<<

. 4
( 11)



>>

nesses, while the families nonetheless retained ultimate control. In Bacardi,
this switch took place in 1996 with the appointment of a professional man-
ager as CEO to replace the great-great-grandson of the founder. Initially this
change in management caused some friction among shareholders (around
500 heirs of Don Facundo Bacardi, the founder of the ¬rm), not only because
the newly appointed CEO was the ¬rst not to be a family member, but also
because he had been previously a mergers and acquisitions specialist in a
Washington law ¬rm with no operating experience in the consumer goods
industry.21
At Heineken, this switch took place in 1989 with the retirement of A. H.
Heineken. The chairmanship passed to a professional manager who had
been working for the company for his whole career. However, like Bacardi,
Heineken remained a family-controlled ¬rm, with Mr. Heineken (and after
his death, his daughter) owning 50.5 percent of the shares of Heineken™s
holding company, which controlled 50.5 percent of the Heineken brewery.22
The switch from personal control systems to managerial control is not,
however, always so straightforward. In some cases, family members remain
as top executives of the ¬rms, sharing the day-to-day management with
hired professionals. In other cases, they delegate the day-to-day man-
agement totally to professionals and become nonexecutive top managers.
Anheuser-Busch, Brown Forman, and Pernod Ricard, are illustrations of the
distinct nature that family control may have.

20 Colin Mayer, “Firm Control,” Inaugural Lecture delivered to the University of Oxford, 18
February 1999.
21 Mr. Reid had been an adviser in the creation of a single global holding company in 1992
to unify ¬ve separate operations of Bacardi in 1992, and had advised Bacardi on the $2
billion acquisition in 1993 of Martini & Rossi, the Italian-owned drinks group. In 2000 he
resigned, after the company decided not to proceed with a planned initial public offering. He
was replaced by Ruben Rodriguez, another hired professional manager who became CEO
and chairman of the group. “A Spirited Strategist,” Financial Times (8 March 1999).
22 Interview with Jan Beijerinck, former Worldwide Client Services Director of Heineken,
Utrecht, 10 March 2001; Heineken, Annual Reports and Accounts (1989); “Head of
Heineken Brewing Family Dies Aged 78,” Financial Times (4 January 2002).
81
Family Ownership and Managerial Control
At the end of the twentieth century, August A. Busch III served as chair-
man of Anheuser-Busch. However, in the year 2000, the ¬rm appointed
a professional manager as CEO and president. He had been working for
Anheuser-Busch for thirty-one years and had a wide knowledge of the ¬rm
and the industry (being involved in building the company™s global leadership
position). However, August A. Busch III remained as chairman.23
At Brown Forman, the chairman and CEO “ Owsley Brown II “ was
also a family member. But, like Augustus Busch III, in 2000 Mr. Brown II
appointed a graduate from Princeton and Harvard Business School, who had
been working for the company since 1963, as president of Brown Forman.24
Since the merger in 1975, Pernod Ricard was run by a professional man-
ager who married into the Pernod family, playing a key role in the creation
of the group Pernod Ricard. In 2000, after his retirement as president, Mr.
Jacquillat became vice-chairman, and Patrick Ricard, another family mem-
ber, became the chairman and CEO of the ¬rm. In addition, two joint man-
aging directors were hired to run the business, neither of whom was a family
member. One had previously served as chairman of Irish Distillers, and the
other had been chairman of Pernod Europe.25
Seagram hired professional managers to run the business after the death
of the founder, Samuel Bronfman, in 1971. One of the managers, Phillip
Beekam, who had previously been president of Colgate International, played
an important role in the introduction of marketing techniques at Seagram
during the 1970s. Throughout this period, Edgar and Charles Bronfman,
the two sons of the founder, kept their positions as chairman and CEO,
thus ensuring tight control of decision taking.26 Seagram reversed the trend
toward independent management in 1990, when Edgar Bronfman Jr.., the
grandson of the founder, took over the management of the ¬rm, until it was
sold to Vivendi in 2001.27


Concentrated Ownership: Does It Matter?
This raises the question of whether it matters if ownership is dispersed or con-
trolled by a family, provided that there is professional management running
the ¬rm. The evidence provided by the alcoholic beverages industry indi-
cates that having ownership concentrated in the hands of a small number of

23 “St. Louis-based Anheuser-Busch announces new CEO,” St. Louis Post-Dispatch (30 July
2000).
24 “Brown Forman Names Street President of the corporation,” PR Newswire (15 November
2000).
25 Pernod-Ricard, Annual Report and Accounts (2000).
26 “Records of the Seagram Company Ltd.,” Record Group 2, Accession 2126, Hagley Museum
and Library; Edgar M. Bronfman, Good Spirits: The Making of a Businessman (New York:
Putman™s, 1998); The Seagram Company Ltd, Annual Report and Accounts (1971, 1977).
27 The Seagram Company Ltd., Annual Report and Accounts (1989, 1990).
82 Global Brands
shareholders, in particular families, has had important implications in the
growth and survival of ¬rms and the evolution of the industry in general.
There are two arguments supporting this. One is of broader application,
and concerns the capacity of ¬rms to overcome the free-rider problems that
af¬‚ict capital markets with dispersed ownership, and also regulation systems
(e.g., such as investor protection).28
The most important reasons are, however, speci¬c to the global alcoholic
beverages industry. In some sectors, such as whisky, port wine or sherry,
brand image is often associated with heritage. Having family members who
also represent the living icons of the brands enhances the heritage image of
those brands. A more important reason concerns the longevity of brands.
Families tend to take more long-term views than professional managers in
their investment decisions. However, the strategies may differ, and depend on
the cultural environment of those ¬rms.29 Similar reasoning may be applied
to the lives of brands. The evidence provided by the world™s most successful
alcoholic beverages brands shows that most are long established, and some
go as far back as the eighteenth and nineteenth centuries.30
A ¬nal reason why family ownership predominates in this industry relates
to the private interests of the entrepreneurs, such as ambition to build an
empire, or preserve the family name, which can be achieved because the cap-
ital required for investment is primarily for marketing, and can be obtained
from retained pro¬ts, without the ¬rm having to recur to capital markets.


The Case of LVMH
Despite the waves of mergers and acquisitions that occurred in the alcoholic
beverages industry from the 1960s, family members often remained as share-
holders of the acquiring ¬rms, taking nonexecutive positions on the boards.
The ownership structure of LVMH after the 1987 merger of Mo¨ t Hennessy
e
with Louis Vuitton is a good illustration of this situation.
The merger between Mo¨ t & Chandon and Hennessy in 1971 united
e
France™s biggest exporters of champagne and cognac, respectively, allowing
the two companies to take advantage of their similarities in terms of the
“personalities” of their brands and the geographical scope of operations, as
well as to economize on costs of, for example, distribution.31

28 A. Schleifer and R. W. Vishny, “Large Shareholders and Corporate Control,” Journal of
Political Economy, Vol. 94 (1986): 461“88; R. La Porta, F. Lopez-de-Silanes, A. Schleifer,
and R. Vishny, “Legal Determinants of External Finance,” Journal of Finance, No. 52 (1997):
1131“50.
29 Jones and Rose, “Family Capitalism”; Church, “The Family.”
30 Paul Duguid, “Developing the Brand: The Case of Alcohol, 1800“1880,” Enterprise &
Society, Vol. 4, No. 3 (2003): 405“41.
31 “Records of Mo¨ t et Chandon, 1971,” LVMH; Mo¨ t-Hennessy, Annual Report and
e e
Accounts (1971).
83
Family Ownership and Managerial Control
In 1987, the families of the newly merged multinational hired a profes-
´
sional manager, Bernard Arnault, who had graduated from the elite Ecole
Polytechnique, worked for his family ¬rm dealing in real estate, and lived
in New York, where he had learned about the aggressiveness of the stock
market. When he took over the management of LVMH, he embarked on
a fast and aggressive process of mergers and acquisitions of other alco-
holic beverages ¬rms, and other French luxury businesses as well, showing
an enormous capacity to detect opportunities and deal with adversity. In
this process, he also became the major shareholder of this French multi-
national.32
By 1988, there were four groups of shareholders: 32 percent of the cap-
ital was held by the consortium Financi` re Agache and Guinness headed
e
by Bernard Arnault; 14 percent was held by the Chandon, Mo¨ t, Mercier,
e
and Hennessy families; 23 percent by the Vuitton family, and 31 per-
cent by the public. The board comprised twelve members, four from the
Agache/Guinness group, four from the Vuitton family, and four from the
Mo¨ t Hennessy family.33 The tight links established between the majority
e
shareholders and the board of directors reduced the risk of opportunistic
behavior by the management of the ¬rm (a risk considered to exist when the
shares are widely dispersed among a large number of shareholders, and no
single shareholder is in a position to control the affairs of the ¬rm). Minority
shareholders had no representation on the board at that time, even though
the ¬rm was publicly quoted.
The case of LVMH also illustrates another trend in the corporate gov-
ernance of alcoholic beverages ¬rms, the creation of interlocking share-
holdings. The British whisky and gin producer Distillers Company had a
profound role in the foundation and rise of Distillers Corporation, later
renamed Seagram. With the formation of LVMH, Guinness became a share-
holder together with Financi` re Agache through a holding company called
e
Jacques Rober (60 percent owned by Financi` re Agache and 40 percent by
e
Guinness). Conversely, LVMH acquired Guinness™s shares, obtaining a 12
percent ownership in that ¬rm. Despite Bernard Arnault™s initial opposi-
tion to the merger of Guinness with Grand Metropolitan to form Diageo in
1997, this interlocking shareholding between the two ¬rms remained until
the beginning of the twenty-¬rst century, and there have been only a few
changes in the percentages of the interlocked shareholdings.34
Together with interlocking shareholdings, it is common for ¬rms to have
interlocking directorships. For example, at the time of the LVMH merger,
Bernard Arnault, who became the company™s chairman, also had a position

