2011 IID Brewing Industry - Università degli Studi di Milano-Bicocca

ECONOMIA E DINAMICA INDUSTRIALE

7 Giugno 2011 Università degli Studi di Milano-Bicocca

INDUSTRIAL DYNAMICS OF THE BREWING INDUSTRY IN U.S. …AND IN ITALY Christian Garavaglia KITeS – Bocconi University and University of Milan-Bicocca, Faculty of Statistics

By the end of this lecture, you should be able to: • Be able to compare/discuss models of industry evolution • Understand research about “Organizational Ecology” and “ResourcePartitioning” • Recognize the sectoral specificities of the brewing industry • Understand the dynamics of firms’ population

REFERENCES •

Swaminathan A. (1995), “The Proliferation of Specialist Organizations in the American Wine Industry, 19411990”, Administrative Science Quarterly, Vol. 40

•

Swaminathan A. (1998), “Entry into new Market Segments in Mature Industries: Endogenous and Exogenous Segmentation in the US Brewing Industry”, Strategic Management Journal, Vol. 19

•

Carroll G. (1985), “Concentration and Specialization: Dynamics of Niche Width in Populations of Organizations”, American Journal of Sociology, Volume 90, Number 6, 1262-1283

•

Carroll G. (1988), Ecological Models of Organizations, Ballinger Publishing Company, Cambridge Mass

•

Carroll G. (1997), “Long-term Evolutionary Change in Organizational Populations: Theory, Models and Empirical Findings in Industrial Demography”, Industrial and Corporate Change, Volume 6, Number 1

•

Carroll G., Hannan M. (1995), Organizations in Industry: Strategy, Structure & Selection, Oxford University Press

•

Carroll G., Swaminathan A. (1992), “The Organizational Ecology of Strategic Groups in the American Brewing Industry from 1975 to 1990”, Industrial and Corporate Change, Volume 1, Number 1, 65-97

•

Carroll G., Swaminathan A. (2000), “Why the Microbrewery Movement? Organizational Dynamics in the US Brewing Industry”, American Journal of Sociology, Volume 106, Number 3, 715-762

•

Peli G., Nooteboom B. (1999), “Market Partitioning and the Geometry of the Resource Space”, American Journal of Sociology, Volume 104, Number 4, 1132-1153

TO SUM UP 1. 2. 3.

4.

“Industry Life Cycle” and “Orgnizational Ecology” These models fail in some predictions Go beyond these models a) Strategic Groups b) “Resource Partitioning” Model Sample: BEER!

“INDUSTRY LIFE CYCLE” STORY (1) • As the industry case studies and the empirical evidence have illustrated, many industries’ evolution follows a particular path. • The firms’ innovative capabilities and the role of size in influencing R&D investment decisions are the key elements that shape the evolution of the market structure. • The driving force of the model is given by the dependence of the returns of process R&D on firm size: bigger firms have more incentive to invest in process R&D because they benefit from a cost reduction on a larger output Economies of Scale!

“INDUSTRY LIFE CYCLE” STORY (2) • This process leads the less innovative and efficient firms to exit the market because the price falls over time as long as output expands. This means that later entrants need to be relatively more efficient in product innovation in order to find entry profitable. However, at some point in time the entry process stops, given that the potential entrants have a larger and larger cumulated disadvantage with respect to incumbent firms. • The RESULT: market concentrates and few large firms dominate the industry in the late stage of its evolution. • EXAMPLES: TVs, radios, penicillin, automobiles, tires … beer (?)

“ORGANIZATIONAL ECOLOGY” APPROACH (1) • Firms’ populations (or more broadly organizations’ populations) long-run evolution is examined in terms two sociological forces. • The process of LEGITIMATION and the process of diffuse COMPETITION drive the evolution of the populations. • The key variable of the model is given by the population density, i.e. the number of organizations (i.e. firms) N in the population (model of DENSITY-DEPENDENCE).

