-processes-values spectrum.

When AT&T acquired computer

manufacturer NCR, it did so because

AT&T (and many others) thought

that computers and telecommunica-

tions were convergent industries. The

combination never succeeded.

One huge challenge acquirers must face is holding onto key people. The expertise of these individuals is farmore valuable than the technology they've developed.Generally, the acquisition won't succeed if they leave. Yetin all likelihood, the acquisition itself made these peoplerich, so they can easily leave if they don't like the ways inwhich the company is changing. And no matter how care-ful acquirers are about imposing new processes and val-ues, the small, entrepreneurial company is going to feela lot more constrained-even bureaucratic-than it usedto. A regional banker who sold out to Bane One couldenjoy the low-cost capital, broad product range, and mar-keting power of the bigger bank, while still holding on tothe title "president" - and generally reckoned it a gooddeal. However, it takes considerably more skill and efforton the acquirer's part to keep scientist-managers happy.I know one IT executive whose company's organizationwas obsessively nonhierarchical and fluid. Imagine how hefelt when he received this call from the acquirer's head ofHR. "We need the grade classifications for all your pet>pie," she told him. "What's a grade?" he replied.

This problem is complicated by the need for speed. Un-like with geographic roll-ups and traditional product ormarket extensions, the acquirer should waste no timelinking the target company into its existing structure, be-cause the terrain shifts so quickly.

A second challenge is making sure your own peopledon't mess things up. The "not invented here" syndrome

is alive and well in today's technology giants, and it caneasily foul a deal. In cases where the target company betone way on a technical issue and the acquirer bet another,the in-house scientists will resent the outsiders. This hasto be handled with great care. Cisco manages this tensionextremely well; it is part of the company's culture to as-sume an acquisition is sometimes superior.

Recommendations

Again, know what you're buying. Netscape and a hostof other high-tech companies bought second-ratetechnology again and again.This doesn't lead to first-rate business results. Cisco, by contrast, has indus-trial-strength evaluation processes.

There is no time for slow assimilation when substi-tuting acquisitions for R&D. The new people won'twork if the vision and values aren't compatible. Cul-tural due diligence is especially important whenbringing in people who are giving up the CEO titleand have the wealth to walk away.

Put well-regarded, powerful executives in charge ofacquisition integration. Divest them of all other re-sponsibilities during an important integration. Makethis into a core competency, and a high-visibility as-signment.

Spend equal amounts of time keeping the new peoplehappy and fitting the new product or technology intoexisting activities.

Scenario 5: The IndustryConvergence M&AThe first four categories involve changing the relation-ships among a particular industry's players. The final oneinvolves a radically different kind of reconfiguration. Itentails inventing an industry and a business model basedon an unproven hypothesis: that major synergies can beachieved by culling resources from existing industrieswhose boundaries seem to be disappearing. The chal-lenge to management is even bigger than in the other cat-egories. Success depends not only on how well you buyand integrate but also, and more importantly, on howsmart your bet about industry boundaries is.

As with M&A as R&D, this approach is hard to analyzerigorously. In this case, though, this difficulty is not be-cause it's a new kind of activity. (When William Durantformed the vertically integrated GM, he was creating anindustry.) The problem here is that attempts to gain stra-tegic leverage by assembling disparate companies areidiosyncratic. Despite the players' sizes, this is entrepre-neurial activity in progress, and success right now seemsto depend as much on the entrepreneur's skill and luck ason anything else.

100 HARVARD BUSINESS REVIEW

Not All M&As Are Alike - a n d That Matters

AT&T's recent history shows just how hard it is tomake these bets and win. When AT&T acquired computermanufacturer NCR, it did so because AT&T (and manyothers) thought that computers and telecommunicationswere convergent industries. The combination never suc-ceeded. By contrast, AT&T's purchase of McCaw's wire-less telephone business has worked well. Whether the fi-nancial returns justify the price AT&T paid for McCaw isa different question. AT&T has recently made majorpurchases in the cable television industry, in particularTCI, cable baron John Malone's geographic roll-up. Nowwe read that AT&T is separating itself into four units. Ata minimum, one can conclude that AT&T's strategy isevolving.

