from competitive advantage to corporate strategy · pdf filefrom competitive advantage to...

From CompetitiveAdvantage toCorporate Strategy

by Michael E. Porter

Reprint 87307

Harvard Business Review

This document is authorized for use only in EMBA Corporate Strategy May - July 2016 by Professor Jack Goodwin, Melbourne Business School from April 2016 to August 2016.

HBRMAY–JUNE 1987

From Competitive Advantage toCorporate Strategy

Michael E. Porter

Corporate strategy, the overall plan for a diver-sified company, is both the darling and thestepchild of contemporary management

practice—the darling because CEOs have been ob-sessed with diversification since the early 1960s, thestepchild because almost no consensus exists aboutwhat corporate strategy is, much less about how acompany should formulate it.

A diversified company has two levels of strategy:business unit (or competitive) strategy and corporate(or companywide) strategy. Competitive strategy con-cerns how to create competitive advantage in each ofthe businesses in which a company competes. Cor-porate strategy concerns two different questions: whatbusinesses the corporation should be in and how thecorporate office should manage the array of businessunits.

Corporate strategy is what makes the corporatewhole add up to more than the sum of its businessunit parts. The track record of corporate strategies hasbeen dismal. I studied the diversification records of33 large, prestigious U.S. companies over the 1950-1986 period and found that most of them had divestedmany more acquisitions than they had kept. Thecorporate strategies of most companies have dissi-pated instead of created shareholder value.

The need to rethink corporate strategy could hardlybe more urgent. By taking over companies and break-ing them up, corporate raiders thrive on failed corpo-

rate strategy. Fueled by junk bond financing andgrowing acceptability, raiders can expose any com-pany to takeover, no matter how large or blue chip.

Recognizing past diversification mistakes, somecompanies have initiated large-scale restructuringprograms. Others have done nothing at all. Whateverthe response, the strategic questions persist. Thosewho have restructured must decide what to do nextto avoid repeating the past; those who have donenothing must awake to their vulnerability. To sur-vive, companies must understand what good corpo-rate strategy is.

A SOBER PICTURE

While there is disquiet about the success of corporatestrategies, none of the available evidence satisfacto-rily indicates the success or failure of corporate strat-egy. Most studies have approached the question bymeasuring the stock market valuation of mergers,captured in the movement of the stock prices ofacquiring companies immediately before and aftermergers are announced.

Michael E. Porter is professor of business administrationat the Harvard Business School and author of CompetitiveAdvantage (Free Press, 1985) and Competitive Strategy(Free Press, 1980).

Copyright © 1987 by the President and Fellows of Harvard College. All rights reserved.


These studies show that the market values mergersas neutral or slightly negative, hardly cause for seri-ous concern.1 Yet the short-term market reaction is ahighly imperfect measure of the long-term success ofdiversification, and no self-respecting executivewould judge a corporate strategy this way.

Studying the diversification programs of a com-pany over a long period of time is a much more tellingway to determine whether a corporate strategy hassucceeded or failed. My study of 33 companies, manyof which have reputations for good management, is aunique look at the track record of major corporations.(For an explanation of the research, see the insert“Where the Data Come From.”) Each company en-tered an average of 80 new industries and 27 newfields. Just over 70% of the new entries were acquisi-tions, 22% were start-ups, and 8% were joint ven-tures. IBM, Exxon, Du Pont, and 3M, for example,focused on start-ups, while ALCO Standard, Beatrice,and Sara Lee diversified almost solely through acquisi-tions (Exhibit 1 has a complete rundown).

My data paint a sobering picture of the success ratioof these moves (see Exhibit 2). I found that on averagecorporations divested more than half their acquisi-tions in new industries and more than 60% of theiracquisitions in entirely new fields. Fourteen compa-nies left more than 70% of all the acquisitions theyhad made in new fields. The track record in unrelatedacquisitions is even worse—the average divestmentrate is a startling 74% (see Exhibit 3). Even a highlyrespected company like General Electric divested avery high percentage of its acquisitions, particularlythose in new fields. Companies near the top of thelist in Exhibit 2 achieved a remarkably low rate ofdivestment. Some bear witness to the success ofwell-thought-out corporate strategies. Others, how-ever, enjoy a lower rate simply because they have notfaced up to their problem units and divested them.

I calculated total shareholder returns (stock priceappreciation plus dividends) over the period of thestudy for each company so that I could compare themwith its divestment rate. While companies near thetop of the list have above-average shareholder re-turns, returns are not a reliable measure of diversifi-cation success. Shareholder return often dependsheavily on the inherent attractiveness of companies’base industries. Companies like CBS and GeneralMills had extremely profitable base businesses thatsubsidized poor diversification track records.

I would like to make one comment on the use ofshareholder value to judge performance. Linkingshareholder value quantitatively to diversificationperformance only works if you compare the share-holder value that is with the shareholder value thatmight have been without diversification. Becausesuch a comparison is virtually impossible to make,

measuring diversification success—the number ofunits retained by the company—seems to be as goodan indicator as any of the contribution of diversifica-tion to corporate performance.

My data give a stark indication of the failure ofcorporate strategies.2 Of the 33 companies, 6 had beentaken over as my study was being completed (see thenote on Exhibit 2). Only the lawyers, investmentbankers, and original sellers have prospered in mostof these acquisitions, not the shareholders.

PREMISES OF CORPORATE STRATEGY

Any successful corporate strategy builds on a numberof premises. These are facts of life about diversifica-tion. They cannot be altered, and when ignored, theyexplain in part why so many corporate strategies fail.

Competition Occurs at the Business Unit Level. Di-versified companies do not compete; only their busi-ness units do. Unless a corporate strategy placesprimary attention on nurturing the success of eachunit, the strategy will fail, no matter how elegantlyconstructed. Successful corporate strategy must growout of and reinforce competitive strategy.

Diversification Inevitably Adds Costs and Con-straints to Business Units. Obvious costs such as thecorporate overhead allocated to a unit may not be asimportant or subtle as the hidden costs and con-straints. A business unit must explain its decisionsto top management, spend time complying with plan-ning and other corporate systems, live with parentcompany guidelines and personnel policies, and forgothe opportunity to motivate employees with directequity ownership. These costs and constraints can bereduced but not entirely eliminated.

Shareholders Can Readily Diversify Themselves.Shareholders can diversify their own portfolios ofstocks by selecting those that best match their pref-erences and risk profiles.3 Shareholders can oftendiversify more cheaply than a corporation becausethey can buy shares at the market price and avoidhefty acquisition premiums.

These premises mean that corporate strategy can-not succeed unless it truly adds value—to businessunits by providing tangible benefits that offset theinherent costs of lost independence and to sharehold-ers by diversifying in a way they could not replicate.

PASSING THE ESSENTIAL TESTS

To understand how to formulate corporate strategy,it is necessary to specify the conditions under which

HARVARD BUSINESS REVIEW May–June 1987 3This document is authorized for use only in EMBA Corporate Strategy May - July 2016 by Professor Jack Goodwin, Melbourne Business School from April 2016 to August 2016.

EXH

IBIT

1D

iver

sific

atio

nPr

ofile

sof

33Le

adin

gU

.S.C

ompa

nies

,195

0–19

86

Com

pany

Num

ber

Tota

lEn

trie

s

All

Entr

ies

into

New

Indus

trie

sPe

rcen

tA

cqui

sitio

nsPe

rcen

t Jo

int

Ven

ture

sPe

rcen

tSt

art

-ups

Entr

ies

into

New

Indus

-tr

ies

That

Rep

rese

nted

Entir

ely

New

Fiel

ds

Perc

ent

Acq

uisi

tions

Perc

ent

Join

tV

entu

res

Perc

ent

Start

-ups

ALC

O S

tand

ard

221

165

99%

0%1%

5610

0%0%

0%A

llied

Cor

p.77

4967

1022

1765

629

Beat

rice

382

204

971

261

970

3Bo

rden

170

9677

419

3275

322

CBS

148

8167

1617

2865

2114

Con

tinen

talG

roup

7547

776

1719

7911

11C

umm

ins

Engi

ne30

2454

1729

1346

2331

Du

Pont

8039

3316

5119

370

63Ex

xon

7956

345

6117

296

65G

ener

alEl

ectri

c16

010

847

2033

2948

1438

Gen

eral

Food

s92

5391

46

2286

59

Gen

eral

Mill

s11

010

284

79

2774

719

W.R

.Gra

ce27

520

283

710

6674

521

Gul

f & W

este

rn17

814

091

46

4888

210

IBM

4638

1818

6316

190

81IC

Indu

strie

s67

4185

312

1788

66

ITT

246

178

892

950

920

8Jo

hnso

n&

John

son

8877

770

2318

560

44M

obil

4132

5316

3115

607

33Pr

octe

r & G

ambl

e28

2361

039

1479

021

Rayt

heon

7058

869

516

8119

6

4 HARVARD BUSINESS REVIEW May–June 1987


EXH

IBIT

1D

iver

sific

atio

nPr

ofile

sof

33Le

adin

gU

.S.C

ompa

nies

,195

0–19

86(C

ontin

ued)