“ORGANIZATIONAL ECOLOGY” APPROACH (2) • LEGITIMATION: L(t) = f (N ) = αNβ t t •

The process of legitimation, indicates the taken-for-granted attitude in considering a given organizational form. A particular organizational form follows a time-process during which it gets more and more considered as the “[…] natural way of doing certain things” (Carroll, 1997). At the very first manifestation, a new organisational forms lacks legitimation. This can be observable in different context: there might be in general hostile behaviours by suppliers, customers and institutional authorities.

•

As the organizational form diffuse, its legitimation raises. The effect of this process, anyway, is decreasing: the legitimation effect added by the additional organization is positive, but it increases with a population’s density at a decreasing rate.

“ORGANIZATIONAL ECOLOGY” APPROACH (3) • COMPETITION: C(t ) = g( Nt ) = γ e

δ Nt2

• The process of competition indicates the competition among firms when they depend on the same set of resources. • The resources are limited and consequently the competition that arises when the number of the same type of organization proliferate becomes more and more fierce. This means that the effect of diffuse competition increases with the increase in N at an increasing rate

“ORGANIZATIONAL ECOLOGY” APPROACH (4) • Graphically L(t) and C(t)

C(t) L(t)

N

• L(t) and C(t) affect entry rates and mortality rates

Entry rate

Mortality rate

N

CONCLUSIONS • Organizational Ecology models predict long-term evolution of firms’ population: the time path of the number of firms follows an S-shaped trajectory. • The Industry Life Cycle model explains “shakeout” and the rise of industry concentration. Both models are unable to explain the rapid increase in the number of organizations of the population during the late stage of the industry evolution that some industries experienced (e.g. beer, telephone companies, winemakers and the newspaper industry. See Carroll, 1985; Barnett, 1997).

»Need an explanation!!

THE NUMBER OF BEER PRODUCERS IN U.S. 1975--1990 1975

Research questions: - How do we explain this pattern? - How do we reconcile the increase in the degree of concentration with the increase in the number of producers?

CONCENTRATION OF BEER PRODUCERS IN U.S.

CONCENTRATION OF BEER PRODUCERS IN U.S. Top 10 U.S. Brewers. Year: 1950 RANK

BREWER

Top 10 U.S. Brewers. Year: 1960 BARRELAGE

RANK

BREWER

BARRELAGE

1

Jos. Schlitz Brewing Co.

5,096,840

1

Anheuser-Busch, Inc.

8,477,099

2


4,928,000

2


5,694,000

3

Ballantine, Inc.

4,375,000

3

Falstaff Brewing Corp.

4,915,000

4

Pabst Brewing Co.

3,418,677

4

Carling Brewing Co.

4,822,075

5

F & M Schaefer Brewing Co.

2,772,000

5

Pabst Brewing Co.

4,738,000

6

Liebmann Bros.

2,965,522

6

P. Ballantine & Sons

4,408,895

7


2,286,707

7

Theo. Hamm Brewing Corp.

3,907,040

8

Miller Brewing Co.

2,105,706

8


3,202,500

9

Blatz Brewing Co.

1,756,000

9

Liebmann Breweries

2,950,268

10

Pfeiffer Brewing Co.

1,618,077

10

Miller Brewing Co.

2,376,543

Tot Barrelage of U.S. Brewers in 1950: 82,830,137 barrels.

Tot Barrelage Of U.S. Brewers in 1960: 87,912,847 barrels.

C10 = 38%

C10 = 52%

CONCENTRATION OF BEER PRODUCERS IN U.S. Top 10 U.S. Brewers. Year: 1980

Top 10 U.S. Brewers. Year: 1970 RANK

BREWER

BARRELAGE

RANK

BREWER

BARRELAGE

1


22,201,811

1


50,200,000

2


15,129,000

2

Miller Brewing Co.

37,300,000

3

Pabst Brewing Co.

10,517,000

3

Pabst Brewing Co.

15,091,000

4

Adolph Coors Co.

7,277,076

4


14,900,000

5


5,749,000

5

Adolph Coors Co.

