A w |\\|\\\\\\\\\\l|\“WWWM\|\\\\\\\\L\|\\|1\\“all

a; _")rnvt A’LA 3 3 1293 10515
hie: no W15” m

This is to certify that the

dissertation entitled

EI’HLIVJ S‘ff‘wlffioc Ehlflj bekl’t‘tnl‘s
Oxnd Prolrlab l (21

presented by

32hr: L. 11,24

has been accepted towards fulfillment
of the requirements for

Pb") degreein ECWMZCS

Major professor a

Date Jam 3 l 7.9/

MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771

OVERDUE FINES:
25¢ per day per item

RETURNING LIBRARY MATERIALS:

Place in book return to ranove
charge from circulation records

 

 

lif!‘

  
  
   

#7 “”1995
wt" '
”ix/‘5‘)

JAR—13186 t2?

   

3&4-
‘ 1

 

 

ENTRY, STRATEGIC ENTRY DETERRENTS, AND PROFITABILITY

BY

John Leonard Fizel

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Economics

1981

f-

'1
4'5
.0“,

m7 -

© Copyright by
JOHN LEONARD FIZEL

1981

ABSTRACT
ENTRY, STRATEGIC ENTRY DETERRENTS, AND PROFITABILITY
BY

John L. Fizel

The primary objective of this dissertation is to extend the concept
of entry barriers, as defined by Bain, to include strategic deterrents
and in so doing test the hypothesis that investment programs adopted by
established firms facing the threat of entry can influence entrants' ex-
pectations regarding their post-entry profit opportunities. Previous
entry-entry barrier studies considered only structural market conditions
to be barriers to entry, positing a fixed industry structure that deter-
‘mines the behavior and performance of the firms in the industry. Em-
ploying the concept strategic barriers -- investments made by established
firms which both signal and permit aggressive action against intruders --
in an analysis of entry conditions allows for the unique possibility that
conduct may have a feedback effect on market structure.

In order to determine whether strategic barriers do reduce the rate
of entry, measures of excess capacity and liquidity were incorporated in-
to equations seeking to explain inter-industry differences in the rate
of entry for a set of 40 oligopolies over the period 1959-1968. By uti—
lizing indices of excess capacity and liquidity as direct determinants of
actual entry, this study points out that strategic barriers do reduce the

rate of entry beyond that which would occur due to structural barriers

John L. Fizel

alone. Consequently, firms can no longer be considered passive agents
within a given economic environment. Rather they can be expected to
actively shape their environment so it is more conducive to their
particular long-run goals.

This dissertation also examines the influence of entry barriers on
profits exclusive of their role in entry deterrence. This study con-
cludes that previous positive correlations between profitability and
scale economies understate the capacity of scale economies to deter
entry whereas such positive relationships between advertising and profits
overstate the negative impact of advertising on entry.

In examining the main elements of the dissertation, a number of re-
lated hypotheses were tested. Given a particular blend of strategic and
structural barriers in an industry, this study shows conclusively that
foreign potential entrants are less likely to be deterred than domestic
entrants. This study also provides evidence suggesting that entry offers

little prospect for constraining oligopolistic behavior.

ACKNOWLEDGMENTS

I would like to extend my thanks to all members of my dissertation
committee. With admirable patience and dedication, my Chairman Kenneth
Boyer guided me through all phases of the dissertation and to him I am
indeed grateful. Special thanks are also due to Stephen Martin and
B.T. Allen for comments and suggestions which led to notable improvements
in this study. I would also like to express my appreciation to Warren
Samuels for his help and encouragement through my graduate years.

In addition I would like to express my gratitude to a colleague,
Stephen Husted, who always was a willing listener, perceptive critic,
and supportive friend.

Finally, I owe substantial thanks to my wife, Lynn, who was my
indefatigable editor. Her efforts to unsnarl sentences that even an
illiterate contortion artist could not deliberately comprise was a task
surely as relentless as mine. Furthermore Lynn provided valuable and
incessant encouragement throughout the dissertation process.

Of course, the author is solely reSponsible for any shortcomings

which may yet remain.

iii

LIST
LIST

I.

II.

III.

IV.

OF TABLES . .
OF FIGURES .

TABLE OF CONTENTS

INTRODUCTION . . . . . . .

Entry and
NOTES . .

Efficiency .

ENTRY BARRIERS AND ENTRY .

Limit Pricing and Entry Barriers
Empirical Analysis of The Role of

Entry Barriers . . .

Summary .
NOTES . .

APPENDIX 2A:

A FRAMEWORK FOR EVALUATING THE IMPORTANCE
OF STRATEGIC ENTRY BARRIERS

A Survey of Entry Studies

Gaskins Mbdel . . . . .
The Theoretical Mbdel .

NOTES . .

STRATEGIC BARRIERS TO ENTRY

Description of Variables

Entry and Strategic Entry Barriers
Results and Interpretation

Empirical
Summary .
NOTES . .

APPENDIX 4A:
APPENDIX 43:

Selected Data for Sample Industries
Ordinary Least Squares Estimation

iv

Page

vi
viii

15
16

31
42
44

51

54

56
59
69

71

71
79
84
99
101

105
107

Page

V. PROFITABILITY, ENTRY BARRIERS, AND ENTRY . . . . . . . . . 112
Dominant Firm Profitability and Entry . . . . . . . . . 113
Profitability and Entry Barriers . . . . . . . . . . . 118

sumary O I O C C O O O O O O O O O O O O O O O O O C O 122

NOTES 0 O O O O O O C O O I O O I O O O O O O O O O O O 124
APPENDIX 5A: Ordinary Least Squares Estimation . . . . . 126

VI. CONCLUSIONS AND POLICY IMPLICATIONS . . . . . . . . . . . 127
NOTES . . . . . . . . . . . . . . . . . . . . . . . . . 133

BIBLIOGRAPI'IY O O O O O O O O O O O O O O O O O O O O O O O O O 134

Table

4.2

4.3

4.4

4.5

4.6

4.7

4.8

‘4.9

4.10

4.11

LIST OF TABLES

Variable List . . . . . . . . . . . . . .