Com

pany

Num

ber

Tota

lEn

trie

s

All

Entr

ies

into

New

Indus

trie

sPe

rcen

tA

cqui

sitio

nsPe

rcen

tJo

int

Ven

ture

sPe

rcen

tSt

art

-ups

Entr

ies

into

New

Indus

-tr

ies

That

Rep

rese

nted

Entir

ely

New

Fiel

ds

Perc

ent

Acq

uisi

tions

Perc

ent

Join

tV

entu

res

Perc

ent

Start

-ups

RCA

5346

3515

5019

3721

42Ro

ckw

ell

101

7573

243

2774

224

Sara

Lee

197

141

961

441

952

2Sc

ovill

5236

970

312

920

8Si

gnal

5345

674

2920

750

25Te

nnec

o85

6281

613

2673

819

3M14

412

554

245

3471

356

TRW

119

8277

1013

2864

1125

Uni

ted

Tech

nolo

gies

6249

5718

2417

2317

39W

estin

ghou

se12

973

6311

2636

613

36W

icke

s71

4783

017

2268

032

Xero

x59

5066

628

1850

1139

Tota

l3,

788

2,64

490

6

Ave

rage

114.

880

.170

.37.

921

.827

.467

.97.

025

.9

Not

es:B

eatri

ce,C

ontin

enta

lGro

up,G

ener

alFo

ods,

RCA

,Sco

vill,

and

Sign

alw

ere

take

nov

eras

the

study

was

bein

gco

mpl

eted

.The

irda

taco

vert

hepe

riod

upth

roug

hta

keov

erbu

tnot

subs

eque

ntdi

vestm

ents.

The

perc

enta

geav

erag

esm

ayno

tadd

upto

100%

beca

use

ofro

undi

ngof

f.


diversification will truly create shareholder value.These conditions can be summarized in three essen-tial tests:

1. The attractiveness test. The industries chosenfor diversification must be structurally attrac-tive or capable of being made attractive.

2. The cost-of-entry test. The cost of entry mustnot capitalize all the future profits.

3. The better-off test. Either the new unit mustgain competitive advantage from its link withthe corporation or vice versa.

Of course, most companies will make certain thattheir proposed strategies pass some of these tests. Butmy study clearly shows that when companies ignoredone or two of them, the strategic results were disas-trous.

How Attractive Is the Industry?In the long run, the rate of return available fromcompeting in an industry is a function of its underly-ing structure, which I have described in another HBRarticle.4 An attractive industry with a high averagereturn on investment will be difficult to enter be-cause entry barriers are high, suppliers and buyershave only modest bargaining power, substitute prod-ucts or services are few, and the rivalry among com-petitors is stable. An unattractive industry like steelwill have structural flaws, including a plethora ofsubstitute materials, powerful and price-sensitivebuyers, and excessive rivalry caused by high fixedcosts and a large group of competitors, many of whomare state supported.

Diversification cannot create shareholder value

Where the data come from

We studied the 1950–1986 diversification histories of33 large diversified U.S. companies. They were cho-sen at random from many broad sectors of the econ-omy.

To eliminate distortions caused by World War II, wechose 1950 as the base year and then identified eachbusiness the company was in. We tracked every acqui-sition, joint venture, and start-up made over this pe-riod—3,788 in all. We classified each as an entry intoan entirely new sector or field (financial services, forexample), a new industry within a field the companywas already in (insurance, for example), or a geo-graphic extension of an existing product or service.We also classified each new field as related or unre-lated to existing units. Then we tracked whether andwhen each entry was divested or shut down and thenumber of years each remained part of the corpora-tion.

Our sources included annual reports, 10K forms, theF&S Index, and Moody’s, supplemented by our judg-ment and general knowledge of the industries in-volved. In a few cases, we asked the companies spe-cific questions.

It is difficult to determine the success of an entry with-out knowing the full purchase or start-up price, theprofit history, the amount and timing of ongoing invest-ments made in the unit, whether any write-offs or write-downs were taken, and the selling price and terms ofsale. Instead, we employed a relatively simple way togauge success: whether the entry was divested or shutdown. The underlying assumption is that a companywill generally not divest or close down a successful

business except in a comparatively few special cases.Companies divested many of the entries in our samplewithin five years, a reflection of disappointment withperformance. Of the comparatively few divestmentswhere the company disclosed a loss or gain, the divest-ment resulted in a reported loss in more than half thecases.

The data in Exhibit 1 cover the entire 1950–1986period. However, the divestment ratios in Exhibit 2and Exhibit 3 do not compare entries and divestmentsover the entire period because doing so would over-state the success of diversification. Companies usuallydo not shut down or divest new entries immediatelybut hold them for some time to give them an opportu-nity to succeed. Our data show that the average hold-ing period is five to slightly more than ten years,though many divestments occur within five years. To ac-curately gauge the success of diversification, we calcu-lated the percentage of entries made by 1975 and by1980 that were divested or closed down as of January1987. If we had included more recent entries, wewould have biased upward our assessment of how suc-cessful these entries had been.

As compiled, these data probably understate therate of failure. Companies tend to announce acquisi-tions and other forms of new entry with a flourish butdivestments and shutdowns with a whimper, if at all.We have done our best to root out every such transac-tion, but we have undoubtedly missed some. Theremay also be new entries that we did not uncover, butour best impression is that the number is not large.



unless new industries have favorable structures thatsupport returns exceeding the cost of capital. If theindustry doesn’t have such returns, the companymust be able to restructure the industry or gain asustainable competitive advantage that leads to re-turns well above the industry average. An industryneed not be attractive before diversification. In fact,a company might benefit from entering before theindustry shows its full potential. The diversificationcan then transform the industry’s structure.

In my research, I often found companies had sus-pended the attractiveness test because they had avague belief that the industry “fit” very closely withtheir own businesses. In the hope that the corporate“comfort” they felt would lead to a happy outcome,the companies ignored fundamentally poor industrystructures. Unless the close fit allows substantialcompetitive advantage, however, such comfort willturn into pain when diversification results in poorreturns. Royal Dutch Shell and other leading oil com-panies have had this unhappy experience in a numberof chemicals businesses, where poor industry struc-tures overcame the benefits of vertical integrationand skills in process technology.

Another common reason for ignoring the attrac-tiveness test is a low entry cost. Sometimes the buyerhas an inside track or the owner is anxious to sell.Even if the price is actually low, however, a one-shotgain will not offset a perpetually poor business. Al-most always, the company finds it must reinvest inthe newly acquired unit, if only to replace fixed assetsand fund working capital.

Diversifying companies are also prone to use rapidgrowth or other simple indicators as a proxy for atarget industry’s attractiveness. Many that rushedinto fast-growing industries (personal computers,video games, and robotics, for example) were burnedbecause they mistook early growth for long-termprofit potential. Industries are profitable not becausethey are sexy or high tech; they are profitable only iftheir structures are attractive.

What Is the Cost of Entry?Diversification cannot build shareholder value if thecost of entry into a new business eats up its expectedreturns. Strong market forces, however, are workingto do just that. A company can enter new industriesby acquisition or start-up. Acquisitions expose it toan increasingly efficient merger market. An acquirerbeats the market if it pays a price not fully reflectingthe prospects of the new unit. Yet multiple biddersare commonplace, information flows rapidly, andinvestment bankers and other intermediaries workaggressively to make the market as efficient as possi-ble. In recent years, new financial instruments suchas junk bonds have brought new buyers into the

market and made even large companies vulnerable totakeover. Acquisition premiums are high and reflectthe acquired company’s future prospects—sometimestoowell.Philip Morris paid more than four times bookvalue for Seven-Up Company, for example. Simplearithmetic meant that profits had to more than qua-druple to sustain the preacquisition ROI. Since thereproved to be little Philip Morris could add in market-ing prowess to the sophisticated marketing wars inthe soft-drink industry, the result was the unsatisfac-tory financial performance of Seven-Up and ulti-mately the decision to divest.