13,800,000

6


5,386,133

6

G. Heileman Brewing Co.

13,270,000

7

Miller Brewing Co.

5,150,000

7

Stroh Brewery Co.

6,161,255

8

Carling Brewing Co.

4,819,000

8

Olympia Brewing Co.

6,091,000

9

Theo. Hamm Brewing Co.

4,470,000

9

Falstaff Brewing Co.

3,901,000

10

Associated Brewing Co.

3,750,000

10

C. Schmidt & Sons

3,625,000


C10= 69%


C10 = 93%

CONCENTRATION OF BEER PRODUCERS IN U.S.

• In 2002: Market shares: Anheuser-Busch, Inc. Miller Brewing Co. Adolph Coors Co Pabst Brewing Co.

55.1% 21.4% 12.2% 4.6%

C4 = 93.4%

PS: One barrel equals 42 US gallons, that is 158 litres.

a) STRATEGIC GROUPS THEORY (Caves and Porter, 1977) • Firms in an industry are heterogeneous. • But heterogeneity is limited: firms present similarities with some but not all other firms in the industry. • We might define a STRATEGIC GROUP: a set of firms with similar structural features that are symbiotically interdependent. • Strategic groups show differences in performance and conduct. • Empirical research shows only weak support to this theory.

… Organizational ecology interpretation of “Strategic Groups” theory • Helps us in identifying different groups in an industry according to some criteria: groups of firms that face different competitive constraints and challenges, and present different properties (stated goals, core technology, marketing strategy) • Organizational ecology identifies 2 broad groups: generalists and specialists

b) RESOURCERESOURCE-PARTITIONING MODEL (Carroll, 1985) Identify different groups in a population: GENERALISTS and SPECIALISTS. Generalists = organizations that make appeals to a broad range of customer tastes (i.e. they choose targets composed of some heterogeneous segments). i.e. they compete in a variety of domains (niches) simultaneously. Specialists = organizations that address specific customer tastes (i.e. they choose narrow targets with specific preferences, niches). Can explain the rapid increase in the number of firms of the population during the late stage of the industry evolution the resurgence of SPECIALIST PRODUCERS!

• The two key concepts in this model are: 1) the resource space and 2) economies of scale. • 1) For industries based on consumer products (such as beer) the resource space is interpreted as consumer demand. The dimensions of this space can be interpreted as different dimensions of taste preferences. The market has a center (or more than one) composed of mainstream tastes: this is the largest consumer resource basis of the mass market. • 2) Firms tend to occupy the center of the market (mainstream tastes), where the resource space (i.e. demand) is most abundant! Here firms can grow to be large… The increase in size yields economies of scale advantages in production, marketing and distribution intense fighting (competition) the big firms get bigger and forcing mediumsize generalists out of the market decline in the number of generalist organizations consequent rise in the concentration ratio.

• Thesis of the model: As a consequence of economies of scale, only few generalist firms survive (and they move towards the center of the market!) this process opens up niches of resources near the periphery of the market (i.e. new opportunities for new firms assessing these peripheral preferences open up) specialist organizations enter the market! • Because resources tend to be thin in these regions, than specialists located there also tend to be small. • When these resources are sufficient to sustain a specialist segment, the market can said be to be “partitioned”! Generalists and specialists do not compete: they depend on different parts of the resource space. – Let’s see the functioning of the model…

• The crucial element of the model is that the life chances of the emerging small specialists are attached to the concentration level of generalists: the more concentrated the market, the more resource pockets for specialists open up. • Generalists fight for the center of the market. Smaller generalists face economies of scale disadvantage compared to big generalists and exit the market. • The winner generalists adjust their offers to the mainstream need at the center of the market: they take over the best parts of the extinct competitors’ market segments. • But as they move towards the center, they leave some customers unsatisfied at the edges: specialist organizations enter these market pockets! …submarkets!

• Look at the brewing industry in more detail… - I - Number of firms: declining! …from 926 in 1934 to 43 in 1981! - II - Economies of scale! - III - Which kind of firms is responsible for the upsurge in the recent decades? - IV - Which resource space has been addressed?