Description of Variables . . . . . . . . . .

Regression Analysis of Traditional Factors
Determining the Entry Condition . . . . .

Instrumental Variables Estimation of Total
Entry Utilizing Specific Entry Barriers
(Scale Variable Discontinuous) . . . .

Instrumental Variables Estimation of Total
Entry Utilizing Specific Entry Barriers
(Scale Variable Continuous) . . . . . .

Instrumental Variables Estimation of Total
Entry Utilizing Aggregate Barrier
Classifications . . . . . . . . . . . . .

Instrumental Variables Estimation of
Domestic and Foreign Entry Utilizing
a Discontinuous Scale Variable . . . . . .

Instrumental Variables Estimation of
Domestic and Foreign Entry Utilizing
a Continuous Scale Variable . . . . . .

Ordinary Least Squares Regression Analysis
of Total Entry Utilizing Specific Entry
Barriers (Scale Variable Discontinuous)

Ordinary Least Squares Regression Analysis
of Total Entry Utilizing Specific Entry
Barriers (Scale Variable Continuous) . . .

Ordinary Least Squares Regression Analysis

of Total Entry Utilizing Aggregate
Barrier Classifications . . . . . . . . .

vi

Page
80

81

85

87

89

91

95

96

107

108

109

Table Page

4.12 Ordinary Least Squares Regression Analysis
of Domestic and Foreign Entry Utilizing
a Discontinuous Scale Variable . . . . . . . . . 110

4.13 Ordinary Least Squares Regression Analysis
of Domestic and Foreign Entry Utilizing
3 Continuous Scale Variable . . . . . . . . . . 111

5.1 Instrumental Variables Estimation of the

Impact of Entry and Entry Barriers on

Dominant Firm Profitability . . . . . . . . . . 116
5.2 The Effect of Entry Barriers on Profitability . . . 120
5.3 Ordinary Least Squares Regression Analysis

of the Impact of Entry and Entry Barriers
on Dominant Firm Profitability . . . . . . . . . 126

vii

Figures
1.1

2.1

2.2
2.3
2.4
2.5
2.6

2.7

LIST OF FIGURES

Monopoly and Allocative Inefficiency .

Absolute Cost Advantage and the

Consequence for Limit Price .

A Potential Entrant's Expected Demand Curve
Scale Economies and Limit Price
Advertising and Limit Price
Limit Price and Scale Economies
Strategic Entry Barriers and Limit Price

Strategic Entry Barriers and The

Probability of Entry

viii

Page

19
19
21
45
46

29

29

CHAPTER ONE

Introduction

Entry, whether actual or potential, has an important assignment
in the framework of microeconomic theory. Under perfect competition,
entry by new firms has been shown to enforce allocative and productive
efficiency, encourage innovation, and eliminate excess profits. Models
of imperfect competition rely on the threat of potential competition --
the probability of entry -- to motivate workably competitive perfor-
mance.1 Considering the critical role of entry it becomes crucial to
continue research which will enable us to better understand the actual
functioning of this phenomenon.

Economists have given extensive consideration to a number of hypoth-
eses concerning the determinants and effects of entry. To date, however,
no study has recognized the empirical significance of strategic barriers
to entry -- investments made by established firms that both signal and
permit aggressive action against intruders. In addition, no study has
accurately scrutinized the entry-entry barrier relationship. Studies
which have attempted to determine the relative effectiveness of entry
barriers have either relied on the implied entry-entry barrier rela—
tionship inherent in the theory of limit pricing, used an incorrect
measure of entry, and/or have excluded strategic barriers from

consideration.

The primary purpose of this dissertation is to extend the concept
of entry barrier, as defined by Bain, to include strategic deterrents
and in so doing test the hypothesis that strategic responses of estab-
lished firms to the threat of entry are important elements of the entry
condition. Contrary to the traditional industrial organization paradigm
which states that structure affects conduct which in turn determines
performance, this inquiry will emphasize the feedback effects of behav-
ior on industry market structure. In addition to identifying the im-
portance of strategic barriers, this analysis will disentangle the func-
tional role of the hypothesized entry barriers by explicitly accounting
for the simultaneity in the profit-entry relationship, determining
whether these variables impact indirectly on profitability through their
direct effects on entry or whether they impact directly on profitability
through an effect independent of entry, or both.

It will be shown that the critical role of entry cannot be fully
understood when the response to entry by existing firms is ignored and
that precise conclusions regarding the actual entry retarding capabil—
ities of specific entry barriers cannot be discerned by tacitly accepting

the fact that entry barriers deter entry.

ENTRY AND EFFICIENCY

 

The role of entry as a promoter of efficient economic performance
originated from the claims for the Pareto optimal condition of a per-
fectly competitive economy. Although it may be unreasonable to expect
the structure of the manufacturing sector of an advanced economy to
conform to the textbook standard of perfect competition, the goals

remain the same nevertheless. Economists focus on how allocative and

productive efficiency may be promoted, how innovation can be encouraged,
and how income and wealth can be equitably distributed. What follows

is a summary of the relationship between entry and each of these goals,
emphasizing the ability or inability of entry to influence industrial

behavior such that performance conforms to the competitive standards.

Allocative Efficiency

 

Inefficient allocation of resources occurs whenever any producer
or group of producers is able to persistently restrict output and charge
a price above marginal cost. Such output restrictions foreclose the
market to consumers who were willing to pay a lower price but a price
which would have covered the opportunity cost of producing the addi-
tional unit(s).2

The role of entry in restraining this form of misallocation is
twofold. If unrestricted, new firms will respond to the abnormal returns
by increasing supply in the focal industry and pushing price to the com-
petitive level (equal to marginal cost). Instances of entry should also
aid in preventing misallocations from occurring in the first place by
reducing the probability that collusive agreements can be invoked and
enforced.