In a start-up, the company must overcome entrybarriers. It’s a real catch-22 situation, however, sinceattractive industries are attractive because their en-try barriers are high. Bearing the full cost of the entrybarriers might well dissipate any potential profits.Otherwise, other entrants to the industry would havealready eroded its profitability.

In the excitement of finding an appealing newbusiness, companies sometimes forget to apply thecost-of-entry test. The more attractive a new indus-try, the more expensive it is to get into.

Will the Business Be Better Off?A corporation must bring some significant competi-tive advantage to the new unit, or the new unit mustoffer potential for significant advantage to the corpo-ration. Sometimes, the benefits to the new unit ac-crue only once, near the time of entry, when theparent instigates a major overhaul of its strategy orinstalls a first-rate management team. Other diversi-fication yields ongoing competitive advantage if thenew unit can market its product through the well-de-veloped distribution system of its sister units, forinstance. This is one of the important underpinningsof the merger of Baxter Travenol and American Hos-pital Supply.

When the benefit to the new unit comes only once,the parent company has no rationale for holding thenew unit in its portfolio over the long term. Once theresults of the one-time improvement are clear, thediversified company no longer adds value to offset theinevitable costs imposed on the unit. It is best to sellthe unit and free up corporate resources.

The better-off test does not imply that diversifyingcorporate risk creates shareholder value in and ofitself. Doing something for shareholders that theycan do themselves is not a basis for corporate strategy.(Only in the case of a privately held company, inwhich the company’s and the shareholder’s risk arethe same, is diversification to reduce risk valuable forits own sake.) Diversification of risk should only bea by-product of corporate strategy, not a prime moti-vator.

Executives ignore the better-off test most of all or


EXH

IBIT

2A

cqui

sitio

nTr

ack

Reco

rds

ofLe

adin

gU

.S.D

iver

sifie

rsRa

nked

byPe

rcen

t Div

este

d,19

50–1

986

Com

pany

All

Acq

uisi

tions

inN

ewIn

dus

trie

s

Perc

entM

ade

by

1980

and

Then

Div

este

d

Perc

entM

ade

by

1975

and

Then

Div

este

d

Acq

uisi

tions

inN

ewIn

dus

trie

sTh

at

Rep

rese

nted

Entir

ely

New

Fiel

ds

Perc

ent

Made

by

1980

and

Then

Div

este

d

Perc

ent

Made

by

1975

and

Then

Div

este

d

John

son

& Jo

hnso

n59

17%

12%

1033

%14

%Pr

octe

r & G

ambl

e14

1717

1117

17Ra

ythe

on50

1726

1325

33U

nite

dTe

chno

logi

es28

2513

1017

03M

6726

2724

4245

TRW

6327

3118

4038

IBM

733

0*3

330*

Du

Pont

1338

437

6075

Mob

il17

3857

950

50Bo

rden

7439

4024

4550

ICIn

dustr

ies

3542

5015

4644

Tenn

eco

5043

4719

2733

Beat

rice

198

4645

5952

51IT

T15

952

5246

6161

Rock

wel

l55

5657

2071

71A

llied

Cor

p.33

5745

1180

67Ex

xon

1962

20*

580

50*

Sara

Lee

135

6265

3980

76G

ener

alFo

ods

4863

6219

9393

Scov

ill35

6477

1164

70Si

gnal

3065

6315

7067

ALC

O S

tand

ard

164

6570

5672

76



EXH

IBIT

2A

cqui

sitio

nTr

ack

Reco

rds

ofLe

adin

gU

.S.D

iver

sifie

rsRa

nked

byPe

rcen

t Div

este

d,19

50–1

986

(Con

tinue

d)

Com

pany

All

Acq

uisi

tions

inN

ewIn

dus

trie

s

Perc

entM

ade

by

1980

and

Then

Div

este

d

Perc

ent

Made

by

1975

and

Then

Div

este

d

Acq

uisi

tions

inN

ewIn

dus

trie

sTh

at

Rep

rese

nted

New

Fiel

ds

Perc

ent

Made

by

1980

and

Then

Div

este

d

Perc

ent

Made

by

1975

and

Then

Div

este

d

W.R

.Gra

ce16

765

7049

7170

Gen

eral

Elec

tric

5165

7814

100

100

Wic

kes

3867

7215

7370

Wes

tingh

ouse

4668

6922

6159

Xero

x33

7179

910

010

0C

ontin

enta

lGro

up36

7172

1560

60G

ener

alM

ills

8675

7320

6560

Gul

f & W

este

rn12

779

7842

7572

Cum

min

sEn

gine

1380

806

8383

RCA

1680

927

8610

0C

BS54

8789

1888

88

Tota

l2,0

21

661

Ave

rage

per

com

pany

†61.2

53.4

%5

6.5

%20.0

61.2

%61.1

%

*C

ompa

nies

with

thre

eor

few

erac

quisi

tions

by th

ecu

toff

year

.†

Com

pani

esw

ithth

ree

orfe

wer

acqu

isitio

nsby

the

cuto

ffye

arar

eex

clud

edfro

mth

eav

erag

eto

min

imiz

e sta

tistic

aldi

storti

ons.

Not

e:Be

atric

e,C

ontin

enta

lGro

up,G

ener

alFo

ods,

RCA

,Sco

vill,

and

Sign

alw

ere

take

nov

eras

the

study

was

bein

gco

mpl

eted

.The

irda

taco

vert

hepe

riod

upth

roug

hta

keov

erbu

tnot

subs

eque

nt d

ives

tmen

ts.


deal with it through arm waving or trumped-up logicrather than hard strategic analysis. One reason is thatthey confuse company size with shareholder value.In the drive to run a bigger company, they lose sightof their real job. They may justify the suspension ofthe better-off test by pointing to the way they managediversity. By cutting corporate staff to the bone andgiving business units nearly complete autonomy,they believe they avoid the pitfalls. Such thinkingmisses the whole point of diversification, which is tocreate shareholder value rather than to avoid destroy-ing it.

CONCEPTS OF CORPORATE STRATEGY

The three tests for successful diversification set thestandards that any corporate strategy must meet;meeting them is so difficult that most diversificationfails. Many companies lack a clear concept of corpo-rate strategy to guide their diversification or pursue aconcept that does not address the tests. Others failbecause they implement a strategy poorly.

My study has helped me identify four concepts ofcorporate strategy that have been put into prac-tice—portfolio management, restructuring, transfer-ring skills, and sharing activities. While the conceptsare not always mutually exclusive, each rests on adifferent mechanism by which the corporation cre-ates shareholder value and each requires the diversi-fied company to manage and organize itself in adifferent way. The first two require no connectionsamong business units; the second two depend onthem. (See Exhibit 4.) While all four concepts ofstrategy have succeeded under the right circum-stances, today some make more sense than others.Ignoring any of the concepts is perhaps the quickestroad to failure.

Portfolio ManagementThe concept of corporate strategy most in use isportfolio management, which is based primarily ondiversification through acquisition. The corporationacquires sound, attractive companies with compe-tent managers who agree to stay on. While acquiredunits do not have to be in the same industries asexisting units, the best portfolio managers generallylimit their range of businesses in some way, in partto limit the specific expertise needed by top manage-ment.

The acquired units are autonomous, and the teamsthat run them are compensated according to the unitresults. The corporation supplies capital and workswith each to infuse it with professional managementtechniques. At the same time, top management pro-vides objective and dispassionate review of business

unit results. Portfolio managers categorize units bypotential and regularly transfer resources from unitsthat generate cash to those with high potential andcash needs.

In a portfolio strategy, the corporation seeks tocreate shareholder value in a number of ways. It usesits expertise and analytical resources to spot attrac-tive acquisition candidates that the individual share-holder could not. The company provides capital onfavorable terms that reflect corporatewide fundrais-ing ability. It introduces professional managementskills and discipline. Finally, it provides high-qualityreview and coaching, unencumbered by conventionalwisdom or emotional attachments to the business.

The logic of the portfolio management conceptrests on a number of vital assumptions. If a company’sdiversification plan is to meet the attractiveness andcost-of-entry test, it must find good but undervaluedcompanies. Acquired companies must be truly under-valued because the parent does little for the new unitonce it is acquired. To meet the better-off test, thebenefits the corporation provides must yield a signifi-cant competitive advantage to acquired units. Thestyle of operating through highly autonomous busi-ness units must both develop sound business strate-gies and motivate managers.

In most countries, the days when portfolio manage-ment was a valid concept of corporate strategy arepast. In the face of increasingly well-developed capi-tal markets, attractive companies with good manage-ments show up on everyone’s computer screen andattract top dollar in terms of acquisition premium.Simply contributing capital isn’t contributing much.A sound strategy can easily be funded; small to me-dium-size companies don’t need a munificent parent.