32 e`
Michel Refait, Mo¨ t & Chandon: de Claude Mo¨ t a Bernard Arnault (Saints-Geosmes:
e
Dominique Gu´ niot, 1998).
e
33 Barclays de Zoete Wedd, LVMH (1988).
34 “Cognac Threat to Diageo Deal,” The Independent on Sunday (23 July 2000).
84 Global Brands
on Guinness™s board of directors. Conversely, Anthony Tennant, Guinness™s
chairman, had a position on LVMH supervisory board.35
But multiple directorships may also exist independently of interlocking
shareholdings. For example, by 2000, Bernard Arnault was also on the board
of directors of other ¬rms, including Vivendi.36 These multiple directorships,
which usually occur among large ¬rms, are thought not only to bring prestige
to the directors and the ¬rms they manage, but also to reduce transaction
costs, when the ¬rms involved have transactions among themselves. In this
sense, multiple directorships can be considered as a hybrid mode for orga-
nizing transactions, which lies midway between markets and hierarchies.37
The case of Bernard Arnault™s participation in Vivendi™s board of directors
provides, however, an additional insight into the mixture of cooperation
and competition that interlocking directorships may create in the alcoholic
beverages industry. He would have had an interesting view of the fate of
Seagram, which, as mentioned earlier, was ¬rst bought by Vivendi, before
its alcoholic beverages business was sold to Diageo (which owns a substan-
tial share of Arnault™s LVMH) and Pernod Ricard, an important French
competitor.


Other Sources of Concentration of Shareholdings
It is not only families that concentrate ownership of ¬rms. Financial interme-
diaries such as banks, insurance companies and pension funds have also con-
tributed to a reconcentration of corporate power in the global alcoholic bev-
erages industry. Even when families control the ¬rms, these institutions tend
to have a substantial number of shares traded through the capital markets.
An example is Pernod Ricard, in which the French bank Soci´ t´ G´ n´ rale
ee e e
38
has an indirect ownership corresponding to 5.6 percent.
Carlsberg, the leading Danish brewer is yet another case. It is controlled
by a foundation formed by the founder in 1906. In 2000, the Carlsberg
Foundation owned 55 percent of the shares, other Danish investors such
as banks, insurance companies, and pension funds controlled 28.5 percent,
and foreign institutional investors controlled 16.5 percent.39 This association
with other institutions through their investment in the capital of alcoholic
beverages ¬rms may bring several bene¬ts for ¬rms, such as obtaining better

35 Guinness, Annual Report and Accounts (1988); LVMH, Annual Report and Accounts
(1988).
36 Vivendi, Annual Report and Accounts (2000).
37 Oliver E. Williamson, Markets and Hierarchies (New York: Free Press, 1975); Frans N.
Stockman, Rolf Ziegler, and John Scott, Networks of Corporate Power: A Comparative
Analysis of Ten Countries (Cambridge: Polity, 1985): 274.
38 Pernod Ricard, Annual Report and Accounts (2001).
39 Carlsberg, Annual Report and Accounts (1999/2000).
85
Family Ownership and Managerial Control
banks loans, as these tend to accept a higher risk in ¬nancing their own
entrepreneurial ventures.40


Conclusion
The United States and United Kingdom tended to develop large ¬rms ear-
lier than Continental Europe and Japan. However, by the beginning of the
twenty-¬rst century, Continental Europe and Japan had also developed lead-
ing multinationals. Regardless of national systems of corporate governance,
family ownership remained predominant in alcoholic beverages, even though
certain managerial ¬rms (such as Diageo and SABMiller) had also become
very important as the industry consolidated internationally.
Initially ¬rms were entrepreneurial based, with ownership and corporate
control concentrated in the hands of a small group of investors, mainly fami-
lies. Over time, they developed distinct institutional arrangements, becoming
marketing based. Ownership tended to remain concentrated in the hands of
families or restricted groups of investors, but corporate control was now in
the hands of professional managers.
There are several reasons behind the apparent discordance between the
existing literature on national systems of corporate governance, and the
actual institutional arrangements that predominate in the global alcoholic
beverages industry by the beginning of the twenty-¬rst century. The ¬rst is
that those studies tend to make generalizations based on industries and ¬rms
that are dominant in their countries™ economies. The second is that different
industries require distinct corporate capabilities. It is the way in which these
capabilities evolved over time that leads ¬rms to adapt their ownership and
control structures.
In some industries such as automotive or pharmaceuticals, which are tech-
nology based, the predominant governance structures of ¬rms adapted by
having dispersed ownership of shares and managerial corporate control. In
other industries such as alcoholic beverages or cosmetics, where the distinc-
tive capabilities required by ¬rms are marketing knowledge, systems adapted
by keeping insider ownership structures but switched to managerial control.
This knowledge, which basically relates to the management of brands, has
two main parts. One part is “sticky” to the ¬rm, and is accumulated over
time. Another part is “smooth,” and may be acquired in the short run by
hiring professional managers. As a result of their growth, most ¬rms tended
to acquire “smooth” knowledge, and to switch from “personal” to “man-
agerial” control. In this process, it was possible for ownership to remain
concentrated due to the characteristics of the businesses, where the main
investments are in marketing and brand management, and where the cash

40 Naomi R. Lamoreaux, Insider Lending (Cambridge: Cambridge University Press, 1994): 9.
86 Global Brands
¬‚ows generated by the operations of ¬rms tend to be suf¬cient to cover those
investments.
This study shows that as we move from looking at countries to look-
ing at industries and ¬rms, the picture is quite different from that claimed
in the literature on the evolution of systems of corporate governance. In
global industries that are not dominant in any single country™s economy, the
predominant systems of corporate governance may evolve into a wide range
of combinations of ownership and control.
5

Channel Management




Introduction
Over time, the manner in which ¬nished products were handled and delivered
to the ¬nal consumers of alcoholic beverages changed substantially. While
in many countries the wholesaler was traditionally the major intermediary
between the producer and the retailer, in some countries other modes of dis-
tribution also developed. Interestingly, a number of direct competitors have
created distribution alliances. This chapter explores the rationale behind the
creation and evolution of different modes of distribution, and, in particu-
lar, of alliances that involved only producers, or producers and distributors
(mainly wholesalers). I analyze the different levels of commitment by multi-
nationals and provide an overview of the long-term patterns in the global
distribution of this product.


Differing Levels of Commitment, 1960s
During the 1960s, alcoholic beverages ¬rms from different parts of the
world had distinct forms of distribution. In the United States, wholesalers
were starting to concentrate at a regional or state level.1 In Europe, retail-
ers entered into direct marketing relationships with producers, increasingly
bypassing the wholesalers. As distributors grew in size and power, they
reduced the number of purchasing channels and suppliers lost bargaining
power. Many were at the mercy of retailers. The development of large-scale
supermarkets and hypermarkets made it very dif¬cult for beverage producers
to integrate vertically into retailing (either outright or through contractual
agreements) and remain competitive, as that would have entailed disposing
of brands and other kinds of products from competing ¬rms.2 Large retail-
ers, such as Sainsbury (in the United Kingdom) and Carrefour (in France)
sold a wide range of consumer goods, often under their own private label.

1 Richard S. Tedlow, New and Improved “ The Story of Mass Marketing in America (Oxford:
Heinemann, 1990); Erdener Kaynak (ed.), Trans-National Retailing (New York: Walter de
Gruyter, 1988); Luca Pellegrini and Srinivas K. Reddy (eds.), Retail and Marketing Channels
(London: Routledge, 1989).
2 Pellegrini and Reddy (eds.), Retail and Marketing Channels:18.

87
88 Global Brands
Cooperating directly with the producers, they eliminated the role of the
wholesaler and sold products at great discount.3
This was also a period when in the United Kingdom there was a widespread
growth of specialist outlets, such as “off-licenses,” that sold only alcoholic
beverages. The end of resale price maintenance and the development of large
retail units such as supermarkets in the 1960s meant that distributors could
now cover wider regions and even neighboring countries.4
The structure of distribution was also greatly affected by the use of new
systems of storage, stockholding, and warehousing, including the develop-
ment of stainless steel casks and aluminum containers. These developments
allowed better cleaning and refrigeration, which, with the advent of pasteur-
ization, helped stabilize the products in containers and reduce maintenance.
Major changes in logistics, especially with the growth of large-scale comput-
erized warehouses, allowed ¬rms to centralize their stockholding and obtain
economies of scale.5
While alliances between direct competitors in alcoholic beverages had
developed in the United Kingdom and the United States since Prohibition in
the 1930s,6 in Europe they were still very unusual as late as the 1960s. These
alliances provided economies of scale and scope in distribution and also min-
imized risk and the uncertainty related to their operations in foreign markets.
In other parts of the world such as Japan, however, the industries were still
very closed to foreign direct investment and alliances were nonexistent.