The proposition that unrestricted entry will eliminate allocative
inefficiency can be simply illustrated in the case of a monopoly.3 In
Figure 1.1 the monopolist faces a short run demand curve AXYB and a
corresponding marginal revenue curve MR. The monopolist maximizes short
run profits by choosing the price-output combination X = (Pm, Qm). As
a result output is restricted below the level at which marginal cost

(MC) equals demand (price) and the firm earns economic profits. In the

PRICE (5)

 

MC

 

2; QUANTITY

 

Figure 1.1 Monopoly and Allocative Inefficiency

long run investors are able to react to these "excess" profits and do
so by generating new capacity within the industry. If entry were in—
stantaneous, unrestricted, and complete, rates of return would be nor-
malized across the entire economy. This implies that the segment of
the monopolist's demand curve above MC is unavailable to the single
seller due to the influx of new competitors. The resulting demand curve
facing a monopolist is PCZYB and the correSponding price-output combi-
nation is Y = (PC,QC).4

Producers attempting to restrict output need an agreement, either
explicit or implicit, among the members of the industry to ensure that
all members do not try to sell more than their allotted shares. A
variety of divergent views and influences must be coordinated and coor-
dination requires communication among the concerned parties.S Because
the number of two-way communications increases disproportionately with
the number of communicators, increasing the number of sellers through
entry increases the difficulty of determining a cooperative agreement.6

Entry also decreases the ability to effectively enforce any col-
lective agreement. As the number of firms expands, "chiselers" will be
more difficult to detect. When violations of an accord are facilitated
and not readily detected, they may flourish.7 Other firms party to the
industry may find that the difference between their profits that result
from industry cooperation and their contributions to the control of
industry supply is in fact negative unless the group has effective moni-
toring methods and disciplinary sanctions. But as the number of firms
increase, methods to detect cheaters must become more sophisticated
consequently increasing policing costs and reducing the returns from

collusive action.

Entry, therefore, exacerbates the problems associated with coor-

dinating and enforcing output restrictions.

Productive Efficiency

 

Consideration of the role entry plays in improving the standard
of industrial performance cannot be deemed complete after reaching con-
clusions concerning only the allocative effects resulting from a non—
competitive pricing program.

Restrictions on entry in any particular industry may allow firms
within that industry to incur excessive costs. As the vigor of compe-
tition is reduced, management may relax cost controls and even tolerate

8 These higher costs may

9

what has become known as "X-inefficiency."
accrue due to excessive staffing, lavish office accommodations,

10

generous labor contracts, and relaxed monitoring of the degree and

k.“

quality of wor The welfare loss attributable to x-inefficiency may

be at least as significant as the losses that accrued due to resource
misallocation.12

Wasteful use of resources may also be realized as firms attempt to
create, protect, and extend monopoly power. This variety of expendi-
tures includes the financing of reserve productive capacity, excessive
product promotion, and political lobbying efforts.13 No social utility
is gained from any of the inputs consumed in efforts to capture market
power. While these efforts may or may not succeed, each case must be
counted as a loss to society.

Entry, by increasing the pressure on profit margins, will decrease

the ability of firms to have deficient cost controls and to incur

monopoly promoting expenses. Severe competition may even induce cost

reductions.14

Dynamic Efficiency

 

Even though imperfectly competitive markets have been shown to
violate certain necessary conditions for static economic efficiency, we
have yet to consider the elements concerned with dynamic performance.
This section is concerned with the relationship between technological
advances and competition.

Technological advances occur in three dimensions: invention, in-
novation, and diffusion. Entry has an impact on each of these phases.

Arguments concerned with the relationship between inventive activ—
ity and entry take one of two divergent paths. One proposal suggests a
direct correlation between the probability of successful invention and
the extent of rivalry in the industry. Increasing the number of inde-
pendent centers of initiative enhances the range of approaches to in-
vention. Entry therefore increases the number of Situations where a
new or non-conforming discovery may result.15

The Schumpeterian hypothesis claims, on the other hand, that inven-
tive activity is motivated by the prospect of acquiring and maintaining
monopoly profits. Furthermore, the hypothesis states that the greater
the monopoly power the more accelerated is a firm's inventive activity.16
Schumpeter's theory incorporates entry as a form of "creative destruc—
tion," but emphasizes that if entry is too rapid, profits are dissipated
and inventive activity may be retarded.17

Although the existence of monopoly may speed the rate of inventive

activity, it may also reduce the rate of adoption and implementation of

any invention. A monopolist may tend to delay innovation because he has
a vested interest in the current technology and may want to permit full
depreciation of existing equipment before introducing a new invention.
Such delays are apt to be extended the more secure the firm's excess

profits.18

A statement by Scherer, reinforced by Loury, claims that
increased rivalry may reduce a firm's incentive for invention but will
lead to an increased probability that an innovation will be introduced
at a future date.19 A firm facing competition is more apt to immedi-
ately press its advantage in hopes of capturing customers away from
other sellers. Fellner furthers this line of reasoning emphasizing
that entry may be the very act needed to ensure the adoption of an in—

20

vention. Evidence from a number of case studies confirms this point.

Investigations of individual innovations find the rate of diffusion
to be more rapid in competitive settings and when encouraged by entry.21
In sum, analysis of the relationship between dynamic efficiency
and entry indicates some degree of market power may increase R & D
effort but that entry tends to promote adoption and imitation of inven-

tions. Therefore entry, on balance, is thought to improve the progres-

sive nature of the firms in the economy.