Other benefits have also eroded. Large companiesno longer corner the market for professional manage-ment skills; in fact, more and more observers believemanagers cannot necessarily run anything in theabsence of industry-specific knowledge and experi-ence. Another supposed advantage of the portfoliomanagement concept—dispassionate review—restson similarly shaky ground since the added value ofreview alone is questionable in a portfolio of soundcompanies.

The benefit of giving business units completeautonomy is also questionable. Increasingly, a com-pany’s business units are interrelated, drawn togetherby new technology, broadening distribution chan-nels, and changing regulations. Setting strategies ofunits independently may well undermine unit per-formance. The companies in my sample that havesucceeded in diversification have recognized thevalue of interrelationships and understood that astrong sense of corporate identity is as important as



slavish adherence to parochial business unit financialresults.

But it is the sheer complexity of the managementtask that has ultimately defeated even the best port-folio managers. As the size of the company grows,portfolio managers need to find more and more dealsjust to maintain growth. Supervising dozens or even

hundreds of disparate units and under chain-letterpressures to add more, management begins to makemistakes. At the same time, the inevitable costs ofbeing part of a diversified company take their toll andunit performance slides while the whole company’sROI turns downward. Eventually, a new managementteam is installed that initiates wholesale divestmentsand pares down the company to its core businesses.The experiences of Gulf & Western, ConsolidatedFoods (now Sara Lee), and ITT are just a few compara-tively recent examples. Reflecting these realities, theU.S. capital markets today reward companies thatfollow the portfolio management model with a “con-glomerate discount”; they value the whole less thanthe sum of the parts.

In developing countries, where large companies arefew, capital markets are undeveloped, and profes-sional management is scarce, portfolio managementstill works. But it is no longer a valid model forcorporate strategy in advanced economies. Neverthe-less, the technique is in the limelight today in theUnited Kingdom, where it is supported so far by anewly energized stock market eager for excitement.But this enthusiasm will wane—as well it should.Portfolio management is no way to conduct corporatestrategy.

RestructuringUnlike its passive role as a portfolio manager, whenit serves as banker and reviewer, a company thatbases its strategy on restructuring becomes an activerestructurer of business units. The new businessesare not necessarily related to existing units. All thatis necessary is unrealized potential.

The restructuring strategy seeks out undeveloped,sick, or threatened organizations or industries on thethreshold of significant change. The parent inter-venes, frequently changing the unit managementteam, shifting strategy, or infusing the company withnew technology. Then it may make follow-up acqui-sitions to build a critical mass and sell off unneededor unconnected parts and thereby reduce the effectiveacquisition cost. The result is a strengthened com-pany or a transformed industry. As a coda, the parentsells off the stronger unit once results are clear be-cause the parent is no longer adding value and topmanagement decides that its attention should bedirected elsewhere. (See the insert “An Uncanny Brit-ish Restructurer” for an example of restructuring.)

When well implemented, the restructuring con-cept is sound, for it passes the three tests of successfuldiversification. The restructurer meets the cost-of-entry test through the types of company it acquires.It limits acquisition premiums by buying companieswith problems and lackluster images or by buyinginto industries with as yet unforeseen potential. In-

An uncanny British restructurer

Hanson Trust, on its way to becoming Britain’s largestcompany, is one of several skillful followers of the re-structuring concept. A conglomerate with units in manyindustries, Hanson might seem on the surface a portfo-lio manager. In fact, Hanson and one or two other con-glomerates have a much more effective corporate strat-egy. Hanson has acquired companies such as LondonBrick, Ever Ready Batteries, and SCM, which the cityof London rather disdainfully calls “low tech.’’

Although a mature company suffering from lowgrowth, the typical Hanson target is not just in any in-dustry; it has an attractive structure. Its customer andsupplier power is low and rivalry with competitors mod-erate. The target is a market leader, rich in assets butformerly poor in management. Hanson pays little ofthe present value of future cash flow out in an acquisi-tion premium and reduces purchase price even furtherby aggressively selling off businesses that it cannot im-prove. In this way, it recoups just over a third of thecost of a typical acquisition during the first six monthsof ownership. Imperial Group’s plush properties in Lon-don lasted barely two months under Hanson owner-ship, while Hanson’s recent sale of Courage Breweriesto Elders recouped £1.4 billion of the original £2.1 bil-lion acquisition price of Imperial Group.

Like the best restructurers, Hanson approaches eachunit with a modus operandi that it has perfectedthrough repetition.

Hanson emphasizes low costs and tight financialcontrols. It has cut an average of 25% of labor costsout of acquired companies, slashed fixed overheads,and tightened capital expenditures. To reinforce itsstrategy of keeping costs low, Hanson carves out de-tailed one-year financial budgets with divisional man-agers and (through generous use of performance-related bonuses and share option schemes) gives themincentive to deliver the goods.

It’s too early to tell whether Hanson will adhere tothe last tenet of restructuring-selling turned-around unitsonce the results are clear. If it succumbs to the allure ofbigness, Hanson may take the course of the failed U.S.conglomerates.


EXH

IBIT

3D

iver

sific

atio

nPe

rform

ance

inJo

intV

entu

res,

Star

t-ups

,and

Unr

elat

edA

cqui

sitio

ns,1

950–

1986

(Com

pani

esin

sam

eor

dera

sin

Exhi

bit2

)

Com

pany

Join

tVen

ture

sas

aPe

rcen

tof

New

Entr

ies

Perc

ent

Made

by

1980

and

Then

Div

este

d

Perc

ent

Made

by

1975

and

Then

Div

este

d

Start

-Ups

as

aPe

rcen

tof

New

Entr

ies

Perc

ent

Made

by

1980

and

Then

Div

este

d

Perc

ent

Made

by

1975

and

Then

Div

este

d

Acq

uisi

tions

inU

nrel

ated

New

Fiel

ds

as

aPe

rcen

tof

Tota

lA

cqui

sitio

nsin

New

Fiel

ds

Perc

ent

Made

by

1980

and

Then

Div

este

d

Perc

ent

Made

by

1975

and

Then

Div

este

d

John

son

& Jo

hnso

n0%

††

23%

14%

20%

0%†

†Pr

octe

r & G

ambl

e0

††

390

09

††

Rayt

heon

960

%60

%5

5050

4640

%40

%U

nite

dTe

chno

logi

es18

5050

2411

2040

0*0*

3M2

100*

100*

452

333

7586

TRW

1020

2513

6371

3971

71IB

M18

100*

†63

2022

3310

0*10

0*D

u Po

nt16

100*

†51

6161

430*

0*M

obil

1633

3331

5056

6760

100

Bord

en4

3333

1917

1321

8080

ICIn

dustr

ies

310

0*10

0*13

8030

3350

50Te

nnec

o6

6767

1367

8042

3340

Beat

rice

1†

†2

00

6359

53IT

T2

0*†

838

5761

6764

Rock

wel

l24

3842

30

035

100

100

Alli

edC

orp.

1010

075

2238

2945

500

Exxo

n5

00

6127

1910

080

50*

Sara

Lee

1†

†4

7510

0*41

7373

Gen

eral

Food

s4

††

667

5042

8683

Scov

ill0

††

310

010

0*45

8010

0Si

gnal

4†

†29

2011

6750

50



EXH

IBIT

3D

iver

sific

atio

nPe

rform

ance

inJo

intV

entu

res,

Star

t-ups

,and

Unr

elat

edA

cqui

sitio

ns,1

950–

1986

(Con

tinue

d)

Com

pany

Join

tVen

ture

sas

aPe

rcen

tof

New

Entr

ies

Perc

ent

Made

by

1980

and

Then

Div

este

d

Perc

ent

Made

by

1975

and

Then

Div

este

d

Start

-Ups

as

aPe

rcen

tof

New

Entr

ies

Perc

ent

Made

by

1980

and

Then

Div

este

d

Perc

ent

Made

by

1975

and

Then

Div

este

d

Acq

uisi

tions

inU

nrel

ated

New

Fiel

ds

as

aPe

rcen

tof

Tota

lA

cqui

sitio

nsin

New

Fiel

ds

Perc

ent

Made

by

1980

and

Then

Div

este

d

Perc

ent

Made

by

1975

and

Then

Div

este

d

ALC

O S

tand

ard

0†

†1

††

6379

81W

.R.G

race

733

3810

7171

3965

65G

ener

alEl

ectri

c20

2033

3333

4436

100

100

Wic

kes

0†

†17

6357

6080

75W

estin

ghou

se11

0*0*

2644

4436

5767

Xero

x6

100*

100*

2850

5622

100

100

Con

tinen

talG

roup

667

6717

140

4083

100

Gen

eral

Mill

s7

7171

989

8065

7767

Gul

f & W

este

rn4

7550

610

010

074

7774

Cum

min

sEn

gine

1750

5029

00

6710

010

0RC

A15

6767

5099

5536

100

100

CBS

1671

7117

8680

3910

010

0

Ave

rage

per

com

pany

‡‡

7.9

%5

0.3

%48.9

%21.8

%44.0

%40.9

%46.1

%74.0

%74.4

%

*C

ompa

nies

with

two

orfe

wer

entri

es.