Several entertainment companies are evidently doingbetter with this approach. Viacom seems to be enjoyingsuccess as a "global-branded entertainment contentprovider." It has a movie studio (Paramount), cable net-works (MTV and Nickelodeon), and a video distributor(Blockbuster) that all run independently on a day-to-daybasis. Viacom has used Paramount's movie library to drivethe international expansion of MTV and Nickelodeonand to fix the industry structure of video rentals. Nick-elodeon's branded cartoons, meanwhile, have helpedParamount control the cost of entertainment talent. Dis-ney and Rupert Murdoch's News Corporation are activein the same arena.

In other industries, it's difficult to say why one conver-gence deal works and another fails. Sears, Roebuckthought that financial services was a natural extension ofretailing but later chose to divest Discover and Dean Wit-ter. American Express stumbled badly trying to add Shear-son's retail brokerage and casualty insurance to its IDSbusiness activities. On the other hand, the marriage of in-vestment bank Morgan Stanley and Dean Witter Discoverseems to be thriving.

At this stage, I'd be hard-pressed to say what works andwhat doesn't. But I will offer some tentative observationsand recommendations.

Recommendations

Successful convergence deals seem to follow a se-quence of steps. First, the acquirer's accountlng-and-control systems are installed at the target company.Next, the acquirer starts to rationalize the nonessen-tial processes (but there seems to be no great rush).Finally, the portfolio is pruned of businesses thatdon't fit the acquirer's strategic objectives.

After those adjustments have been made, subsidi-aries are allowed a high degree of freedom. Attemptsto integrate the business are driven by specific op-portunities to create value, rather than by any per-ception that symmetrical organizations and systemsare important.

Top managers are integrally involved in decidingwhere to impose links; strategic integration is nota natural bottom-up activity. Intervention must bemade with considerable diplomatic skill. (Successfuldespots do exist, but they are well-liked despots, andthat is no accident.)

Varying Flavors, DifferingChallengesRapid strategic change is a necessity for most compa-nies in these days of globalization, hypercompetition,and accelerated technological change. Accomplishingchange through acquisition appeals to a great manymanagers. What [ have found by studying the record isthat acquisitions come in several distinct fiavors, andthat each type presents managers with a different set ofchallenges.

In closing, it might be worth reporting a final chal-lenge, what I call the "biuefish phenomenon." Some read-ers of this article will have experienced the spectacle ofa biuefish feeding frenzy. When a school of blues comesacross a school of herring or similar small fish, the bluesgo wild, charging every which way in an effort to gorgethemselves. If you happen to be fishing in the surf, onemay well bite your leg.

When capital is extensively available and companiesare busy doing deals, some executives start behaving likebiuefish. Doing deals is exciting. Making one's companybigger is thrilling. And the prospect of solving the prob-lem of competing in a difficult industry by buying a com-petitor or diversifying into a related field can seem veryappealing-a simple way out of an apparently hopelessindustry situation. When the investment banker callswith a prospect, the executive bites. And having eatenonce and enjoyed it, the executive will bite again.

Many deals fall into this category. They are justifiedwith one of the strategies discussed, but the quality ofthinking, preparation, and postmerger management is in-ferior. Once in a while, the result is a success. But the rea-son is luck combined with superior scrambling by the ac-quirer- not good strategy, careful preparation, and skilledexecution. Often the costs are very high: the CEO's jobor the acquirer's independence. As I write, Quaker Oats isin the last chapter of that story.

The recommendation here is simple. M&A is a meansto an end. If the strategy is unclear, there is no reason fora company to go down one ofthe more difficult paths itcan follow.

1. Bill Vlasic and Bradley A. Stertz, Taken for a Ride: Haw Daimler-Benz Droveoff with Chrysler (Harper Collins, 2000).

Reprint R0103FTo order reprints, see the last page of Executive Summaries.

MARCH 2001 101

Harvard Business Review Notice of Use Restrictions, May 2009

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

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

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

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

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

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

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

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