Income-Wealth Distribution

 

Due to an inability to analyze interpersonal utility, economists
have generally been reluctant to evaluate distributional issues. Con-
sequently this section will only outline, not judge, the ways in which
restricted entry affects income and wealth distribution in the U.S.

Excess profits resulting from market power indicate a flow of

income from the average consumer to the owners, primarily stockholders,

of the firm(s). Because the wealthiest 6 percent of individuals in the
country own more than 72 percent of all corporate securities, such
transfers of income are to the relatively rich.22 Additional income
transfers to the affluent may arise in the form of generous compensation
to corporate executives.23
The redistribution of wealth parallels the redistribution of
income. Wealth is created in the process of capitalizing abnormal
returns into the market value of the firm. Wealth is directed away from
the consuming public to the original owners of the firm. A model made
to measure this distributive effect has found it considerable. Comanor
and Smiley found that in the absence of market power the share of the
wealthiest 2.4 percent of total households would decline by nearly 50
percent. Additional results indicate that the relative wealth position
of 93 percent of the households in the country would be improved.24

In conclusion, we find that serious questions concerning equity condi-

tions will arise when entry restrictions allow market power to exist.

Conclusion

 

The analysis of this chapter establishes that when entry is re-
stricted in any industry in the economy, welfare losses arise. A
relevant question to ask is -- how important are such welfare losses?

Any attempt to quantify welfare losses must serve as a rough
approximation due to the dearth of statistical data able to be used
directly for this purpose. Arnold Harberger made an imaginative attempt
to determine the order of magnitude involved for the U.S. economy.25

Using data for the years 1924 — 1928 Harberger concluded that the wel-

fare loss due to monopoly was less than .1 of 1 percent of GNP. Stigler

10

remarked that: "If this estimate is correct economists might serve a
more useful purpose if they fought fires or termites instead of monop-

"26 However he went on to add that there were a number of reasons

oly.
for believing the estimate was too low. A fundamental criticism of
Harberger's evaluation is that it accounts for, at best, allocative

inefficiency.27

Subsequent studies attempting to refine the measurement
of allocative inefficiency and to include other facets of monopoly in-
efficiency found results ranging from 1 to 10 percent of GNP.28 One
study, using a unique approach based on the concept of monopoly capital,
calculated the aggregate loss to near or even above 50 percent of GNP.29
Clearly, no resolution on the extent of inefficiency in the economy has
been made.

Even if the aggregate welfare loss attributable to monopoly is
small, losses may be high in individual industries. A study by Siegfried
and Tiemann measured the total welfare loss in the U.S. in 1963 and

30 The signif-

found results comparable to Harberger's initial estimate.
icant finding of this study was that 67 percent of this loss could be
attributed to five industries, with the automobile industry alone

31 It therefore seems

accounting for 44 percent of the total loss.
prudent to continue to devote attention to the conditions and activ—
ities affecting entry.

In summary, the net effect of entry is to revive competition in
industries in which market power exists thereby promoting allocative
and productive efficiency, stimulating adoption and diffusion of in-
ventions, and advancing an equitable distribution of income.

Does actual entry alter the behavior and performance of an industry

as theory suggests? If not, what factors inhibit its functioning? Are

11

these factors inherent to an industry's structure or are they manipuable
by the incumbent firms in an industry? It is these and other questions

to which this study now turns.

NOTES

CHAPTER ONE

1See for example: William Baumol, Elizabeth Bailey and R.D.
Willig, "Weak Invisible Hand Theorems on Pricing and Entry in a
Multiproduct Natural Monopoly," American Economic Review 67(June
1977): 350-365.

 

2Stated another way, the higher prices violate a basic
condition for Pareto Optimality. The ratio of prices between
goods is no longer equal to the marginal rate of substitution.
Misallocation of resources and a deadweight loss to society
directly follows.

3The following discussion finds its roots in: J.B. Clark,
Essentials of Economic Theory (New York: MacMillan, 1907). The
geometric analysis is similar to that utilized by William Shepherd
in his book The Economics of Industrial Organization (Englewood
Cliffs, N.J.: Prentice-Hall, 1979), pp. 288-289.

 

4Existence of conditions which prevent a complete response to
the excess profits and/or which delay the intrusion of new competition
will reduce the elasticity of the demand curve (i.e., the elasticity
will be finite above MC). In such cases all allocative inefficiency
may not be neutralized. Even so the welfare loss is diminished when
entry occurs than when entry is completely blockaded. See, Raymond
DeBondt, "On the Effects of Retarded Entry," European Economic Review
8(August 1977): 361-371 and Raymond DeBondt, "Limit Pricing,
Uncertain Entry, and Entry Lag," Econometrics 44(September 1976):
939—946.

5Note the analysis in F.M. Scherer, Industrial Market Structure
and Economic Performance (Chicago: Rand McNally College Publishing
Co., 1980), p. 200.

 

 

 

6Disagreements in communications may be derived from cost
asymmetries of the engaged parties. R.L. Bishop discusses this
point in "Duopoly: Collusion or Warfare?" American Economic Review
50(December 1960): 933-961.

 

7See George J. Stigler, "A Theory of Oligopoly," Journal of
Political Economy 72(February 1967): 44—61.

 

 

12

 

83

SLAB a
hfis,19
91
Williamsi
Cliffs,
1C

(Boston

1
the Fir

of Tat
1967)
Regul

mon0p
Signi
Prime
Revie
———‘._

Remna
(New ‘

PP- 83
0f exi
COmple
POSiti

Profit:
St '
%

Under R
374 and

 

13

8See the works of Harvey Leibenstein, primarily "Allocative
Efficiency vs. X-Efficiency," American Economic Review 56(June 1966):
372-415 and Beyond Economic Man (Cambridge, Mass.: Harvard University
Press, 1976).

9Leibenstein, Beyond Economic Man, pp. 29-46 and Oliver
Williamson, Corporate Control and Business Behavior (Englewood
Cliffs, N.J.: Prentice Hall, Inc., 1970), pp. 41-88.