†N

oen

tries

inth

isca

tego

ry.

‡‡ A

vera

geex

clud

esco

mpa

nies

with

two

orfe

wer

entri

esto

min

imiz

esta

tistic

aldi

storti

ons.

Not

e:Be

atric

e,C

ontin

enta

lGro

up,G

ener

alFo

ods,

RCA

,Sco

vill,

and

Sign

alw

ere

take

nov

eras

the

study

was

bein

gco

mpl

eted

.The

irda

taco

vert

hepe

riod

upth

roug

hta

keov

er,b

utno

tsub

sequ

entd

ives

tmen

ts.


tervention by the corporation clearly meets the bet-ter-off test. Provided that the target industries arestructurally attractive, the restructuring model cancreate enormous shareholder value. Some restructur-ing companies are Loew’s, BTR, and General Cinema.Ironically, many of today’s restructurers are profitingfrom yesterday’s portfolio management strategies.

To work, the restructuring strategy requires a cor-porate management team with the insight to spotundervalued companies or positions in industriesripe for transformation. The same insight is neces-sary to actually turn the units around even thoughthey are in new and unfamiliar businesses.

These requirements expose the restructurer to con-siderable risk and usually limit the time in which thecompany can succeed at the strategy. The most skill-ful proponents understand this problem, recognizetheir mistakes, and move decisively to dispose ofthem. The best companies realize they are not justacquiring companies but restructuring an industry.Unless they can integrate the acquisitions to create awhole new strategic position, they are just portfoliomanagers in disguise. Another important difficultysurfaces if so many other companies join the actionthat they deplete the pool of suitable candidates andbid their prices up.

Perhaps the greatest pitfall, however, is that com-panies find it very hard to dispose of business unitsonce they are restructured and performing well. Hu-man nature fights economic rationale. Size supplantsshareholder value as the corporate goal. The companydoes not sell a unit even though the company nolonger adds value to the unit. While the transformedunits would be better off in another company that hadrelated businesses, the restructuring company in-stead retains them. Gradually, it becomes a portfoliomanager. The parent company’s ROI declines as theneed for reinvestment in the units and normal busi-ness risks eventually offset restructuring’s one-shotgain. The perceived need to keep growing intensifiesthe pace of acquisition; errors result and standardsfall. The restructuring company turns into a con-glomerate with returns that only equal the average ofall industries at best.

Transferring SkillsThe purpose of the first two concepts of corporatestrategy is to create value through a company’s rela-tionship with each autonomous unit. The corpora-tion’s role is to be a selector, a banker, and an inter-venor.

The last two concepts exploit the interrelation-ships between businesses. In articulating them, how-ever, one comes face-to-face with the often ill-definedconcept of synergy. If you believe the text of thecountless corporate annual reports, just about any-

thing is related to just about anything else! But imag-ined synergy is much more common than real synergy.GM’s purchase of Hughes Aircraft simply becausecars were going electronic and Hughes was an elec-tronics concern demonstrates the folly of paper syn-ergy. Such corporate relatedness is an ex post factorationalization of a diversification undertaken forother reasons.

Even synergy that is clearly defined often fails tomaterialize. Instead of cooperating, business unitsoften compete. A company that can define the syner-gies it is pursuing still faces significant organizationalimpediments in achieving them.

But the need to capture the benefits of relationshipsbetween businesses has never been more important.Technological and competitive developments al-ready link many businesses and are creating newpossibilities for competitive advantage. In such sec-tors as financial services, computing, office equip-ment, entertainment, and health care, interrelation-ships among previously distinct businesses areperhaps the central concern of strategy.

To understand the role of relatedness in corporatestrategy, we must give new meaning to this ill-de-fined idea. I have identified a good way to start—thevalue chain.5 Every business unit is a collection ofdiscrete activities ranging from sales to accountingthat allow it to compete. I call them value activities.It is at this level, not in the company as a whole, thatthe unit achieves competitive advantage. I groupthese activities in nine categories. Primary activitiescreate the product or service, deliver and market it,and provide after-sale support. The categories of pri-mary activities include inbound logistics, operations,outbound logistics, marketing and sales, and service.Support activities provide the inputs and infrastruc-ture that allow the primary activities to take place.The categories are company infrastructure, humanresource management, technology development, andprocurement.

The value chain defines the two types of interrela-tionships that may create synergy. The first is acompany’s ability to transfer skills or expertiseamong similar value chains. The second is the abilityto share activities. Two business units, for example,can share the same sales force or logistics network.

The value chain helps expose the last two (andmost important) concepts of corporate strategy. Thetransfer of skills among business units in the diversi-fied company is the basis for one concept. While eachbusiness unit has a separate value chain, knowledgeabout how to perform activities is transferred amongthe units. For example, a toiletries business unit,expert in the marketing of convenience products,transmits ideas on new positioning concepts, promo-tional techniques, and packaging possibilities to a



newly acquired unit that sells cough syrup. Newlyentered industries can benefit from the expertise ofexisting units and vice versa.

These opportunities arise when business unitshave similar buyers or channels, similar value activi-ties like government relations or procurement, simi-larities in the broad configuration of the value chain(for example, managing a multisite service organiza-tion), or the same strategic concept (for example, lowcost). Even though the units operate separately, suchsimilarities allow the sharing of knowledge.

Of course, some similarities are common; one canimagine them at some level between almost any pairof businesses. Countless companies have fallen intothe trap of diversifying too readily because of simi-larities; mere similarity is not enough.

Transferring skills leads to competitive advantageonly if the similarities among businesses meet threeconditions:

1. The activities involved in the businesses aresimilar enough that sharing expertise is mean-ingful. Broad similarities (marketing intensive-ness, for example, or a common core processtechnology such as bending metal) are not asufficient basis for diversification. The result-ing ability to transfer skills is likely to havelittle impact on competitive advantage.

2. The transfer of skills involves activities impor-tant to competitive advantage. Transferringskills in peripheral activities such as govern-ment relations or real estate in consumer goodsunits may be beneficial but is not a basis fordiversification.

3. The skills transferred represent a significantsource of competitive advantage for the receiv-ing unit. The expertise or skills to be transferredare both advanced and proprietary enough to bebeyond the capabilities of competitors.

The transfer of skills is an active process thatsignificantly changes the strategy or operations of thereceiving unit. The prospect for change must be spe-cific and identifiable. Almost guaranteeing that noshareholder value will be created, too many compa-nies are satisfied with vague prospects or faint hopesthat skills will transfer. The transfer of skills does nothappen by accident or by osmosis. The company willhave to reassign critical personnel, even on a perma-nent basis, and the participation and support of high-level management in skills transfer is essential.Many companies have been defeated at skills transferbecause they have not provided their business unitswith any incentives to participate.

Transferring skills meets the tests of diversifica-tion if the company truly mobilizes proprietary ex-pertise across units. This makes certain the company

can offset the acquisition premium or lower the costof overcoming entry barriers.

The industries the company chooses for diversifi-cation must pass the attractiveness test. Even a closefit that reflects opportunities to transfer skills maynot overcome poor industry structure. Opportunitiesto transfer skills, however, may help the companytransform the structures of newly entered industriesand send them in favorable directions.

The transfer of skills can be one-time or ongoing.If the company exhausts opportunities to infuse newexpertise into a unit after the initial postacquisitionperiod, the unit should ultimately be sold. The cor-poration is no longer creating shareholder value. Fewcompanies have grasped this point, however, andmany gradually suffer mediocre returns. Yet a com-pany diversified into well-chosen businesses cantransfer skills eventually in many directions. If cor-porate management conceives of its role in this wayand creates appropriate organizational mechanismsto facilitate cross-unit interchange, the opportunitiesto share expertise will be meaningful.