United Kingdom “ Development of Alliances
Between Competitors
The United Kingdom was the ¬rst to develop distribution alliances between
direct competitors. The growth of incomes in the 1960s was accompanied
by a liberalization of prices. Changes in the taxation of alcoholic bever-
ages and licensing laws allowed “off-licenses” to keep shops open for long

3 Peter Jones, Colin Clarke-Hill, David Hillier, and Peter Shears, “A Case of Bargain Booze,”
British Food Journal, Vol. 103, No. 7 (2001): 453“59; Asa Briggs, Wine for Sale: Victoria Wine
and the Liquor Trade, 1860“1984 (London: Bastford, 1985): 160; Frederick F. Clairmonte
and John Cavanagh, Merchants of Drink: Transnational Control of World Beverages (Penang:
Third World Network, 1988): 176.
4 See Appendix 6 for a de¬nition of distributors, supermarkets, and hypermarkets.
5 Tim Unwin, The Wine and the Vine (London: Routledge, 1992): 546.
6 For example, Distillers from the United Kingdom formed several alliances in the United States
and Canada after Prohibition with companies such as National Distillers, and Distillers Cor-
poration (later renamed as Seagram). Ronald B. Weir, The History of Distillers Company
1877“1939 (Oxford: Oxford University Press, 1995): 261, 271“72, 277“78; Peter Foster,
Family Spirits: The Bacardi Saga (Toronto: MacFarlane Walter & Ross, 1990): 53; “Agree-
ment Between Distillers Company Ltd (in the United Kingdom) and Distillers Corporation
(in Canada),” 1927, Box 93 (Seagram Collection, Hagley Museum and Library).
89
Channel Management
hours and made it much easier for restaurants to obtain liquor licenses. The
growth and geographical dissemination of supermarkets and hypermarkets
that could sell wine all day at cheaper prices played a very important role
in stimulating wine consumption and in ending the elite image of wine. By
offering information about how to combine wine with food, these markets
encouraged an increasing number of consumers in this traditionally beer and
spirit drinking nation to acquire wine drinking habits.7 Internationally, most
¬rms relied on third parties to distribute their beverages.
Distillers Company ranked as the largest alcoholic beverages ¬rm in sales
volume until the late 1960s. Foreign sales accounted for nearly 50 percent
of its activity, with whisky and gin corresponding to around 80 percent of
total sales (the remaining 20 percent came from chemicals and plastics).
Essentially, its strategy was based on high volume, low prices, and interna-
tional distribution. Its foreign direct investment, in France, Australia, Brazil,
Canada, the United States, and South Africa, was primarily in the production
of gin. Distillers Company also had a few investments in distribution. For
example, in 1958 it opened a warehouse and bottling unit in New Zealand
in order to overcome progressive restrictions to imports. In 1967 it acquired
a 70 percent share in the French company Simon Fr` res Ltd., which had
e
been the distributor for Johnnie Walker & Sons, Ltd. In the domestic mar-
ket, Distillers Company acquired several old scotch whisky brokers such as
Messrs. Ross & Coulter Ltd. in 1954. Nonetheless, for most of its sales in
foreign markets, Distillers Company relied on independent distributors.8
Allied Breweries was the second largest British ¬rm in sales volume during
the 1960s. It concentrated on the production of beer for the domestic market
and the sale of alcoholic beverages through a vast chain of licensed houses
(pubs and inns). It owned Victoria Wine, a large wine and spirits retail chain,
which had been taken over by Ind. Coope in 1959.9 The acquisition in 1966
of Showerings, a leading British wine merchant (which had itself acquired
Harveys), allowed Allied Breweries to gain control of the distribution of
its beverages in the United Kingdom. It was also able to acquire marketing
knowledge about the domestic market, and the management of brands in
the wines and spirits business in general.
Guinness was the third largest ¬rm operating in the United Kingdom dur-
ing this period. The ¬rm had some production operations in developing
countries of the Commonwealth. In Nigeria, for instance, Guinness had a
joint venture operation with the Anglo-Dutch consumer goods multinational

7 Unwin, The Wine and the Vine: 341.
8 The Distillers Company Limited, Annual Report and Accounts (1961“1970); D.C.L. Gazette,
Winter (1967): 209“11; Ronald Weir, “List of DCL Acquisitions, 1940“1986” (York, 1999,
mimeo).
9 Allied Breweries, Annual Reports and Accounts (1961); Briggs, Wine for Sale: 134.
90 Global Brands
Unilever,10 allowing the company to acquire market knowledge about the
operation without incurring much risk.
The other large British ¬rms, Bass, Scottish & Newcastle, Truman, Wat-
ney Mann, and Whitbread, were all in the brewing business. They tended
to have essentially national coverage, distributing either through their own
subsidiary companies, independent bottling companies (traditionally very
important in the case of beer), or through reciprocal trade agreements that
provided economies of scale and scope in distribution. Reciprocal trade
agreements had the additional advantage of giving the owners of the brands
national coverage without the costs of acquiring and managing a national
chain of public houses. Whitbread was the leading practitioner of this strat-
egy. The ¬rm frequently cemented these arrangements by acquiring a small
but signi¬cant share of the trading partner™s equity.11


North America “ Domestic Distributors of Foreign Brands
In the United States, the ¬rms that had developed very rapidly during this
period were Anheuser-Busch, Seagram, Hiram Walker, Heublein, Schenley
and National Distillers.12 National Distillers, the leading alcoholic beverages
¬rm in the 1960s, had an important import business that involved several
alliances (long-term distribution agreements) with producers of successful
brands from Europe, Canada, and the Caribbean. These alliances were in
part the result of attempts by foreign ¬rms to overcome the high barriers
to entry imposed by the U.S. three-tier distribution system. This system had
emerged after Prohibition to prevent a repeat of the pre-Prohibition “tied
house evils,” where saloons were controlled by distillers and brewers and
therefore had a vested interest in encouraging excessive drinking. The new
distribution system, which still applies today, did not permit ¬rms to be
vertically integrated, and required that the channels used for the distribu-
tion of alcoholic beverages be distinct according to their type and level of
alcohol content. Thus, beer is usually distributed through different chan-
nels from wines and spirits. Under the three-tier system producers are pro-
hibited from shipping directly to retailers and consumers, as goods have
to pass physically through the hands of at least one intermediary. A simi-
lar situation applies to importers of alcoholic beverages, where wholesaling

10 Terry Gourvish and Richard G. Wilson, The British Brewing Industry, 1830“1980 (Cam-
bridge: Cambridge University Press, 1994): 453; S. R. Dennisson and Oliver MacDonagh,
Guinness 1886“1939: From Incorporation to the Second World War (Dublin: Cork Univer-
sity Press, 1998): 202.
11 K. H. Hawkins and C. L. Pass, The Brewing Industry (London: Heinemann, 1979): 53, 57.
See, e.g., Whitbread, Annual Reports and Accounts (1961): 8“9.
12 Although Seagram and Hiram Walker were two Canadian ¬rms, they were running most of
their operations in the United States.
91
Channel Management
(usually handled at a state level) and retailing have to be performed by distinct
parties.13
During this period, National Distillers started to consolidate its position in
the domestic market by acquiring distributors/importers. In 1962, it bought
Peel Richards Ltd., a company that for the previous twelve years had been
responsible for the distribution in New York of most of National Distillers™
brands. In 1963, it acquired another wholesaler, Munson G. Shaw Co. Inc.,
which added well-accepted imported brands such as Cockburn port, Cossart-
Gordon madeira, Bertolli Italian wines, and the wines of Baron Philippe de
Rothschild to its portfolio. The acquisition also expanded the sales force of
the ¬rm.14
Schenley was another ¬rm with a very large imports business and sales
organization in the United States. Brands imported by Schenley through
distribution agreements included Dewar™s Scotch whisky and Mateus Ros´ , e
a Portuguese wine. Although it also owned sales organizations in foreign
markets, by 1970 exports only accounted for 6.6 percent of total sales (dis-
tributed essentially through sales organizations in eleven countries).15
Heublein developed very rapidly in the 1960s, mainly as a result of its
activity in the domestic market, and, in particular, the success achieved with
the vodka brand Smirnoff. It also distributed other beverages as a result of
the alliances formed with leading European beverage brand owners. These
included Harveys Bristol Cream (beginning in 1959), Lancers Vin Rose, a
Portuguese wine (that became the largest selling imported wine of its type in
the United States from 1965), Gilbey™s Black Velvet, and McMaster Canadian
whiskeys (from 1967). Several of the alliances involved reciprocal trans-
actions. For example, Heublein™s alliance with Gilbey™s also involved the
license for the latter to produce Smirnoff vodka in several markets includ-
ing Ireland and the United Kingdom. As a result of its successful growth,
Heublein appeared among the top 500 U.S. corporations in 1966 for the
¬rst time. In 1967, it became the largest U.S. importer of wines and spirits,
with a sales force spread throughout the country.16 Heublein acquired sev-
eral distributors and formed strategic alliances in production. For example,
in 1968 it acquired Don Q Imports Inc., the U.S. distributor of Puerto Rico™s
leading brand of rum. In the same year, it also formed an alliance with Jos´ e
Maria da Fonseca in Portugal to set up a plant that would more than double
production of Lancers.17
The changing nature of alliances and the role and level of control of each
partner over the management of the brand also led to the termination of some

13 Brian Newkirk and Rob Atkinson, “Buying Wine Online “ Rethinking the 21st Amendment
for the 21st Century,” Policy Report (January 2003).
14 National Distillers, Annual Reports and Accounts (1962, 1963).
15 Schenley, Annual Report and Accounts (1963, 1970).
16 17 Ibid.
Heublein, Annual Report and Accounts (various years).
92 Global Brands
alliances. For instance, in 1965, Heublein terminated the arrangements with
Guinness and L. Rose & Co. for Guinness Stout and Rose™s Lime Juice,
respectively. This was the result of disagreements concerning the marketing
of the brands in the United States, where the producing ¬rms wanted to have
more control over their management than Heublein would allow.18
U.S. ¬rms tended to use different modes for distributing their own and
their partner™s brands. Brown Forman, for instance, used a separate sales
force to distribute imported brands. Some of the imported brands they dis-
tributed during the 1960s were Veuve Cliquot champagne; Usher™s Green
Stripe Scotch; Bols liqueurs, brandies, and gin; Ambassador Scotch; Old
Bushmills Irish whiskey, and Martell.19
There were two leading Canadian alcoholic beverages ¬rms, Hiram
Walker and Seagram, that had an important position in the U.S. market
during this period. They had their own distribution channels that dealt with
the wholesalers in various regions of the country. Apart from selling in the
United States, Seagram was doing business in several other parts of the world.
It had invested in wholly owned operations in countries such as Belgium,
Argentina, and Venezuela.20