 

 

 

10John Kenneth Galbraith, Economics and the Public Purpose
(Boston: Houghton Mifflin Co., 1973), pp. 186-187.

11Harvey Leibenstein, "Aspects of the X—Efficiency Theory of
the Firm," The Bell Journal of Economics 6(Autumn 1975): 580-606.

12

 

See Leibenstein, "Allocative Efficiency."

13Examples of this view are: Cordon Tullock, "The Welfare Costs
of Tariffs, Monopolies, and Theft," Western Economic Journal 5(June
1967): 224-232 and Richard Posner, "The Social Costs of Mbnopoly and
Regulation," Journal of Political Economy 83(August 1975): 807-827.

14A study which compared the productive efficiency of complete
monopolies versus duopolies in electrical utilities found costs to be
significantly lower where some competition existed. See Walter J.
Primeaux, "An Assessment of X-Efficiency Gained Through Competition,"
Review of Economics and Statistics 59(February 1977): 105-108.

 

 

 

15Sidney G. Winter, "Satisficing, Selection, and the Innovating
Remnant," Quarterlnyournal of Economics 85(May 1971): 237-261.

16See Joseph Schumpeter, Capitalism, Socialism, and Democracy
(New York: Harper, 1942).

17Schumpeter in Capitalism, Socialism, and Democracy, especially
pp. 83-106, states that growth should be achieved through replacement
of existing products or processes by improved ones. This is to be
completed by having successive monOpolists, each having attained its
position through inventive and innovative activity.
18Morton Kamien and Nancy Schwartz, "Potential Rivalry, Mbnopoly
Profits, and the Pace of Inventive Activity," Review of Economics and
Statistics 45(October 1978): 547-557.

19See F.M. Scherer, "Research and Development Resource Allocation
Under Rivalry," Quarterly Journal of Economics 81(August 1967): 359-
374 and Glenn Loury, "Market Structure and Innovation," Quarterly Journal
of Economics 93(August 1979): 395-410.

 

 

 

 

 

14

20William Fellner, "The Influence of Market Structure on
Technological Progress," Quarterly Journal of Economics 65(November
1951): 560-567.

 

21Quantitative analyses include Edwin Mansfield, "Technological
Change and the Rate of Imitation," Econometrics 29(October 1961): 741-
766 and Anthony Romeo, "Interindustry and Interfirm Differences in the
Rate of Diffusion of an Innovation," Review of Economics and Statistics
57(August 1975): 311—319.

22

 

 

Scherer, Industrial Market Structure, p. 472.

 

23Oliver Williamson, "Managerial Discretion and Business
Behavior," American Economic Review 53(December 1963): 1040—1047.

24William Comanor and Robert Smiley, "Monopoly and the
Distribution of Wealth," Quarterly Journal of Economics 89(May 1971):
327-336.

25Arnold Harberger, "Monopoly and Resource Allocation," American
Economic Review 44(May 1954): 77-87.

26George Stigler, "The Statistics of Monopoly and Merger,"
Journal of Political Economy 4(February 1956): 33-35.

27One of the most sophisticated critiques of Harberger's analysis
is that of Abram Bergson, "On Monopoly Welfare Losses," American
Economic Review 63(December 1973): 853-870.

 

 

 

 

 

28Leibenstein in "Allocative versus X—Efficiency" finds productive
efficiencies ranging around 10 percent of costs. For an analysis of
aggregate loss see Keith Cowling and Dennis Mueller, "The Social Costs
of Monopoly Power," Economic Journal 88(December 1978): 724-748.

29Joseph Phillips, "Estimating the Economic Surplus," an appendix
in Paul Baran and Paul Sweezy, Monopoly Capital (New York: Monthly
Review Press, 1966).

30John Siegfried and Thomas Tiemann, "The Welfare Cost of
Monopoly: An Inter-Industry Analysis," Economic Inquiry 12(June
1974): 190-202.

31The other industries accounting for the major portion of the
welfare loss are plastic materials and synthetics, petroleum refining,
computers, and drugs.

 

 

 

CHAPTER TWO

Entry Barriers and Entry

The theory of limit pricing has given rise to the concept entry
barrier, a differential advantage established firms have over their
potential competitors, and the tacit acceptance that entry barriers do
indeed deter entry. Specific entry barriers such as scale economies,
capital requirements, and advertising intensity have been given wide-
spread attention by economists, yet to date, strategic barriers have
been omitted from empirical consideration. Profitability—market
structure studies attempting to examine the effects of entry barriers
have neglected to examine the entry-entry barrier relationship directly.
Rather their inquiries have relied on the implied entry—entry barrier
relationship inherent in the limit pricing theory. Their findings,
interesting as they may be, fail to discern specific entry-entry barrier
relationships but instead provide insight into a number of peripheral
questions. Studies which have attempted to directly study the ability
of entry barriers to deter entry by entering these hypothesized barriers
directly into equations explaining the extent of entry, have been beset
by faulty entry measures and the exclusion of specific responses to entry
by going concerns. To date, no study has adequately dealt with the
question of entry barriers and strategic barriers and their ability to

deter entry.

15

l6

LIMIT PRICING AND ENTRY BARRIERS

 

Early limit price theorists implicitly restricted their analyses to
situations in which economies of scale relative to total market size
were assumed to be negligible and potential entrants were assumed to be
at no absolute disadvantage in relation to their established counter-
parts. In such situations, the limit price -- the highest established
firms could charge without inducing entry -- was equal to the compet-
itive price.1

The theory of limit pricing was subsequently advanced by Bain to
include and revolve around the concept barrier to entry.2 An entry
barrier is a specific industry condition or characteristic which creates
a cost disadvantage for potential entrants. The cost differential,
between incumbent firms and prospective entrants, enables the incumbent
firms in certain industries to capture abnormal profits because even at
higher prices the elevated costs experienced by entrants prevent them
from securing normal profits.