By using both acquisitions and internal develop-ment, companies can build a transfer-of-skills strat-egy. The presence of a strong base of skills sometimescreates the possibility for internal entry instead of theacquisition of a going concern. Successful diversifiersthat employ the concept of skills transfer may, how-ever, often acquire a company in the target industryas a beachhead and then build on it with their internalexpertise. By doing so, they can reduce some of therisks of internal entry and speed up the process. Twocompanies that have diversified using the transfer-of-skills concept are 3M and Pepsico.

Sharing ActivitiesThe fourth concept of corporate strategy is based onsharing activities in the value chains among businessunits. Procter & Gamble, for example, employs acommon physical distribution system and sales forcein both paper towels and disposable diapers. McKes-son, a leading distribution company, will handle suchdiverse lines as pharmaceuticals and liquor throughsuperwarehouses.

The ability to share activities is a potent basisfor corporate strategy because sharing often en-hances competitive advantage by lowering cost orraising differentiation. But not all sharing leads tocompetitive advantage, and companies can encoun-ter deep organizational resistance to even beneficialsharing possibilities. These hard truths have ledmany companies to reject synergy prematurely andretreat to the false simplicity of portfolio manage-ment.

A cost-benefit analysis of prospective sharing op-portunities can determine whether synergy is possi-


EXH

IBIT

4C

once

pts

ofC

orpo

rate

Stra

tegy

PORT

FOLIO

MA

NA

GEM

ENT

REST

RUC

TURI

NG

TRA

NSF

ERRI

NG

SKI

LLS

SHA

RIN

G A

CTI

VITI

ES

Stra

tegi

cPr

ereq

uisit

esSu

perio

rins

ight

into

iden

tifyi

ngan

dac

quiri

ng u

nder

valu

edco

mpa

nies

Will

ingn

ess

to s

ell o

fflo

sers

qui

ckly

orto

opp

ortu

nisti

cally

dive

stgo

odpe

rform

ers

whe

nbu

yers

are

will

ing

topa

yla

rge

prem

ium

s

Broa

d gu

idel

ines

fora

ndco

nstra

ints

onth

ety

pes

of u

nits

inth

epo

rtfol

ioso

that

seni

orm

anag

emen

t can

play

the

revi

ewro

leef

fect

ivel

y

A p

rivat

eco

mpa

ny o

rund

evel

oped

capi

talm

arke

ts

Abi

lity

tosh

ift a

way

from

portf

olio

man

agem

ent a

s th

eca

pita

lmar

kets

getm

ore

effic

ient

or th

eco

mpa

nyge

ts un

wie

ldy

Supe

rior i

nsig

htin

toid

entif

ying

restr

uctu

ring

oppo

rtuni

ties

Will

ingn

ess

and

capa

bilit

yto

inte

rven

e to

trans

form

acqu

ired

units

Broa

d sim

ilarit

ies

amon

gth

eun

itsin

the

portf

olio

. Will

ingn

ess

tocu

tlo

sses

by

selli

ng o

ff un

itsw

here

restr

uctu

ring

prov

es u

nfea

sible

Will

ingn

ess

to s

ell u

nits

whe

nre

struc

turin

gis

com

plet

e, th

ere

sults

are

clea

r, an

dm

arke

tco

nditi

ons

are

favo

rabl

e

Prop

rieta

ry s

kills

in a

ctiv

ities

impo

rtant

toco

mpe

titiv

ead

vant

age

in ta

rget

indu

strie

s

Abi

lity

to a

ccom

plish

the

trans

fer o

fski

llsam

ong

units

onan

ongo

ing

basis

Acq

uisit

ions

of b

each

head

posit

ions

inne

win

dustr

ies

as a

base

Act

iviti

esin

exi

sting

units

that

can

besh

ared

with

new

busin

ess

units

to g

ain

com

petit

ive

adva

ntag

e

Bene

fits

of s

harin

gth

atou

t wei

ghth

eco

sts

Both

start-

ups

and

acqu

isitio

ns a

s en

tryve

hicl

es

Abi

lity

to o

verc

ome

orga

niza

tiona

lre

sista

nce

tobu

sines

sun

itco

llabo

ratio

n



EXH

IBIT

4C

once

pts

ofC

orpo

rate

Stra

tegy

(Con

tinue

d)

PORT

FOLIO

MA

NA

GEM

ENT

REST

RUC

TURI

NG

TRA

NSF

ERRI

NG

SKI

LLS

SHA

RIN

GA

CTI

VITI

ES

Org

aniz

atio

nal

Prer

equi

sites

Aut

onom

ous

busin

ess

units

Ave

rysm

all,

low

-cos

t,co

rpor

ate

staff

Ince

ntiv

es b

ased

larg

ely

onbu

sines

s un

itre

sults

Aut

onom

ous

busin

ess

units

A c

orpo

rate

org

aniz

atio

nw

ithth

eta

lent

and

reso

urce

sto

over

see

the

turn

arou

nds

and

strat

egic

repo

sitio

ning

s of

acqu

ired

units

Ince

ntiv

es b

ased

larg

ely

onac

quire

dun

its’ r

esul

ts

Larg

ely

auto

nom

ous

but

colla

bora

tive

busin

ess

units

Hig

h-le

vel c

orpo

rate

sta

ffm

embe

rs w

hose

e th

eir r

ole

prim

arily

as

inte

grat

ors

Cro

ss-b

usin

ess-u

nit

com

mitt

ees,

task

forc

es,a

ndot

her f

orm

s to

ser

ve a

sfo

cal

poin

ts fo

r cap

turin

gan

dtra

nsfe

rring

ski

lls

Obj

ectiv

esof

line

man

ager

sth

at in

clud

esk

ills

trans

fer

Ince

ntiv

es b

ased

in p

art o

nco

rpor

ate

resu

lts

Stra

tegi

cbu

sines

sun

itsth

atar

een

cour

aged

to s

hare

activ

ities

An

activ

e str

ateg

ic p

lann

ing

role

atgr

oup,

sect

or,a

ndco

rpor

ate

leve

ls

Hig

h-le

vel c

orpo

rate

staff

mem

bers

who

see

thei

r rol

espr

imar

ily a

s in

tegr

ator

s

Ince

ntiv

es b

ased

hea

vily

on

grou

pan

dco

rpor

ate

resu

lts

Com

mon

Pitfa

llsPu

rsui

ng p

ortfo

lio m

anag

emen

tin

coun

tries

with

effi

cien

t cap

ital

mar

ketin

gan

da

deve

lope

dpo

olof

prof

essio

nal m

anag

emen

t tal

ent

Igno

ring

the

fact

that

indu

stry

struc

ture

is no

tattr

activ

e

Mist

akin

gra

pid

grow

thor

a“h

ot”

indu

stry

as s

uffic

ient

evid

ence

of a

restr

uctu

ring

oppo

rtuni

ty

Lack

ing

the

reso

lve

or re

sour

ces

to ta

ke o

ntro

uble

dsit

uatio

nsan

dto

inte

rven

ein

man

agem

ent

Igno

ring

the

fact

that

indu

stry

struc

ture

is n

otat

tract

ive

Payi

nglip

ser

vice

tore

struc

turin

gbu

tact

ually

prac

ticin

gpa

ssiv

epo

rtfol

iom

anag

emen

t

Mist

akin

gsim

ilarit

yor

com

fort

with

new

busin

esse

sas

suffi

cien

t bas

isfo

rdi

vers

ifica

tion

Prov

idin

g no

prac

tical

way

for

skill

s tra

nsfe

rto

occu

r

Igno

ring

the

fact

that

indu

stry

struc

ture

is no

tattr

activ

e

Shar

ing

fori

ts ow

n sa

kera

ther

than

bec

ause

it le

ads

toco

mpe

titiv

ead

vant

age

Ass

umin

gsh

arin

g w

illoc

cur

natu

rally

with

outs

enio

rm

anag

emen

t pla

ying

anac

tive

role

Igno

ring

the

fact

that

indu

stry

struc

ture

isno

tat

tract

ive


ble. Sharing can lower costs if it achieves economiesof scale, boosts the efficiency of utilization, or helpsa company move more rapidly down the learningcurve. The costs of General Electric’s advertising,sales, and after-sales service activities in major appli-ances are low because they are spread over a widerange of appliance products. Sharing can also enhancethe potential for differentiation. A shared order-pro-cessing system, for instance, may allow new featuresand services that a buyer will value. Sharing can alsoreduce the cost of differentiation. A shared servicenetwork, for example, may make more advanced,remote servicing technology economically feasible.Often, sharing will allow an activity to be whollyreconfigured in ways that can dramatically raise com-petitive advantage.