- I - THE NUMBER OF BEER PRODUCERS IN U.S. 19751975-1990: MORE DETAILS! Specialists

Generalists

- II - ECONOMIES OF SCALE IN BEER BREWING The reasons at the foundations of technological change in beer brewing: 1) Innovations in mechanics have contributed to improve the speed of the whole production process and its efficienct: beer production is a good example for the exploitation of large scale plants. 2) Canning and bottling lines 3) Refrigeration 4) Advancements in information technology also helped in improving automation in the various stages of the production process. 5) Improvements in trasportation and infrastructures have helped in lowering traspostrtation costs that consequently represented an advatage for large scale production. 6) These factors lowered the cost of production of beer. For example, the average real cost of a barrel of beer in U.S. in 1955 was 26.90 dollars (in 1972 dollars) and it declined to 23.92 in 1975 notwistanding the higher costs of the workforce and raw materials (Tremblay, 1987).

- III - SPECIALIST FIRMS IN BEER PRODUCTION • Microbrewery = A business whose primary activity is the brewing, bottling/kegging, and distribution of beer. It may offer a tasting room or beer garden as an attraction, but this a secondary activity and consumption of product is mainly off-site. • Brewpub = A business which has two primary activities: the brewing of beer for on-site consumption and the provision of full restaurant services. It may bottle/keg some product for distribution, but this is a secondary activity.

GENERALIST FIRMS IN BEER PRODUCTION • Mass Producers = Large firms catering to mass market a “standardized” product and benefiting from economies of scale in production, advertising and distribution.

• Our focus is on the resource space that lies outside the generalists target areas: let’s analyse this idea graphically!

- IV - THE GEOMETRY OF THE RESOURCE SPACE (Carroll, 1985; Pelì and Nooteboom, 1999) • Assumptions: - Resources are distributed across multiple dimensions. - Resources represent potential consumers in the market. - Dimensions represent consumers’ preferences (or characteristics) i.e. consumers’ tastes are intended as preferences for product characteristics. - The resource space is represented by an uneven distribution of customers over the space. - Represent organizations as circles.

Graphically… Graphically … • 3D representation of the resource space

• 2D representation of the resource space

Resources are more abundant in the center! (or in a given point in the resource space. It makes no difference for the theory)

• Generalist organizations aim at the center of the market, and they slightly differentiate themselves for attracting more and different consumers: this is represented by the overlapping circles in the figure, where the common area corresponds to those consumers who purchase the product of all the generalist organizations (examples: of multipapers readers; customers that drink different brands of beers).

• Remember our focus: the resource space that lies outside the generalists target areas… that is: the amount of space (i.e. demand) available for specialists! … Is the amount of space available ouside the generalists bigger when the number of generalists is big or small??? That is: is the amount of space available ouside the generalists bigger when concentration is low or high???

• Resource space in an unconcentrated market (A) and in a highly concentrated one (B).

Market A

Market B

There is more “room” left for specialists in Market B, that is the more concentrated situation!

CONSEQUENCES!! Number of mass producers (because of Ec. Of Scale) ↓ Concentration ↑ Products become more and more standardized! Some opinions by beer experts: - “If you taste one American beer, you’ve pretty much tasted them all” (Eckhardt, 1999). - “There is little difference left in the big brewers, it’s all generic… People are looking for something different and unique” (Cattani, 1998). New niches for specialist producers open up

EMPIRICAL PREDICTIONS OF THE “ORGANIZATIONAL ECOLOGY” MODELS • Hypothesis 1: Because of the interaction between legitimation and competition forces, entry rates of microbreweries and brewpubs will increase and then decline (inverted U-shape) as the density (number) of each organizational form increases. • Hypothesis 2: Because of the interaction between legitimation and competition forces, mortality rates of microbreweries and brewpubs will decline and then increase (U-shape) as the density (number) of each organizational form increases.