Bain defined the margin by which established firms can maintain
price above the competitive level as the "condition of entry."

Specifically

tn
0
II

(PL"PC)/PC
where
EC = the condition of entry,
EL = the maximum price that can be charged without
allowing profitable entry (the limit price),
PC = the competitive price; PC a LAC = the minimum
technological cost of established firms.
Solving for PL we find

PL = PC (1+EC)

17

which explicitly demonstrates that the condition of entry (EC) is the
maximum premium accruing to established firms without inducing entry.
Bain contended that the entry condition was an increasing function
of barriers to entry prevailing in the industry in question. He dis-
tinguished three such barriers: l) absolute cost advantages; 2) econ-
omies of scale; and 3) product differentiation. A detailed examination

of each reveals how the cost differentials arise.

Absolute Cost Advantages

 

The reasons for an existing firm incurring costs less than those to
be experienced by a potential entrant at any comparable level of output
are termed absolute cost advantages. For example the availability of
inputs, whether due to nearness in location, purity of a lode, or control
over access, can alter acquisition costs for any level of output.

Patents represent another absolute cost advantage. No entrant may
introduce a product that is identical to a patented product. Patents,
in preventing access to and utilization of a unique technique, force
entrants to use alternative processes which may be less efficient and
more costly.

The required capital needed to accomplish entry may in itself deter
entry. New firms have no past credit rating or acquired reputation that
an established firm may possess. Because investors lack complete infor-
mation about their capabilities, the newcomer is afforded less acces-
sibility to borrowed funds. When funds are obtained, entrants are apt
to pay a risk premium in the form of a higher interest rate. Entrants
will be more hampered the greater the required initial investment for

viable entry. More capital is needed as economies of scale warrant a

18

larger, in absolute size, minimum efficient plant.3 Capital require-
ments is, then another absolute cost disadvantage.

A portrayal of the limit pricing situation where an established
firm holds an absolute cost advantage over a potential rival is depicted
in Figure 2.1. The cost curves of the prospective entrant and a dom-
inant firm are labelled LACe and LACf respectively. The demand expe-
rienced by the dominant firm is DD. The demand conditions to be expe—
rienced by the potential entrant depend on the entrant's assumption
about how the current producer(s) will react to entry. Usually a
potential entrant's behavior is analyzed by using Sylos Postulate, the
assumption that existing producers will maintain their pre-entry output.
Given this assumption, the demand curve facing the potential newcomer
is the segment of the market demand curve to the right of the pre-entry
level of output.4 See Figure 2.2. If this residual demand curve rests
above the entrant's cost curve at any positive level of output, entry
will occur.

The relationship between absolute cost advantages and limit price
hinges upon the use of Sylos Postulate. Is this the policy that estab—
lished firms will adopt? Although current policies may be regarded as
some sort of statement of the firm's future policy after entry, no a
priori reason exists so as to believe this policy will continue when
firms are faced with actual entry. The price that the entrants' product
will attract -- the post-entry price —- is a function of the contri-
bution in output of the entrant and the output response of the incum—
bents. That response could consist of an accommodating output with-

drawal or a hostile output expansion which would be better or worsen,

(S)

($)

19

 

 

l D
QM MR
Output

Figure 2.1 Absolute Cost Advantage and its
Consequences for the Limit Price

    
  
 

D - market demand curve

D - entrant demand given pre-
entry output equal to Q1

D - entrant demand given pre-
entry output equal to Q2

 

 

 

Output

Figure 2.2 A Potential Entrant's Expected Demand Curve

20

respectively, the condition of entry assumed under Sylos Postulate. By
ignoring these responses, the limit pricing model ignores the oligOpo-
listic interdependence between established firms and potential entrants

it seeks to explain.

Economies of Scale

 

The presence of economies of scale, while lowering costs, may
provide a substantial barrier to entry. The larger the plant at which
scale economies are exhausted, the larger the market share an entrant
must capture to produce at minimum costs. But the greater the market
share the entrant must necessarily meet, the greater the price decline
in the entered market. A lower expected post-entry price given any pre-
entry price will lower expected profits of entry making such a course of
action less attractive.

If significant scale economies exist, the prospective entrant may
contemplate entering the industry with a suboptimal size plant. In so
doing the firm would incur per unit costs higher than those of an estab-
lished firm exploiting all available economies. The cost differential
will be exacerbated the steeper the LAC curve. As a result, economies
of scale may have a significant impact on the limit price.

The precise relationship between scale economies and the limit
price as introduced by Modigliani is as follows:5

P
L

PC [ (1+?) / (XCN)]
where

PL = the limit price,

PC = competitive price,
X = output of a minimum efficient size plant,
XC = competitive output or alternatively the
size of the market,
N = price elasticity of market demand.

($)

21

This relationship is represented in Figure 2.3.6 Given the cost and
demand conditions, LAC and DD respectively, an incumbent firm is able
to sell quantity XL without inducing entry. If an entrant offers the
additional output i, the minimum able to be produced optimally, total
industry output will equal the competitive output X0 and post-entry
price will eQual the competitive price PC. Consequently economic
profits are unavailable to the prospective entrant. The larger the
minimum optimal scale the higher that both the limit and corresponding

profits may be without encouraging new competition.