Continental Europe “ In Strategic Markets
There were many differences between the development of ¬rms in Conti-
nental European countries and those from the United States and the United
Kingdom. Leading Continental European ¬rms, for instance, tended to have
a high ratio of exports to total sales, whereas only a few Anglo-Saxon ¬rms
(such as Distillers Company) generated a signi¬cant part of their sales in for-
eign markets.21 Continental European ¬rms were smaller in size, remained
family owned and controlled, and were part of distribution systems that were
much more fragmented.22 The ¬rms tended to concentrate their sales among
a smaller number of distributors. In some cases, they integrated vertically into
distribution in their domestic markets and in foreign markets considered to

18 Heublein, Annual Reports and Accounts (1965).
19 Brown Forman, Annual Reports and Accounts (various years); William F. Lucas, “Nothing
Better in the Market: Brown Forman™s Century of Quality 1870“1970,” The Newcomen
Society in North America (New York: Newcomen Society, 1970).
20 Distillers Corporation “ Seagrams Ltd., Annual Report and Accounts (1960“1970); Seagram,
Annual Report and Accounts (1984); Samuel Bronfman, “ . . . From Little Acorns . . . ,” in
Distillers Corporation “ Seagrams Ltd., Annual Report and Accounts (1970): 73.
21 For example, Mo¨ t & Chandon in 1960 exported 49 percent of its sales, to 113 countries
e
(Mo¨ t & Chandon, Annual Report and Accounts, 1960). Distillers Corporation, Annual
e
Report and Accounts (1961“1970).
22 James B. Jeffreys and Derek Knee, Retailing in Europe: Present Structure and Future Trends
(London: Macmillan, 1962): 25. For example, in 1955 the number of retail establishments
in France was 755,000 and in the United Kingdom was 596,000, corresponding respectively
to 57 and 86 inhabitants per retail establishment.
93
Channel Management
be strategic.23 For example, in 1957, Mo¨ t & Chandon acquired its former
e
agent in the United Kingdom, Simon Brothers & Co. At the time of this
acquisition, that market represented 20 percent of the ¬rm™s total exports,
in a period when the total number of markets of destination for their cham-
pagne was 108.24
It is also during the 1960s that continental European ¬rms started to
form distribution alliances with direct competitors, sometimes from dif-
ferent countries, at other times from the same country. For example, by
1967, Mo¨ t & Chandon had alliances with leading alcoholic beverages ¬rms
e
such as Heineken and Distillers Company. Through its sales organization,
S.A. France Champagne, which consisted of eighteen agents each working a
given region and using subagents and representatives, Mo¨ t & Chandon sold
e
Heineken beer, J&B Rare Scotch whisky, and the liqueur Erven Lucas Bols.25
Another example was the alliance created in 1967 between the French ¬rms
Cointreau, Izarra, and R´ my Martin, which were direct competitors. This
e
alliance led to the creation of Rivi` re Distribution, which sold their products
e
in both the domestic and foreign markets, particularly in Europe.26
In the Netherlands, there were several ¬rms that by the 1960s were already
highly internationalized. Heineken, for example, had been managed in the
United States by a distributor, Van Munching, since 1945. In 1992, Heineken
acquired this distributor, as Van Munching™s heirs had no interest in the busi-
ness and also because this family ¬rm was losing its competitiveness given the
changes that were occurring in distribution. In the United States, this brand
was positioned as a beer for special occasions, consumed primarily in the
on-premise market. This helps explain why Heineken™s management decided
never to produce locally.27 The ¬rm formed alliances in other foreign markets
to overcome entry barriers such as high import duties and high transportation
costs associated with large geographical distances. In the United Kingdom,
Heineken had an alliance with Whitbread during the 1960s. This permit-
ted the ¬rm to penetrate that market in a period in which distribution was
dominated by breweries, most of whom owned pubs.28 The early alliances
established by Heineken in France to overcome the ceilings established by the


23 See, e.g., Philippe Roudi´ , “Une V´ n´ rable Entreprise Bordelaise de Liqueurs Marie Brizard
e ee
et Roger,” in Alain Huetz de Lemps and Philippe Roudi´ (eds.), Eaux-de-Vie et Spiritueux
e
(Paris: Centre National de Recherche Scienti¬que, 1985): 295“300.
24 25 Ibid (1968).
Mo¨ t & Chandon, Annual Report and Accounts (1957).
e
26 Jacques Jeanneau, “La Soci´ t´ Cointreau, D™Angers au March´ Mondial,” and Pierre
ee e
Laborde, “La Soci´ t´ Izarra de Bayonne,” both in Huetz de Lemps and Roudi´ , Eaux-de-Vie:
ee e
307“20.
27 Interview with Jan Beijerinck, former Worldwide Client Services Director of Heineken,
Utrecht, 10 March 2001.
28 Heineken, Annual Reports and Accounts (1960“1961); M. G. P. A. Jacobs and W. H. G.
Mass, Heineken History (Amsterdam: De Bussy Ellerman Harms bv., 1992): 264, 270,
287“88.
94 Global Brands
local government on imports of alcoholic beverages after World War II were
not very successful. It was only after 1961 when Heineken appointed Mo¨ t e
& Chandon as its general importer that sales started to develop. Mo¨ t & e
Chandon had a powerful, countrywide, modern sales organization and dis-
tribution network and could obtain a higher level of economies of scale and
scope in distribution, selling different brands.29 This alliance ended in 1972,
when Heineken acquired a majority interest in Alsaci` ne de Brasserie group
e
30
¨
(Albra), owner of the Mutzig beer brand. De Kuyper, another Dutch ¬rm
that before the War had been relatively large and had an internationalized
Genever and liqueurs business, did not pro¬t as much as it could have from
the economic prosperity of the 1960s. However, it kept its old established
alliances with companies such as National Distillers.31
The Danish brewers, Carlsberg and Tuborg, held almost a monopoly posi-
tion in Denmark in the 1960s. There, they used exclusive distributors, which
were only allowed to carry brands of a single brewery. In foreign markets
they used local distributors, which handled competing beer brands. These
distinct strategies matched the different objectives the ¬rm had for each mar-
ket. In Denmark, the aim was to remain a leader in the industry and extract
as much rent as possible, drawing to a great extent on their reputation and
image; in foreign markets, minimization of risk and uncertainty were the
main concerns. Often, the distributors in foreign markets were leading local
alcoholic beverages ¬rms. For instance, in the United Kingdom, Carlsberg
had a distribution agreement with Grand Metropolitan that enabled its beers
to be distributed through the latter™s pubs and retail outlets.32


Japan “ Focus on Distribution of Domestic Beverages
Distribution in Japan also evolved after World War II, following the changing
habits of consumption and the transformation in the structure of the alco-
holic beverages industry. Rapid economic growth and remarkable economic
changes took place. Concentration within both the manufacturing and retail-
ing sides forced changes in the traditional distribution structure.33 Although
Asahi Breweries, Kirin, Sapporo, and Suntory developed during this period,

29 Interview with Jan Beijerink, former Worldwide Client Services Director, Utrecht, 10 March
2001; Jacobs and Mass, Heineken: 294“96.
30 Heineken, Annual Reports and Accounts (1972“1973).
31 K. E. Sluyterman and H. H. Vleesenbeek, Three Centuries of De Kuyper 1695“1995
(Shiedam: Prepress Center Assen, 1995): 69, 79.
32 Hans Chr. Johansen, “Marketing and Competition in Danish Brewing,” in Geoffrey Jones
and Nicholas J. Morgan (eds.), Adding Value: Brands and Marketing in Food and Drink
(London: Routledge, 1994): 126“38.
33 Kazutoshi Maeda, “The Evolution of Retailing Industries in Japan,” in Akio Okochi and
Koichi Shimokawa (eds.), Development of Mass Marketing: The International Conference
on Business History, 7 (1981): 265“89.
95
Channel Management
Japanese ¬rms had no incentive to form alliances with leading multinationals
of alcoholic beverages to distribute their brands locally because there was
still no demand for international brands, and consumption of alcohol was
growing slowly.
By 1954, Kirin had become an industry leader, ranking among the largest
breweries in the world. Kirin™s activities focused essentially on the Japanese
brewing and soft drinks businesses, but it also had some international activ-
ity, the United States being its most important market. In Japan, Kirin dis-
tributed its beer through retailing companies, and also through wholly owned
vending machines.34 Distribution to foreign markets was essentially carried
out by subsidiary companies engaged in various transport operations.
Suntory™s spirits business took off after the War, because Suntory whisky
had built a reputation in the U.S. military of¬cers™ clubs in Japan. The success
with the military forces gave the brand an association with the American
lifestyle. The ¬rm took advantage of this association in aggressive marketing
campaigns. Distribution of Suntory whisky spread to every part of Japan.
The launching of Suntory bars in 1955 played an important part in this
subsequent success. In 1963, Suntory expanded into brewing, using its wide
distribution network in the food and drinks industries to sell a variety of
products.35

Globalization, Vertical Integration, and Joint Venture
Formation, 1970s“1980s
The convergence of strategies in the 1970s and 1980s did not entail only
mergers and acquisitions of ¬rms that owned brands with the potential to
become global. In the 1970s, it also involved ¬rms that owned distribution
channels. And in the 1980s, it involved the creation of distribution joint
ventures between leading multinationals to cover multiple markets, includ-
ing Japan. The trend toward concentration in the off-premise market by
supermarkets and hypermarkets created a danger of cartel formation.36

Vertical Integration
In the United Kingdom, where brewers had traditionally been vertically inte-
grated, there was a trend toward disintegration, with brewers concentrating
on production and branding. On the other hand, there was a tendency for