Sharing must involve activities that are significantto competitive advantage, not just any activity.P&G’s distribution system is such an instance in thediaper and paper towel business, where products arebulky and costly to ship. Conversely, diversificationbased on the opportunities to share only corporateoverhead is rarely, if ever, appropriate.

Sharing activities inevitably involves costs that thebenefits must outweigh. One cost is the greater coor-dination required to manage a shared activity. Moreimportant is the need to compromise the design orperformance of an activity so that it can be shared. Asalesperson handling the products of two businessunits, for example, must operate in a way that isusually not what either unit would choose were itindependent. And if compromise greatly erodes theunit’s effectiveness, then sharing may reduce ratherthan enhance competitive advantage.

Many companies have only superficially identifiedtheir potential for sharing. Companies also mergeactivities without consideration of whether they aresensitive to economies of scale. When they are not,the coordination costs kill the benefits. Companiescompound such errors by not identifying costs ofsharing in advance, when steps can be taken to mini-mize them. Costs of compromise can frequently bemitigated by redesigning the activity for sharing. Theshared salesperson, for example, can be provided witha remote computer terminal to boost productivityand provide more customer information. Jamming

Adding value with hospitality

Marriott began in the restaurant business in Washing-ton, D.C. Because its customers often ordered takeoutson the way to the national airport, Marriott eventuallyentered airline catering. From there, it jumped intofood service management for institutions. Marriott thenbegan broadening its base of family restaurants andentered the hotel industry. More recently, it has movedinto restaurants, snack bars, and merchandise shops inairport terminals and into gourmet restaurants. In addi-tion, Marriott has branched out from its hotel businessinto cruise ships, theme parks, wholesale travel agen-cies, budget motels, and retirement centers.

Marriott’s diversification has exploited well-devel-oped skills in food service and hospitality. Marriott’skitchens prepare food according to more than 6,000standardized recipe cards; hotel procedures are alsostandardized and painstakingly documented in elabo-rate manuals. Marriott shares a number of importantactivities across units. A shared procurement and distri-bution system for food serves all Marriott units throughnine regional procurement centers. As a result, Marri-ott earns 50% higher margins on food service thanany other hotel company. Marriott also has a fully inte-grated real estate unit that brings corporatewidepower to bear on site acquisitions as well as on the de-signing and building of all Marriott locations.

Marriott’s diversification strategy balances acquisi-tions and start-ups. Start-ups or small acquisitions areused for initial entry, depending on how close the op-portunities for sharing are. To expand its geographicbase, Marriott acquires companies and then disposesof the parts that do not fit.

Apart from this success, it is important to note thatMarriott has divested 36% of both its acquisitions andits start-ups. While this is an above-average record,Marriott’s mistakes are quite illuminating. Marriott haslargely failed in diversifying into gourmet restaurants,theme parks, cruise ships, and wholesale travel agen-cies. In the first three businesses, Marriott discovered itcould not transfer skills despite apparent similarities.Standardized menus did not work well in gourmet res-taurants. Running cruise ships and theme parks wasbased more on entertainment and pizzazz than thecarefully disciplined management of hotels and mid-price restaurants. The wholesale travel agencies wereill fated from the start because Marriott had to com-pete with an important customer for its hotels and hadno proprietary skills or opportunities to share withwhich to add value.



business units together without such thinking exac-erbates the costs of sharing.

Despite such pitfalls, opportunities to gain advan-tage from sharing activities have proliferated becauseof momentous developments in technology, deregu-lation, and competition. The infusion of electronicsand information systems into many industries cre-ates new opportunities to link businesses. The corpo-rate strategy of sharing can involve both acquisitionand internal development. Internal development isoften possible because the corporation can bring tobear clear resources in launching a new unit. Start-upsare less difficult to integrate than acquisitions. Com-panies using the shared-activities concept can alsomake acquisitions as beachhead landings into a newindustry and then integrate the units through sharingwith other units. Prime examples of companies thathave diversified via using shared activities includeP&G, Du Pont, and IBM. The fields into which eachhas diversified are a cluster of tightly related units.Marriott illustrates both successes and failures insharing activities over time. (See the insert “AddingValue with Hospitality.”)

Following the shared-activities model requires anorganizationalcontextinwhichbusinessunitcollabo-ration is encouraged and reinforced. Highly autono-mous business units are inimical to such collabora-tion. The company must put into place a variety ofwhat I call horizontal mechanisms—a strong sense ofcorporate identity, a clear corporate mission state-ment that emphasizes the importance of integratingbusiness unit strategies, an incentive system that re-wards more than just business unit results, cross-busi-ness-unittaskforces, and other methods of integrating.

A corporate strategy based on shared activitiesclearlymeets thebetter-offtestbecausebusinessunitsgain ongoing tangible advantages from others withinthe corporation. It also meets the cost-of-entry test byreducing the expense of surmounting the barriers tointernal entry. Other bids for acquisitions that do notshareopportunitieswillhavelowerreservationprices.Even widespread opportunities for sharing activitiesdo not allow a company to suspend the attractivenesstest, however. Many diversifiers have made the criti-calmistakeofequating theclose fitof a target industrywithattractivediversification. Target industriesmustpass the strict requirement test of having an attractivestructure as well as a close fit in opportunities if diver-sification is to ultimately succeed.

CHOOSING A CORPORATE STRATEGY

Each concept of corporate strategy allows the diver-sified company to create shareholder value in a dif-ferent way. Companies can succeed with any of the

concepts if they clearly define the corporation’s roleand objectives, have the skills necessary for meetingthe concept’s prerequisites, organize themselves tomanage diversity in a way that fits the strategy, andfind themselves in an appropriate capital market en-vironment. The caveat is that portfolio managementis only sensible in limited circumstances.

A company’s choice of corporate strategy is partlya legacy of its past. If its business units are in unat-tractive industries, the company must start fromscratch. If the company has few truly proprietaryskills or activities it can share in related diversifica-tion, then its initial diversification must rely on otherconcepts. Yet corporate strategy should not be a once-and-for-all choice but a vision that can evolve. Acompany should choose its long-term preferred con-cept and then proceed pragmatically toward it fromits initial starting point.

Both the strategic logic and the experience of thecompanies studied over the last decade suggest thata company will create shareholder value throughdiversification to a greater and greater extent as itsstrategy moves from portfolio management towardsharing activities. Because they do not rely on supe-rior insight or other questionable assumptions aboutthe company’s capabilities, sharing activities andtransferring skills offer the best avenues for valuecreation.

Each concept of corporate strategy is not mutuallyexclusive of those that come before, a potent advan-tage of the third and fourth concepts. A company canemploy a restructuring strategy at the same time ittransfers skills or shares activities. A strategy basedon shared activities becomes more powerful if busi-ness units can also exchange skills. As the Marriottcase illustrates, a company can often pursue the twostrategies together and even incorporate some of theprinciples of restructuring with them. When itchooses industries in which to transfer skills or shareactivities, the company can also investigate the pos-sibility of transforming the industry structure. Whena company bases its strategy on interrelationships, ithas a broader basis on which to create shareholdervalue than if it rests its entire strategy on transform-ing companies in unfamiliar industries.

My study supports the soundness of basing acorporate strategy on the transfer of skills or sharedactivities. The data on the sample companies’ diver-sification programs illustrate some important char-acteristics of successful diversifiers. They have madea disproportionately low percentage ofunrelated acqui-sitions, unrelated being defined as having no clearopportunity to transfer skills or share important ac-tivities (see Exhibit 3). Even successful diversifierssuch as 3M, IBM, and TRW have terrible recordswhen they have strayed into unrelated acquisitions.


Successful acquirers diversify into fields, each ofwhich is related to many others. Procter & Gambleand IBM, for example, operate in 18 and 19 interre-lated fields respectively and so enjoy numerous op-portunities to transfer skills and share activities.

Companies with the best acquisition records tendto make heavier-than-average use of start-ups andjoint ventures. Most companies shy away frommodes of entry besides acquisition. My results castdount on the conventional wisdom regarding start-ups. Exhibit 3 demonstrates that while joint venturesare about as risky as acquisitions, start-ups are not.Moreover, successful companies often have very goodrecords with start-up units, as 3M, P&G, Johnson &Johnson, IBM, and United Technologies illustrate.When a company has the internal strength to start upa unit, it can be safer and less costly to launch acompany than to rely solely on an acquisition andthen have to deal with the problem of integration.Japanese diversification histories support the sound-ness of start-up as an entry alternative.

My data also illustrate that none of the concepts ofcorporate strategy works when industry structure ispoor or implementation is bad, no matter how relatedthe industries are. Xerox acquired companies in re-lated industries, but the businesses had poor struc-tures and its skills were insufficient to provideenough competitive advantage to offset implementa-tion problems.