LAC = LAC
e

 

x D 9

Output

Figure 2.3 Scale Economies and Limit Price

Product Differentiation

 

Products are differentiated when a firm's product is preferred over
competing products at a given price due to either real or perceived dif-

ferences in the products. Advertising and other selling activities aim

22

at intensifying these differences. Cumulatively these activities may
result in brand loyalties, the preference of buyers to purchase the
product of an established firm. In addition, when products are complex
consumers are more ready to depend on product reputation, an attitude
favoring established firms that have brands already established as
household words. To surmount the goodwill surrounding existing products,
the newcomer must incur promotional costs greater than that of the
incumbent.7

The strength of product differentiation as an entry barrier depends
on many factors. A few of the most important are: 1) "conspicuous
consumption" -- the prestige accorded some existing firms' products;
2) the complexity of the product -- the more complex the more a buyer
relies on information from previous purchasers and their successes; and
3) the size of advertising expenditures required to establish goodwill
for a new product.8

The magnitude of these costs is of particular importance. Studies
of advertising expenditures in the cigarette industry indicate that the
costs run four to five times more for new brands entering the market

than for established brands.9

Testimony in the Proctor and Gamble-
Clorox merger case indicate that P & G felt the entry barrier to be so
high in the liquid bleach marekt that the only economical way P & G
could enter the market and obtain a sufficient market share was through

the purchase of Clorox.10

Strategies in the Limit Pricinngramework

 

The strategy of limit pricing as developed by Bain and Modigliani

permitted existing firms to opt for one of two pricing policies. One

23

option was to price at the short run monopoly maximizing level, permit
entry and face a reduction in market share. The alternative was to
adopt the limit price, forestall all entry and capture the profits
obtainable due to the difference between the limit price and the com-
petitive price. In Figure 2.3 that price is PL.ll The former choice
permits higher short-run profits but the precise level of future gains
is uncertain since the precise rate at which entry penetrates the market
is indeterminate. Existing firms must therefore compare the present
discounted value of the certain and persistent profits associated with
limit pricing to the present discounted value of the profits from the
"gamble" alternative, choosing the scheme yielding the greatest return.

The dichotomous pricing strategy is based on the assumption that
the condition of entry is constant over time. As described by Bain,
"it is definitely posited -- on the basis of extensive empirical inves—
tigation -- that the condition of entry, as defined, and its ultimate
determinants are usually stable and slowly changing through time."12
However, existing firms have a number of alternative ways to deter
entry, including the use of excess capacity and liquidity reserves. The
result is that some barriers are significantly endogenous and are able
to be manipulated by established firms in order to alter the structural
entry conditions. The implication is that an analysis of the expected
responses of established firms to entry is required to explain entry
behavior.

Bhagwati, by incorporating growth into the limit pricing model, was

able to make the initial demonstration of the inadequacies of the dichot-

omous pricing choice of previous limit pricing models.13 If demand grows

24

by the proportion 'A' of which 'K' percent accrues to the entrant, and
if KA exceeds the minimum optimal scale, (R), prevention of entry is
impossible by a pricing policy alone. In Bhagwati's words, "the theo-
retical problem thus shifts from an entry preventing price to a formal
analysis of the non-price factors which determine the share of the
existing firm in a growing market and the ways in which these are within
the range of influence of these firms."14

Realizing the incompleteness of the Bain-Modigliani limit pricing
choice, economists have attempted to refine the limit pricing model.
Recent extensions of the limit pricing idea have permitted the oppor—
tunity to choose a trajectory of prices over time rather than a partic-
ular limit price and have introduced concepts of entry uncertainty, the
recognition that entry takes time and will occur at different rates
depending on expected profits. Important as these refined models are,
they continue to fundamentally disregard the influences of endogenous
entry barriers.

For example, using optimal control theory Gaskins formulated a
limit pricing model which enabled him to demonstrate that the optimal
policy for a firm facing entry was to choose a price path rather than

to engage in discrete price changes.15

Additionally, Gaskins concluded
that the optimal equilibrium price will always be below the short run
profit maximizing price and above the entry forestalling price.

The latter of these two conclusions is sensitive to the assumption
denoting who is to receive the benefits of demand growth. Gaskins uses
the restrictive hypothesis that all increases in demand are absorbed by

the incumbent firm. If this assumption is relaxed, allowing only a

share of total demand growth to accrue to the established firm, the

25

dominant firm's long run price must fall to the entry forestalling
level.16 It has been suggested that "once the rate of entry does not
depend exclusively on a price policy but simultaneously on other
policies of the dominant firm," then the firm will ultimately be able
to charge a higher long run price.17 Modifying Gaskins' model by in-
cluding a strategy to purchase certain newcomers to the industry,
Jacquemin and Thisse found that a dominant firm would claim higher long
run profits than if a pricing policy alone were used.18 This finding
indicates that the conduct of a firm can influence the structure of the
industry through its effects on entrant behavior, and consequently that
non—price factors play an important role in entrant behavior.

Kamien and Schwartz extended the dynamic theory of limit pricing

to a probablistic framework.19

Existing firms are concerned with an
optimal pre-entry price, one that will maximize discounted expected
profits while being constrained by the uncertainty as to when new
rivals will appear. They conclude that the optimal pre-entry price
typically lies below the short run monopoly price but above the limit
price. Including risk preferences in this particular model, Baron
found that increasing risk aversion leads to a lower pre-entry price.20
If the incumbent firm, considering the risk of entry, undertook an in—
vestment program designed for the contingency of entry, entry risk may
be reduced and the long run profits of the firm enhanced. Therefore the

feedback from conduct to structure must be considered when established

firms design a strategy to counter potential entry.

26

Strategic Barriers

 

A critical element in determining profitability of an entrant in
the response incumbent firms will be expected to make as new firms
attempt to penetrate their markets. In the preceding discussion entry
was predicted whenever pre-entry price was high enough to offset the
potential entrant's cost disadvantage. This conclusion was based on the
assumption that established firms would react to entry in a possibly
irrational fashion by continuing production of their pre-entry output
and depending on the questionable presence of effective barriers to
entry. In atomistic markets an individual producer will by definition
have no noticeable effect on other producers; therefore we may expect
entry not to provoke reaction by incumbent firms. In oligopolistic
industries established firms are definitionally aware of and affected by
actions of other firms and may be expected to react to entry. If the
reactions of the going firms include anything from price wars to accom-
modation then the knowledge of the existence of entry barriers does not
tell us where the entry forestalling price will be unless we know the
reaction to entry assumed by the entrant.