34 Kirin, Annual Report and Accounts (1954, and 1961“1969).
35 Interview with Kunimasa Himeno, Manager International Division of Suntory and Yoshi
Kunimoto, Executive Vice President of Suntory-Allied, Tokyo, 16 September 1999; Christo-
pher Fielden, A Drink Dynasty: The Suntory Story (Throwbride: Wine Source, 1996): 29“32.
36 Indeed, some believe that has already happened in the United Kingdom, where Tesco, Sains-
bury™s, Asda and Safeway are often accused of bringing down the prices paid to suppliers
without passing these savings onto consumers.
96 Global Brands
wines and spirits ¬rms to create wholly owned distribution channels. There
was a perception that leading ¬rms needed to control strong distribution net-
works either on their own or through alliances. Ownership of distribution
also aimed at overcoming the problem of the so-called parallel distribution,
whereby brands were sold in the same market at different prices. Networks
would enable ¬rms to increase control over the marketing of their brands
and acquire thorough knowledge about their markets. Simultaneously, they
would minimize the risk of distributors or agents switching to a competi-
tor. Firms, therefore, needed to have sales and marketing teams as well as
wide portfolios of brands in order to be able to control their marketing.
This vertical integration into distribution took the form of acquisitions of
small family-owned distributors. These distributors usually had carried the
acquiring ¬rm™s brands and thus had played an important role in the creation
of their success in particular markets. In this period of high competition, the
acquisitions prevented competitors from taking over these distributors and
acquiring the valuable knowledge they possessed.
In the United States, despite the three-tier system of distribution, a simi-
lar trend toward vertical integration emerged. Leading multinationals from
Europe acquired their former importers, which until then had acted as sales
agents handling the local advertising and marketing of the brands. This
activity, which only required a small number of employees, provided very
good margins. These margins were much higher in the United States than in
Europe, where importers and distributors were usually the same entity and
distribution was much more fragmented.37
Beginning in the 1980s when demand started to stagnate in the West, the
largest multinationals of alcoholic beverages made their biggest investments
in distribution. Only a few of the leading North American ¬rms survived
independently. Seagram, which had invested strongly in Europe and also in
the Far East, was one of the few exceptions. Many North American ¬rms with
strong distribution channels in the domestic market were acquired by Euro-
pean ¬rms. As previously mentioned, the major acquisitions by European
¬rms in North America were Hiram Walker by Allied Lyons in 1986, Schen-
ley by Guinness in 1987, Liggett & Myers by Grand Metropolitan in 1982,
and Heublein also by Grand Metropolitan in 1987. The family-controlled
U.S. ¬rms of Anheuser-Busch, Brown Forman, and Coors remained inde-
pendent, while Miller Brewing fell under the control of the acquisitive U.S.
tobacco company Philip Morris.38
Grand Metropolitan™s acquisition of Liggett brought with it two impor-
tant marketing companies in the United States, Carillon Importers and

37 Interview with James Espey, former Chairman of Seagram Distillers PLC, former Managing
Director of United Distillers/Guinness Plc and former Chairman of IDV-UK, Wimbledon, 4
September 2000.
38 Miller was subsequently acquired by South African Breweries in 2001.
97
Channel Management
Paddington Corp. Paddington Corp. had helped develop the U.S. market
for European spirits, in particular whisky brands such as Grand Metropoli-
tan™s J&B Rare.39 The hostile acquisition of Liggett & Myers allowed Grand
Metropolitan to gain control of the management of its own brand, J&B
Rare, in a market where it was very successful. In the early 1980s, Grand
Metropolitan™s own distribution channels were selling about half of the total
volume of their major brands.
There are other examples of European multinationals that during this
period integrated vertically by acquiring North American distribution com-
panies. Pernod Ricard™s acquisition of Austin Nichols in 1980, which was
also part of Liggett & Myers, is one case. Austin Nichols owned the bourbon
Wild Turkey, and was a major importer of European brands such as Baileys
Irish Cream and Campari, as well as French wines. Initially, the manage-
ment of Austin Nichols resisted being acquired by Pernod Ricard, and as
a gesture of goodwill, the two parties agreed that Austin Nichols would
be the importer of Pernod in the United States. When the threat of a hos-
tile takeover by Grand Metropolitan appeared, however, the management
of the U.S. distribution ¬rm ¬nally agreed to sell the business to Pernod
Ricard.40 Another case is Schieffelin & Co., which had been the U.S. distrib-
utor for Mo¨ t Hennessy since 1945 and was acquired by Mo¨ t Hennessy in
e e
1980. At the time of the acquisition, Schieffelin accounted for roughly 40
percent of Hennessy™s and close to 15 percent of the champagne Mo¨ t & e
41
Chandon™s sales.
Guinness only started to integrate vertically very late, after the acquisition
of Distillers Company in 1986. The company chose vertical integration to
create new wholly owned distribution channels. But there were problems
with that strategy as it could lead to parallel pricing. Selling brands through
existing distribution channels as well as through new wholly owned channels
might imply that the beverages would reach the ¬nal customer at higher
prices than if the distribution was controlled by the ¬rm. This concern led
Guinness to buy out its own distributors. Guinness™s sales through its own
distribution channels moved from 25 percent at the outset of 1987 to more
than 70 percent in 1988.42 The ownership of distribution channels gave
Guinness a margin available on distribution as well as critical control of the
marketing activity of its brands at the local level.

39 Interview with Sir John Bull, former Chairman of Grand Metropolitan and Diageo, London,
19 November 2003; interview with Tim Ambler, former consultant for Grand Metropolitan,
London, 12 July 2000.
40 Interview with Thierry Jacquillat, former CEO of Pernod Ricard, London, 20 January 2004,
“The Austin Nichols Story,” Accession 2126, Box 773, Seagram Collection.
41 Mo¨ t & Chandon, Annual Report and Accounts (31 July 1946); Mo¨ t Hennessy, Annual
e e
Report and Accounts (1980).
42 James Capel & Co. Ltd., “United Distillers Group™” (November 1988): 1, 25; Distillers
Archive, Diageo.
98 Global Brands
With the move to wholly owned distribution, Guinness rationalized its
list of third-party distributors. It went from a situation of 1,304 agents/
distributors worldwide and zero owned distribution channels in 1986, to
a situation of 470 agents and thirty-seven owned distribution channels
in 1989. However, despite this move into distribution, Guinness often used
different channels to serve the same market. In Asia, for example, the ¬rm had
two joint venture alliances with distinct partners and also used wholly owned
distribution channels, with each arrangement covering different regions
and brands.43
Seagram followed a similar strategy. In order to gain control over the dis-
tribution of its beverages, Seagram acquired several leading family ¬rms that,
apart from having well-established channels of distribution in certain mar-
kets, also owned successful brands. One illustration is Sandeman, a leading
port and sherry producer with a very strong distribution network in Europe,
which was acquired in 1979. Another is Martell, acquired in 1989, which
apart from the very successful cognac brand also owned a strong distribu-
tion network in France.44 In 1984, Seagram integrated into retailing in the
United Kingdom and France by acquiring two retail outlet chains, Odd-
bins Limited, and Gough Brothers from Scottish & Newcastle brewery.45
Specialist distributors such as Oddbins played an important role in wean-
ing middle-class British customers off consumption of French and German
wines by introducing them to new world and other wines.46
Despite its investments in foreign markets, Seagram still relied extensively
on third-party distributors in the 1970s and 1980s. Important independent
channels of distribution were the worldwide Duty Free Stores and other
special markets such as military bases, where the ¬rm was one of the largest
suppliers in the industry. In the United States, Seagram had several sales
organizations (each one specializing in a particular set of brands and markets)
and a network of independent distributors most of which had been working
for the ¬rm since the repeal of Prohibition or at least since World War II.47
Some Japanese ¬rms also established wholly owned subsidiaries in the U.S.
market during this period. One example was Kirin, which formed “The
Cherry Company Ltd.” in 1981, the sole importer of Kirin beer in the United
States.

43 Ibid.
44 Interview with George Sandeman, Chairman of the House of Sandeman, and Managing
Director of Seagram Iberia, Oporto, 19 January, 2000; Teresa da Silva Lopes, “A Evolucao
¸˜
¸˜
das Estruturas Internacionais de Comercializacao de Vinho do Porto no s´ culo XX,” Revista
e
de Historia Economica e Social, S´ rie 2, No.1 (2001): 91“132; Seagram, Annual Report and
´ ´ e
Accounts (1980, 1989).
45 Seagram, Annual Report and Accounts (1985).
46 In 2001 Seagram disposed of its drinks business, and in 2002 Oddbins was sold to Castle
Fr` res, a giant French wholesaler, wine merchant, and wine producer.
e
47 Seagram, Annual Report and Accounts (1972).
99
Channel Management

Spread of Joint Ventures With Local Partners
During the 1970s and 1980s, alliances were also an important mode through
which ¬rms distributed their brands worldwide. Leading multinationals con-
tinued to form alliances with local partners, but there was also a proliferation
of alliances between direct competitors “ such as large multinationals and
leading local partners in emerging markets of Asia and South America “ to
produce, bottle, and distribute locally.
During this period, consumption of alcoholic beverages increased in Japan.
Japanese ¬rms formed multiple alliances with ¬rms from other countries both
to produce beer and spirits locally and also to sell imported beverages locally.
These alliances provided Japanese ¬rms with the opportunity to acquire
marketing knowledge while simultaneously obtaining economies of scale
and scope in various levels of activity, especially distribution. Foreign ¬rms
were able to minimize the risk associated with entering such a different
cultural and geographical market as Japan. Japanese ¬rms also started to
internationalize their alcoholic beverages businesses (mainly through exports
and licensing agreements), but overall they remained essentially domestic
¬rms up until the beginning of the twenty-¬rst century.
Kirin and Seagram formed one of the ¬rst joint ventures in Asia in 1972,
at a time when ¬rms wanted to consolidate their positions in their domes-
tic markets and maximize market share. Kirin owned 50 percent, Seagram
45 percent, and Chivas Brothers 5, and they planned to produce whiskey at
the foot of Mount Fuji, famous for its fresh water and clean air. Seagram
brought its manufacturing techniques and Kirin its sales network and mar-
ket knowledge. Apart from selling locally produced whiskey, Kirin also sold
the Seagram™s brands Robert Brown, Dunbar, Emblem, Burnett™s Gin, and
Nikolai vodka in Japan. This joint venture, which relied on both a long-term
relationship based on a contract and on mutual trust, became part of Dia-
geo, after its joint acquisition with Pernod Ricard of the Seagram alcoholic
beverages business worldwide in 2002.48
Suntory also began to form alliances with Western partners in the 1970s.
Its ¬rst distribution agreement was with Brown Forman in 1970. In the
1980s, it allied with other leading multinationals, including Seagram for
the local distribution of the brand Martell, and Guinness for the brand Haig
Scotch. However, the instability associated with these short-term agreements
led to the creation of a joint venture with Allied Lyons in 1988. For Sun-
tory, this new alliance brought the reputation and pro¬tability of a leading
British multinational in a period when demand for imported brands was