An Action ProgramTo translate the principles of corporate strategy intosuccessful diversification, a company must first takean objective look at its existing businesses and thevalue added by the corporation. Only through suchan assessment can an understanding of good corpo-rate strategy grow. That understanding should guidefuture diversification as well as the development ofskills and activities with which to select further newbusinesses. The following action program provides aconcrete approach to conducting such a review. Acompany can choose a corporate strategy by:

1. Identifying the interrelationships among al-ready existing business units. A companyshould begin to develop a corporate strategy byidentifying all the opportunities it has to shareactivities or transfer skills in its existing port-folio of business units. The company will notonly find ways to enhance the competitive ad-vantage of existing units but also come uponseveral possible diversification avenues. Thelack of meaningful interrelationships in theportfolio is an equally important finding, sug-gesting the need to justify the value added by

the corporation or, alternately, a fundamentalrestructuring.

2. Selecting the core businesses that will be thefoundation of the corporate strategy. Success-ful diversification starts with an understandingof the core businesses that will serve as the basisfor corporate strategy. Core businesses are thosethat are in an attractive industry, have the po-tential to achieve sustainable competitive ad-vantage, have important interrelationshipswith other business units, and provide skills oractivities that represent a base from which todiversify.

The company must first make certain its corebusinesses are on sound footing by upgradingmanagement, internationalizing strategy, orimproving technology. The study shows thatgeographic extensions of existing units,whether by acquisition, joint venture, or start-up, had a substantially lower divestment ratethan diversification.

The company must then patiently dispose ofthe units that are not core businesses. Sellingthem will free resources that could be betterdeployed elsewhere. In some cases disposal im-plies immediate liquidation, while in others thecompany should dress up the units and wait fora propitious market or a particularly eagerbuyer.

3. Creating horizontal organizational mecha-nisms to facilitate interrelationships amongthe core businesses and lay the groundwork forfuture related diversification. Top manage-ment can facilitate interrelationships by em-phasizing cross-unit collaboration, groupingunits organizationally and modifying incen-tives, and taking steps to build a strong sense ofcorporate identity.

4. Pursuing diversification opportunities that al-low shared activities. This concept of corporatestrategy is the most compelling, provided acompany’s strategy passes all three tests. Acompany should inventory activities in existingbusiness units that represent the strongestfoundation for sharing, such as strong distribu-tion channels or world-class technical facilities.These will in turn lead to potential new busi-ness areas. A company can use acquisitions asa beachhead or employ start-ups to exploit in-ternal capabilities and minimize integratingproblems.

5. Pursuing diversification through the transferof skills if opportunities for sharing activitiesare limited or exhausted. Companies can pur-sue this strategy through acquisition, althoughthey may be able to use start-ups if their existing



units have important skills they can readilytransfer.

Such diversification is often riskier becauseof the tough conditions necessary for it to work.Given the uncertainties, a company shouldavoid diversifying on the basis of skills transferalone. Rather it should also be viewed as astepping-stone to subsequent diversification us-ing shared activities. New industries should bechosen that will lead naturally to other busi-nesses. The goal is to build a cluster of relatedand mutually reinforcing business units. Thestrategy’s logic implies that the companyshould not set the rate of return standards forthe initial foray into a new sector too high.

6. Pursuing a strategy of restructuring if this fitsthe skills of management or no good opportu-nities exist for forging corporate interrelation-ships. When a company uncovers underman-aged companies and can deploy adequatemanagement talent and resources to the ac-quired units, then it can use a restructuringstrategy. The more developed the capital mar-kets and the more active the market for compa-nies, the more restructuring will require a pa-tient search for that special opportunity ratherthan a headlong race to acquire as many badapples as possible. Restructuring can be a per-manent strategy, as it is with Loew’s, or a wayto build a group of businesses that supports ashift to another corporate strategy.

7. Paying dividends so that the shareholders canbe the portfolio managers. Paying dividends isbetter than destroying shareholder valuethrough diversification based on shaky under-pinnings. Tax considerations, which some com-panies cite to avoid dividends, are hardly legiti-mate reasons to diversify if a company cannotdemonstrate the capacity to do it profitably.

CREATING A CORPORATE THEME

Defining a corporate theme is a good way to ensurethat the corporation will create shareholder value.Having the right theme helps unite the efforts ofbusiness units and reinforces the ways they interre-late as well as guides the choice of new businesses toenter. NEC Corporation, with its “C&C” theme,provides a good example. NEC integrates its com-puter, semiconductor, telecommunications, and con-

sumer electronics businesses by merging computersand communication.

It is all too easy to create a shallow corporatetheme. CBS wanted to be an “entertainment com-pany,” for example, and built a group of businessesrelated to leisure time. It entered such industries astoys, crafts, musical instruments, sports teams, andhi-fi retailing. While this corporate theme soundedgood, close listening revealed its hollow ring. Noneof these businesses had any significant opportunityto share activities or transfer skills among them-selves or with CBS’s traditional broadcasting andrecord businesses. They were all sold, often at signifi-cant losses, except for a few of CBS’s publishing-related units. Saddled with the worst acquisitionrecord in my study, CBS has eroded the shareholdervalue created through its strong performance inbroadcasting and records.

Moving from competitive strategy to corporatestrategy is the business equivalent of passing throughthe Bermuda Triangle. The failure of corporate strat-egy reflects the fact that most diversified companieshave failed to think in terms of how they really addvalue. A corporate strategy that truly enhances thecompetitive advantage of each business unit is thebest defense against the corporate raider. With asharper focus on the tests of diversification and theexplicit choice of a clear concept of corporate strat-egy, companies’ diversification track records fromnow on can look a lot different.

1. The studies also show that sellers of companies capture a largefraction of the gains from merger. See Michael C. Jensen andRichard S. Ruback, “The Market for Corporate Control: The Sci-entific Evidence,” Journal of Financial Economics (April 1983): 5,and Michael C. Jensen, “Takeovers: Folklore and Science,” Har-vard Business Review (November–December 1984): 109.

2. Some recent evidence also supports the conclusion thatacquired companies often suffer eroding performance after acqui-sition. See Frederick M. Scherer, “Mergers, Sell-Offs and Manage-rial Behavior,” in The Economics of Strategic Planning, ed. LacyGlenn Thomas (Lexington, Mass.: Lexington Books, 1986), p. 143,and David A. Ravenscraft and Frederick M. Scherer, “Mergers andManagerial Performance,” paper presented at the Conference onTakeovers and Contests for Corporate Control, Columbia LawSchool, 1985.

3. This observation has been made by a number of authors. See,for example, Malcolm S. Salter and Wolf A. Weinhold, Diversifi-cation Through Acquisition (New York: Free Press, 1979).

4. See Michael E. Porter, “How Competitive Forces ShapeStrategy,” Harvard Business Review (March–April 1979): 86.

5. See Michael E. Porter, Competitive Advantage (New York:Free Press, 1985).


Harvard Business ReviewHBR Subscriptions Harvard Business Review

U.S. and Canada Outside U.S. and CanadaSubscription Service Tower HouseP.O. Box 52623 Sovereign ParkBoulder, CO 80322-2623 Lathkill Street

Market HarboroughLeicestershire LE16 9EF

Telephone: (800) 274-3214 Telephone: 44-85-846-8888Fax: (617) 496-8145 Fax: 44-85-843-4958American Express, MasterCard, Visa accepted. Billing available.

HBR Article ReprintsHBR Index and Other CatalogsHBS CasesHBS Press Books

Harvard Business School PublishingCustomer Service - Box 230-560 Harvard WayBoston, MA 02163Telephone: U.S. and Canada (800) 545-7685Outside U.S. and Canada: (617) 495-6117 or 495-6192Fax: (617) 495-6985Internet address: [email protected]

HBS Management ProductionsVideos

Harvard Business School Management Productions videos areproduced by award-winning documentary filmmakers. You’ll findthem lively, engaging, and informative.

HBR Custom Reprints Please inquire about HBR’s custom service for quantity discounts.We will print your company’s logo on the cover of reprints,collections, or books in black and white, two color, or four color. Theprocess is easy, cost effective, and quick.

Telephone: (617) 495-6198 or Fax: (617) 496-8866

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

For permission to re-publish please write or call:

Permissions EditorHarvard Business School PublishingBox 230-560 Harvard WayBoston, MA 02163Telephone: (617) 495-6849


from competitive advantage to corporate strategy · pdf filefrom competitive advantage to...

Documents