EMPIRICAL PREDICTIONS OF THE “RESOURCE PARTITIONING” MODEL • Hypothesis 3: Entry rates of microbreweries and brewpubs will increase as overall market concentration rises. • Hypothesis 4: Mortality rates of microbreweries and brewpubs will decline as overall market concentration rises. • Hypothesis 5: Mortality rates of mass producers will decline with organizational size.

EMPIRICAL ECONOMETRIC RESULTS Support the theories in many industries (newspaper, winemakers, auditing firms…)!

•

• • • •

In the brewing industry: Microbrewery density and brewpub density have a significant nonmonotonic effect on the entry rate of microbreweries and brewpubs (Hypothesis 1) Hypothesis 2 is also supported: density of microbreweries and brewpubs have non-monotonic effect on mortality rates. Industry concentration has a positive effect on microbrewery and brewpub founding rates (Hypothesis 3) Death rates decline with increased market concentration (Hypothesis 4) Hypothesis 5: for mass producers, the greater the competition with larger competitors, the higher their mortality rate (see Carroll and Swaminathan, 2000, AJS)

AN ALTERNATIVE EXPLANATION • PELI and NOOTEBOOM (1999) show that the the amount of space available ouside the generalists areas gets bigger as the number of dimensions of the resource space increases. • Given that the resource space is the demand, an increase in its dimensions reflects consumers’ tastes elaboration (stands for a more diversified and sophisticated demand). • CONCLUSION: when demand becomes more sophisticated the percentage of total resources accessible for generalists becomes less and less, and more opportunities for specialists open up! • This can be interpreted in the brewing industry as an increasing dissatisfaction of consumers toward a more and more standardized product It can also be interpreted as the recent growth of demand for biological, healthy, genuine products

IN ITALY?

THE EVOLUTION OF THE ITALIAN BREWING INDUSTRY • A similar story… • Bacause of economies of scale and Merger & Acquisitions the degree of industry concentration increased in the last decades. • By the end of the XIX century there were 140 breweries in Italy, producing 161 000 hectoliters. • The number of producers constantly declined and concentration rose during the decades. • Now we have only 7 groups: the industrial structure is highly concentrated and dominated by foreign big mass producers… it is a duopoly! Now only Forst and Menabrea, that joined in 1991, represent the Italian tradition in brewing (togher with 3 other regional producers: Hausbrandt-Theresianer, Castello di Udine and Tarricone).

CONCENTRATION IN ITALY ↑ The Italian brewing industry in 2005: a duopolistic structure! (000 hl.)

%

5649,00

31,81

3814,00

21,47

1246,00

7,02

775,00

4,36

765,00

4,31

19,00

0,11

288,00

1,62

5054,40

28,46

Gruppo

Gruppo Heineken Gruppo Peroni-SABMiller Gruppo Carlsberg Italia Birra Forst-Menabrea Castello di Udine Hausbrandt Trieste Tarricone a

Importazioni di terzi Altro (Micro, depositi…) b

Totale

150,00

0,84

17760,40

100,00

NEW ENTRANTS: SPECIALIST PRODUCERS!

anno

20 05

20 03

20 01

19 99

19 97

19 95

30 27 24 21 18 15 12 9 6 3 0 19 93

# nuovi entranti

Andamento degli ingressi di microbirrifici nel mercato italiano della birra (1993 - 2007)

19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06

NEW ENTRANTS: TOTAL NUMBER!

160

140

120

100

80

60

40

20

0

FUTURE RESEARCH • I am studying the evolution of the Italian brewing industry and the resource partitioning models • Open questions: - will more micro enter? - will we see a shakeout in micro producers? - will mass producers respond to the micro threat? - but… is the micro movement a real threat for mass producers? - study the characteristics of demand (…for beer of course!): see Garavaglia WP (2008) - further develop resource partitioning arguments (…in analytical models??) -

Stay tuned!

The American Wine Industry 1941-1990 • Swaminathan (1995)

Industria Birraria in Svizzera