48 Kirin, Annual Report and Accounts (1973), 13; Seagram, Annual Report and Accounts
(1972). For a detailed analysis of different modes of coordinating international activities by
multinationals, and for a discussion of long-term close trading relationships based on trust
in Japan see also Mari Sako, Prices, Quality and Trust “ Inter-Firm Relationships in Britain
and Japan (Cambridge: Cambridge University Press, 1992).
100 Global Brands
starting to expand very rapidly. In addition, Suntory gained access to a wide
distribution network of alcoholic beverages in Europe and North America.
For Allied Lyons, it provided the opportunity to enter a new market where
consumption of alcoholic beverages was growing very fast. Suntory took
Allied Lyons™s brands in Japan, and Hiram Walker took Suntory™s brands
in the United States. This agreement also involved cross-shareholding, with
Suntory investing £89 million to acquire 2.5 percent of Allied™s share capi-
tal and Allied Lyons investing £28 million to become Suntory™s ¬rst outside
shareholder.49
The joint venture formed in 1986 between Guinness and Mo¨ t-Hennessy
e
was a landmark agreement. It covered multiple markets including Japan,
Hong Kong, Singapore, Malaysia, Thailand, France, Eire, and the United
States. This global alliance between two leading multinationals became a
model in the industry owing to the bene¬ts it brought to the ¬rms involved.
It permitted a careful marketing of limited volumes of deluxe and premium
quality brands for high margins and status positioning. The complementarity
of these brands in terms of their beverage type (champagne, cognac, scotch
whisky, and gin) permitted them to be sold in the same markets, using a
single sales force, with complete cost sharing, thereby obtaining economies of
scale and scope and ¬‚exibility in distribution. In the United States, this joint
venture brought together Schieffelin & Co. Importers (acquired by Mo¨ t e
Hennessy) and Somerset Importers (acquired by Guinness). Apart from the
cost bene¬ts, this alliance gave Mo¨ t Hennessy access to an operation with
e
superior marketing skills and Guinness/Distillers a very good sales team.
This alliance soon created a general trend in the industry, especially among
the leading multinationals, but the geographical focus on the Far East and
South American markets was quite clear. Brewers and spirits ¬rms with com-
plementary brands formed alliances, relying essentially on distinct position-
ing of the beers in the markets. Some examples are the alliances formed in the
1980s between Kirin and Heineken (to market Dutch beer in Japan), Asahi
¨ ¨
Breweries and Lowenbrau (to market its specialty German beer in Japan),
Pripps of Sweden and Asahi (to bottle and sell Swedish beer in Japan), Sap-
poro and both Miller Brewing and Guinness, and also Suntory and Anheuser-
Busch (to produce Budweiser beer in Japan). In the spirits business, examples
of alliances in emerging markets during this period include Guinness and
Jinro in South Korea, Seagram and China Distillery Shanghai in China, and
the joint venture involving Mo¨ t-Hennessy, Cinzano, Monteiro, and Aranha
e
in Brazil.

49 Interview with Mr. Kozo Chiji, Manager of the Corporate Planning Department of Suntory,
Tokyo, 16 September 1999; Interview with Yoshi Kunimoto, Executive Vice President of
Suntory-Allied, and Kunimasa Himeno, Manager of the International Division of Suntory,
both in Tokyo, 16 September 1999; Allied Lyons, Annual Report and Accounts (1989): 8;
Canadean Ltd, “Suntory” (Hants, 1999).
101
Channel Management
In Europe, there was a proliferation of alliances in distribution between
leading multinationals. They also aimed at developing greater control over
distribution and obtaining economies that would otherwise not have been
available. For example, Guinness created joint ventures with Bacardi (in
Spain and Germany), Uderberg (in Germany), Codorniu (in Spain), Boutari
(in Greece), and Real Companhia Velha (in Portugal). By allying its sub-
stantial scotch whisky and gin interests with those of major local operators,
scale and scope economies were achieved much more quickly than would
have otherwise been the case.50
In 1988, Grand Metropolitan also had joint ventures in France, Germany,
the Benelux, Japan, and Australia. Allied Lyons and Whitbread had created a
jointly owned European Worldwide Cellars in 1985, with the aim of merging
the wine interests of both companies in the United Kingdom and worldwide,
and Brown Forman formed an agreement with Martell for the distribution of
Jack Daniels in France. The arrangement in 1984 between American Brands
and Grand Metropolitan, whereby the latter undertook the selling of Whyte
& Mackay whiskies in the “on-license” trade and in “cash-and-carry” outlets
in the United Kingdom, is yet another illustration.51


Alliances Between Competing Partners From the 1990s
In the beginning of the 1990s, consolidation in the retailing and wholesal-
ing businesses accelerated. Large retailers and wholesalers demanded wider
portfolios of brands of wines, spirits, and beer from different geographical
origins, sourced from just a few distribution networks at low costs and in
short periods of time.52
In countries such as the United States where each state has its own laws
about distribution, labeling, packaging, and retailing, there were very few
companies with national distribution channels. The distribution of wines,
spirits, and beer is still dominated by family businesses such as Southern Wine
& Spirits of America, Charmer-Sunbelt Group, and National Distributing.
These wholesalers have regional coverage (either one- or multistate) and
often handle competing brands.
By the beginning of the twenty-¬rst century, there were very few wines
and spirits ¬rms integrated vertically into retailing. LVMH is one of the
few leading multinationals vertically integrated into retailing through its
international chain Duty Free Stores, aimed especially at tourists. The con-
centration by ¬rms on one activity in the value added chain rather than in
50 James Capel, “United Distillers Group”: 25, Distillers Archive, Diageo.
51 Grand Metropolitan, Annual Reports and Accounts (various years).
52 ¸˜ ¸˜
Teresa da Silva Lopes, Internacionalizacao e Concentracao no Vinho do Porto, 1945“1995
(Porto: GEHVID/ICEP, 1998); Nirmalya Kumar, “The Power of Trust in Manufacturer-
Retailer Relationships™” in Harvard Business Review on Managing the Value Chain (Boston,
Mass: Harvard Business School Press, 2000): 92“93.
102 Global Brands
vertical integration into retailing is due in part to the very high overhead
costs of distribution and the distinct kinds of capabilities necessary to run
such businesses.
E-commerce also started to develop at the beginning of the twenty-¬rst
century, even though this channel still did not account for a substantial
amount of trade in alcoholic beverages.53 Nevertheless, this activity was
expanding as companies and retailers sought to increase the scope of their
businesses, transmit the imagery of their brands to their customers, and
position themselves strategically for future modes of competition.
During this period, the way business was conducted had changed, and
logistics in distribution had gained increasing importance as ¬rms had to
manage larger stocks of different brands and deal with larger retailers. Tech-
nology had improved logistics, integrating information systems to provide
quicker and easier mechanisms for decision taking within multinationals.
By the end of the 1990s, however, many ¬rms had failed to achieve their
aims and had begun to disintegrate vertically. Distribution operations often
involved high ¬xed costs, which were not always covered by the gains
obtained from sales. The maturation of product categories also led ¬rms to
disintegrate vertically.54 Many ¬rms had overexpanded, offering very wide
portfolios of brands. Their distribution channels caused them to face the
basic con¬‚ict of being both in the distribution business and in the manage-
ment of brands. The sale by Seagram of its wholly owned distribution chan-
nels in Austria, Scandinavia, and Australia (in, respectively, 1997, 1998, and
1999), for which alliances with local partners were subsequently substituted,
demonstrates this trend.
Leading multinationals such as Diageo, Allied Domecq, Seagram, Bacardi,
and Pernod-Ricard, which were traditionally vertically integrated, started to
rationalize their operations in existing markets, covering new regions with
interlocking alliances. One of the major aims in the formation of Diageo in
1997 was the creation of new ef¬ciencies at the distribution level. Putting
Guinness and Grand Metropolitan together produced a stronger brand port-
folio and provided £300“400 million ($495“660 million) in annual cost sav-
ings, some for the bottom line and some for investing in brands.55 After the
acquisition of Seagram™s brands in 2002, Diageo concentrated its distribution

53 For example, the Internet business accounted for 33.5 million pounds of Bass worldwide
sales, but that ¬gure was expected to treble in 2000 [Interview given by Sir Ian Prosser,
Bass™s chairman to Lucy Killgrem in “Bass put £5m into last minute,” Financial Times
(4 February 2000)].
54 Seagram, Annual Report and Accounts (1997, 1998, 1999). For a theoretical analysis of the
advantages and disadvantages of wholly owned distribution channels, see Erin Anderson
and Anne T. Coughlan, “International Market Entry and Expansion via Independent or
Integrated Channels of Distribution,” Journal of Marketing, Vol. 51 (1997): 72.
55 Interview with Jack Keenan, Chairman of Diageo, London, 3 June 2003; Diageo, Annual
Report and Accounts (1998).
103
Channel Management
on fewer wholesalers, with whom they formed long-term distribution agree-
ments.56
The creation of Maxxium in 1999, a distribution joint venture formed
between R´ my Cointreau, Highland Distillers, and Fortune Brands, and later
e
extended to Vin & Sprit, was in part a reaction to the creation of Diageo.
Maxxium sought to form a global distribution company for premium wines
and spirits that would operate in ¬fty markets outside the United States.
This move cemented some long-standing business relationships, and helped
the ¬rms to obtain lower costs and greater effectiveness and thus to grow
their brands into many major overseas markets where they did not have a
presence.
Alliances in distribution among smaller ¬rms owning different types of
alcoholic beverages brands (in wines, beer, or spirits) became even more
common than in previous decades. These alliances permitted smaller ¬rms
to explore the complementarities between the products of several companies.
They also allowed them to achieve strategic objectives (such as ¬lling country
and portfolio gaps in demand), to obtain a higher control of the marketing
of their brands, and to achieve economies of scale and scope in distribution.
Some examples are Brown Forman and the old Swedish Liquor Monopoly,
which owned the famous vodka brand Absolut produced by Vin & Sprit.
The close ties between the two family businesses Bacardi-Martini and
Brown Forman led to the creation of Gemini, a distribution alliance that
covered different markets. Through this alliance, Bacardi distributed Brown
Forman™s Jack Daniels whiskey in Europe where it had a strong distribution
network, especially after the acquisition of Martini Rossi in 1993. Prior to
that, the two family-controlled companies had also collaborated, partnering
in an on-premise distribution initiative.
By the beginning of the twenty-¬rst century, most of the top brewers still
relied on their home countries for the bulk of their sales. However, with trade
barriers falling within regions such as Europe and Latin America, and with
production synergies becoming possible, pressure increased for local giants
to expand outside their home markets. There had been a similar process in the
United States twenty years earlier when domestic consolidation occurred very
rapidly. These changes in the external environment suggested two possible
ways for brewers to develop. One was through mergers and acquisitions of
brewers in other countries. The other was through the creation of alliances.
For instance, after the late 1990s many companies entered the Asian market.
South African Breweries was the ¬rst multinational brewer to enter China,
taking a 49 percent stake in a joint venture with the second leading brewer in
China, China Resource Beverage Ltd. Anheuser-Busch followed by forming

56 While the cost of distributing a case of spirits is about $16.25, a box of soft drinks costs on
average $5.3 and beer $6.25. “Diageo aims to outstrip the market with its new distribution
model,” Impact International, Vol. 33, Nos. 3 and 4 (2003).
104 Global Brands
a joint venture with Tsingao Breweries, the leading Chinese brewer (with a
market share of around 10 percent).
Alliances also provided important learning experiences for wines ¬rms.
For example, the Californian wines producer Mondavi started making efforts
to become a global company in the 1970s through the 1990s by producing
California wines and selling them globally. But it was through a joint venture
with French wine maker Baron Philippe de Rothschild that Mondavi learned
what it took to succeed on a global scale. The managers of the ¬rm realized
that to be a global wine business they not only had to produce and sell wines
from California on a global basis, but they also had to produce wines in
many of the great countries of the world. They had to market those wines by
emphasizing a unique style and character that represented the cultures and
people from those places.57


Long-Term Patterns in Distribution
The alcoholic beverages industry thus changed decisively in the years fol-
lowing the 1960s, consumption was country-speci¬c, distribution was frag-
mented, competition was essentially domestic, and ¬rms had little marketing
knowledge about the management of brands and distribution or the markets
and their social and cultural speci¬cities. The most frequent modes through
which ¬rms distributed their beverages were those that minimized risk and
uncertainty. Their lack of experience in international distribution meant that
¬rms had no control over the logistics or the marketing of their brands and
therefore could hardly acquire any marketing knowledge about the perfor-
mance of their brands in different markets. Therefore, they tended to sell their
brands primarily through agents and distributors. These provided a quick
entry into markets and allowed ¬rms to economize on information costs
associated with the risk of exposure to uncertainty.58 Only when contracts
with distributors were well formulated could ¬rms acquire some knowledge
of those markets.
In the 1960s, communication and transportation costs decreased, mar-
kets opened to foreign direct investment, competition intensi¬ed and became
global, and as the distribution activity of consumer goods started to consol-
idate, new imperfections in intermediate product markets emerged. Control
became an important incentive for internalization, leading to the creation
of long-term alliances or hierarchical relations in distribution. At that stage,
several alcoholic beverages ¬rms acquired marketing knowledge either from
their operations in their domestic markets or internationally. This allowed

57 Interview with Robert Mondavi, Impact International (2002).
58 Mark Casson, “The Organisation and Evolution of the Multinational Enterprise: An Infor-
mation Cost Approach,” Management International Review, Vol. 39 (1999): 119; Oliver E.
Williamson, The Economic Institutions of Capitalism (New York: Free Press, 1985): 57.
105
Channel Management
them to change the modes through which they distributed their beverages,
¬nding better combinations of risk and control for the marketing of their
brands. This also provided new economies of scale and scope in their opera-
tions, including distribution in different regions of the world. They were able
to obtain reliable feedback and better information about the preferences of
consumers and the performance of their brands.
The joint ventures formed between Western multinationals and partners in
geographically and culturally distant markets of the Far East from the 1980s
allowed ¬rms to acquire marketing knowledge about the distribution systems
and social habits in those markets. They could simultaneously overcome
tariff or other types of constraints imposed by the institutional environment.
Later in the 1980s, the economies of scale and scope in distribution pro-
vided by long-term alliances became increasingly important. Consumption
stagnated, distribution was concentrated, competition was played at a mul-
timarket level, and the costs and bene¬ts of the alternative channel designs
became very different, threatening the independent survival of nonef¬cient
¬rms.59 By aggregating different beverages and brands and targeting distinct
market segments, it was possible to obtain declining average costs associ-
ated with increasing output of a single line of commerce. At the same time it
was possible to reduce the risk and uncertainty associated with distribution,
thereby lowering the cost of market entry to the internationalizing com-
panies. Appendix 7 offers a schematic representation of the main types of
alliances used by alcoholic beverages ¬rms over time.
Overall, it is possible to identify four signi¬cant trends in distribution of
alcoholic beverages. First, in the initial part of the twentieth century the
modes of distribution of alcoholic beverages were similar to those used in
other consumer products in general.60 One striking difference about this
industry, however, is that, despite the high risk of failure of alliances, they
remained an important alternative governance structure used for distribution
even in periods such as the 1980s when ¬rms were integrating vertically
by merging and acquiring their distributors.61 While the types of alliances
formed by ¬rms may have changed over time, this did not necessarily imply
that the existing ownership structures were wrong, but that these structures
often represented transitional circumstances.

59 About multimarket competition see, e.g., Satish Jayachadran, Javier Gimeno, and P. Rajan
Varadarajan, “The Theory of Multimarket Competition: A Synthesis and Implication for
Marketing Strategy,” Journal of Marketing, Vol. 63 (1999): 49“66.
60 See, e.g., Geoffrey Jones, Merchants to Multinationals (Oxford: Oxford University Press,
2000).
61 For an analysis of failure in alliances see, e.g., Bruce Kogut, “A Study of the Life Cycle
of Joint Ventures,” in F. J. Contractor and P. Lorange (eds.), Cooperative Strategies; and
Andrew C. Inkpen and Paul W. Beamish, “Knowledge, Bargaining Power, and the Instabil-
ity of International Ventures,” Academy of Management Review, Vol. 22, No. 1 (1997):
177“202.
106 Global Brands
A second trend involves the tendency of ¬rms to use several types of gover-
nance structures simultaneously to distribute their brands, each one adapted
to the strategy of a particular market or region, to the product being traded
(wines, beer, or spirits), or the speci¬c brand being marketed. A third trend
is that alliances in distribution tended at ¬rst to involve one large ¬rm that
owned brands that were leaders in speci¬c market segments and types of
beverages. But early in the twenty-¬rst century, it was common for alliances
to be formed between ¬rms of all sizes, including leading multinationals that
were direct competitors.
A fourth trend relates to the apparent correlation between the country
of origin and the main geographical scope of activities of the largest ¬rms,
and the governance structures used to distribute their products. In countries
such as the United States and the United Kingdom, which developed capital
markets where shares of most of the large ¬rms could be publicly quoted
(even when families kept control of the shares), alliances in distribution
developed very early in the century. In continental Europe and Japan, by
contrast, this type of alliances became popular only later. Here, shares tended
to be concentrated in the hands of a small number of investors; often families,
banks, and governments, and interlocking shareholdings were common.
Cross-border alliances in distribution developed beginning in the 1930s
between ¬rms from Anglo-Saxon countries, in particular from the United
Kingdom, United States, and Canada. In the 1970s and 1980s they spread
to ¬rms from European countries and Japan. Volume and pro¬tability
seemed to have been the main determinants for ¬rms to integrate vertically
through wholly owned distribution units. Nonetheless, by the beginning of
the twenty-¬rst century, alliances were very common in the industry and were
used even between ¬rms of smaller size and those from emerging markets.
It is in this period that a convergence of strategies occurred in the industry
(even though sometimes these did not result in the most ef¬cient decisions or
appear to re¬‚ect genuine economies), especially between the world™s largest
multinationals. Publicly quoted large ¬rms, which did not follow these con-
vergent strategies, were often acquired by others that did. Only ¬rms in
which families controlled the shares were able to remain independent while
following distinct strategies. Alliances in distribution focused essentially on
obtaining economies of scale and scope in logistics, as ¬rms sought to retain
control over their marketing operations and minimize externality costs.
In addition, in a world where smooth marketing knowledge is becoming
more important for growth and survival, the creation of alliances rather than
hierarchies facilitates the acquisition of vital knowledge. The frequency of
global alliances between competitors in alcoholic beverages also shows the
increasingly strategic role of distribution in the value-added chain of ¬rms.
6

Diversi¬cation Strategies




Introduction
The vulnerability of the specialized ¬rm to fast and unexpected changes in the
environment in the last half of the twentieth century meant that ¬rms in many
industries chose to diversify as a way to grow and survive.1 The development
of the multiproduct ¬rm is often considered to be related to factors such as
excess capacity and its creation, market imperfections, and the peculiarities
of organizational knowledge, particularly its fungibility and tacit character.2
This chapter explores the role of marketing knowledge and brands as under-
utilized resources. I want to explain how despite following apparently dif-
ferent strategies of related and unrelated diversi¬cation, a group of multi-
national ¬rms from different countries achieved similar global leadership
positions.
It is very dif¬cult to classify ¬rms™ strategies over long periods of time as
involving only related or unrelated diversi¬cation. Nonetheless, despite the
unique ways through which ¬rms respond to imperfections in markets and
other factors,3 it is possible to ¬nd common patterns in their diversi¬cation
strategies. Commonalities exist not only in the products and geographical
markets they selected, but also in their vertical integration strategies and

1 See, e.g., H. I. Ansoff, “Strategies for Diversi¬cation,” Harvard Business Review, Vol. 35,
No. 5 (1957): 113“24; idem, “A Model for Diversi¬cation,” Management Science, Vol. 4
(1958): 392“414; Edith Penrose, The Theory of the Growth of the Firm (Oxford: Oxford
University Press, 1959/1995): 111“12; Robin Marris, The Economic Theory of Managerial
Capitalism (London: Macmillan, 1964); M. Gort, Diversi¬cation and Integration in Amer-
ican Industry (Princeton, NJ: Princeton University Press, 1962); Richard Whittington and
Michael Mayer, The European Corporation (Oxford: Oxford University Press, 2000). In
business history, the seminal work of Alfred Chandler in Strategy and Structure highlighted
the importance of the diversi¬ed ¬rm in the development of modern economy. Alfred D.
Chandler Jr., Strategy and Structure (Cambridge, Mass: The MIT Press, 1962).
2 Penrose, The Theory; Alfred D. Chandler Jr., The Visible Hand (Cambridge, Mass: Harvard
University Press, 1977); David J. Teece, “Towards and Economic Theory of the Multiproduct
Firm,” Journal of Economic Behaviour and Organization, Vol. 3 (1982): 39“63; Michael E.
Porter, “From Competitive Advantage to Corporate Strategy,” Harvard Business Review,
Vol. 65, No. 3 (1987): 43“59.
3 Richard R. Nelson, “Why Do Firms Differ, and How Does it Matter?” Strategic Management
Journal, No. 14 (1991): 61“74.

107
108 Global Brands
the physical and knowledge linkages they created. Some ¬rms even adopted
strategies of double diversi¬cation, engaging simultaneously in geographical
and industrial diversi¬cation.4
In this chapter, I provide empirical evidence on the imperfections in alco-
holic beverages ¬rms and markets that led ¬rms to diversify into related and
unrelated areas over time and to create different kinds of physical and knowl-
edge linkages.5 I also explain the cycles of diversi¬cation of ¬rms within the
industry. Finally, I highlight the increasing role played by brand management
in creating the knowledge linkages required for successful diversi¬cation.



Shifts in Diversi¬cation Strategies Over Time
The changes in the imperfections generated in markets and in ¬rms create
costs that affect ¬rms™ ef¬cient operations in multiple ways and ultimately
may lead them to reassess their diversi¬cation strategies.6 Imperfections in
markets are created by changes in the external environment. They include
declining demand, competitive shocks, country barriers, and policy distor-
tions such as tax and antitrust policy. These prevent ¬rms from economically
exploiting ownership advantages in any way other than by internalizing the
market.7 To minimize these costs and simultaneously take advantage of the
bene¬ts that the internalization of new linkages might provide, ¬rms often
substitute market transactions for a hierarchy or for hybrid governance struc-
tures.8
Imperfections in ¬rms are created by changes that occur inside the ¬rms.
They include the development of excess resources (tangible, intangible, or


4 Robert D. Pearce, “The Internationalisation of Sales by Leading Enterprises: Some Firm,
Industry and Country Determinants,” The University of Reading: Discussion Papers in Inter-
national Investment and Business Studies, No.101 (1987).
5 This chapter draws to a great extent on concepts from Peter J. Buckley and Mark Casson,
The Future of the Multinational Enterprise (London: Macmillan, 1976); and from Mark
Casson, Enterprise and Competitiveness: A Systems View of International Business (Oxford:
Clarendon, 1990); idem, “Internalisation Theory and Beyond,” in Peter J. Buckley (ed.),
Recent Research of Multinational Enterprise (Aldershot: Elgar, 1991); idem, Economics of
International Business (Cheltenham: Elgar, 2000): chapter 3.
6 Robert D. Pearce, The Growth and Evolution of Multinational Enterprise (Aldershot: Elgar,
1993).
7 Oliver E. Williamson, Markets and Hierarchies (New York: Free Press, 1975); idem,
“Transaction-Cost Economics: The Governance of Contractual Relations,” Journal of Law
and Economics, Vol. 22 (1979): 3“61; Benjamin Klein, Robert Crawford, and Armen Alchian,
“Vertical Integration, Appropriable Rents and the Competitive Contracting Process,” Journal
of Law and Economics, No. 21 (1978): 297“326; Buckley and Casson, The Future of the
Multinational.
8 David Teece, “Economies of Scale and Scope of the Enterprise,” Journal of Economic
Behaviour and Organisation, Vol. 1, No. 3 (1980): 223“47.
109
Diversi¬cation Strategies
¬nancial), or shifts in managerial motives and shareholder interests.9 They
may lead ¬rms to internalize physical or knowledge linkages that had not
previously existed, not even through the market.10 The costs they generate
may be of two types “ transfer costs or information costs. In this discussion
only transfer costs are analyzed.11
Transfer costs are the costs of actually moving the resources from one
location to another. In the case of physical resources in alcoholic beverages,
transfer costs include transportation costs, tariffs, and costs of overcoming
nontariff barriers. In the case of knowledge resources in alcoholic beverages,
transfer costs include costs of training. The linkages that result from the
attempt to reduce these costs are of two kinds “ physical or knowledge
linkages.
Physical linkages involve tangible assets, and are characterized by one-
way product ¬‚ows (inputs or outputs) that run from the supplier to the
consumer of those products. Plant capacity and the equipment necessary to
manufacture a product are examples of such linkages. Tangible assets such
as specialized manufacturing equipment are not very ¬‚exible in facilitating
diversi¬cation because of their indivisibility and the likelihood of creating
excess capacity.12 Often, the excess assets (plant and equipment) can only
be used for very closely related products, especially those requiring similar
manufacturing technology. Another limitation relates to the fact that these
types of assets eventually become physically exhausted.
Knowledge linkages relate to the ¬rms™ human capital, its expertise such as
marketing knowledge, or a knowledge of the technologies that can improve
the business assets of a new domain being considered for investment. This
knowledge accrues to the ¬rm over time and involves intangible linkages.
These linkages ¬‚ow from the supplier to the customer but may also be
acquired by the supplier due to its linkage with its customer.13 Knowledge is
easily transferred between separate activities, is less imitable than physical
assets, and can be repeatedly used in different products with little cost in
the effectiveness of the original operations. It is this fungible character of
knowledge assets and the excess resources that the ¬rm may generate that

9 See, e.g., Richard P. Rumelt, Strategy, Structure and Economic Performance (Cambridge
Mass: Harvard University Press, 1974).
10 The two different imperfections that may lead to diversi¬cation are not mutually exclusive.
In fact change may simultaneously produce imperfections in markets and in ¬rms.
11 Information costs may take the form of communication costs (which are costs of agreeing on
the price and quantity of the resource to be transferred, assuming honesty), or of assurance
or transaction costs (which are costs incurred in dealing with misinformation or dishonesty).
Mark Casson, Information and Organization: A New Perspective on the Theory of the Firm
(Oxford: Clarendon, 1997).
12 Penrose, The Theory; Robert E. Hoskisson and Michael A. Hitt, “Antecedents and Perfor-
mance Outcomes of Diversi¬cation: A Review and Critique of Theoretical Perspectives,”
Journal of Management, Vol. 16, No. 2 (1990): 461“509.
13 Casson, Information and Organization.
110 Global Brands
are critical in the understanding of a ¬rms™ diversi¬cation into new as well
as existing lines of business. As mentioned before, marketing knowledge
is the knowledge within the ¬rm about marketing methods, branding, and
distribution. Smooth marketing knowledge can be shared among different
industries; and sticky marketing knowledge is more limited in scope “ it is
relevant in the operation of particular geographical markets or in the indus-
try for which it was developed, and is not easily shared with other business
activities.


From Physical Linkages to Knowledge Linkages
In the beginning of the 1960s, physical linkages were more important than
knowledge linkages in determining the diversi¬cation strategies of ¬rms.
Most of the world™s largest alcoholic beverages ¬rms were either not diver-
si¬ed at all, or had low levels of diversi¬cation.14 Over time, as the size of
¬rms expanded and professional managers took on a larger role, knowledge
linkages gained increasing importance. Many ¬rms evolved into medium
or highly diversi¬ed businesses. By the beginning of the twenty-¬rst cen-
tury, ¬rms were refocusing in areas where they could obtain cost ef¬ciencies
through both physical and knowledge linkages. Appendix 8 provides the
ratio of sales in alcoholic beverages to total sales between 1960 and 2005
for a group of ¬rms.15
In each decade, the incentives for internalization were created by differ-
ent imperfections in markets and in ¬rms. As previously mentioned, in the
1960s, consumption was growing in the Western world and competition
was still largely local. There were no incentives for ¬rms such as Seagram
and IDV to diversify into other businesses rather than spirits and wines. For

<<

. 4
( 11)



>>