The response expected by entrants hinges on the preparedness of the
established firms for the contingency of entry. Aggressive action and
an aggressive image are more likely when provisions have been made in
the pre-entry period and are specifically designed to counter the thrust
of a new firm. Aggressive ability is important in that if current pro-
ducers are incapable of injuring an entrant, the threat of retaliation
is no longer credible. "An investment analysis therefore clarifies

incumbent firms' conduct toward potential and actual entrants."21

27

Retaliatory capability may be reflected by investments in reserve
productive capacity.22 Holding excess capacity in the pre-entry period
enables an incumbent firm, when threatened by entry, to expand output
and reduce the residual demand curve facing the potential entrant. If
the post-entry residual demand lies below the entrant's anticipated
costs the entrant is deprived of a profitable return in the particular
industry. Investment into reserve capacity may not necessitate its use.
Once in place excess capacity presents a credible price war threat which
potential entrants may not wish to challenge. Therefore the potential
foreclosure of entrants' post-entry profit opportunities also deters
entry.

Retaliatory capability may also be expressed in the large holdings
of liquid assets. Ample cash reserves enhance the ability of engaging
in a price war. These funds may be used to finance the costs of ex-
panding output, particularly if the use of excess capacity entails
Operating on the rising part of the firm's average total cost curve.23
The funds may also be used to alleviate any risk of default on prior
obligations during a sustained price war. Only if an entrant builds an
equivalent cash reserve will he be able to stand toe-to—toe with the
established firm.24 This additional capital requirement represents an
additional absolute cost disadvantage for the potential entrant and as
such is an additional deterrent to entry.

Retaliation and reduction of entry may also result from expanded
investment into advertising thus solidifying the loyalty consumers have
for the established brand(s).25 The potential entrant is therefore

hampered either by increased penetration costs or disadvantages asso-

ciated with economies of scale in advertising, or both. The damage may

28

be most acute when the extended advertising program of the established
firms is directed toward a specific entrant. Anticipating such a reaction
from Kodak, DuPont opted not to enter the color film market.26

Each of the investments highlighted above, whether used individually
or concurrently, signal to the potential entrant that retaliation is
possible upon entry into the industry. The acquired aggressive image
of the established firms alters the expectations of the potential
entrants resulting in an increase in entry deterrence and consequently
a higher limit price.

To aid in the clarification of the effect of strategic barriers to
entry consider Figure 2.6. Given the cost differentials between an
existing firm (LACf) and potential entrants (LACe) and the industry
demand curve DD, the traditional limit price is PL. Assume that the
established firm maintains excess capacity which may be translated into
(XL-X1) units of output. In this case the established firm can maintain
a pre-entry price of P1 without inducing entry if potential entrants
anticipate that the established firm will engage its excess capacity
when faced by entry. Consequently, the investment into the strategic
barrier excess capacity has permitted the established firm to adopt a
pre-entry price higher than would be predicted by the traditional limit
pricing model.

The influence of strategic barriers on pricing behavior may be ex-
tended to a probabilistic framework. Referring to Figure 2.7, the

traditional limit price is P In a world of certainty, without

L.
strategic barriers, entry would occur at any price greater than PL'

However, a shroud of indeterminancy often surrounds the entry decision.

29

($) ‘

LAC

!
xL

 

 

D

v

 

 

H _——- —-

Output

Figure 2.6 Strategic Entry Barriers and Limit Price

 

 

 

 

1 3

PL PS
1.0
Probability

of
Entry
0.0 :
ATC PL P5 Pre-entry Price

Figure 2.7 Strategic Entry Barriers and the Probability of Entry

30

Independent of strategic barriers, the entry decision depends on a number
of factors including the behavior of other potential entrants, the re-
sources available to the entrant, and projected production costs. Each
of these factors affect the probability that an entrant will meet with
failure. Consequently, at any given pre-entry price there is some risk
to enter an industry. However as pre-entry price moves to levels higher
than and farther from PL, subjective entry risk is reduced and entry is
more likely to occur. The probability of entry at various price levels
is represented by PLPL'

Investments into strategic entry barriers increase the risk of
entry. Resources well equipped for retaliation represent the ability
of established firms to reduce the post-entry prospects for a potential
entrant at any pre-entry price. This results in a reduction in the
likelihood of entry at all pre-entry price levels. Established firms
now expect to be able to charge a pre-entry price, PS’ without including
entry.

Retaliatory capability investments have reduced the likelihood of
entry making PSPS the relevant probability relationship between the
likelihood of entry and any given pre-entry price. The horizontal dif-
ference between PLPL and PSPé can be described as the extent (height)
of strategic barriers to entry. This analysis points out that pre-entry
pricing need not be as low as traditional limit pricing theory suggests.

To summarize, the earliest models of limit pricing permitted firms
to price to either maximize short run profits or to forestall all entry.
Subsequent pricing models found intermediate strategies to be more

profitable for the firm. Essentially, each of these models may be

31

criticized for ignoring the response to entry by existing firms. This
is especially important since the expected profits of an entrant are a
function of the actual (or expected) response made by the incumbent
firms. The particular response anticipated by a potential entrant is
manipuable by the incumbent firms. Certain investments make credible
the threat of price warfare and are brandished to deter entry. In
deterring entry these strategic investments allow existing firms to
capture higher profits than would be captured given only a pricing

policy.

EMPIRICAL ANALYSIS OF THE ROLE OF ENTRY BARRIERS

 

Profit-Market Structure Studies

 

The limit pricing model implies a direct correlation between prof-
itability and entry barriers based on the hypothesis that if barriers
to entry exist, firms are able to price above cost and earn economic
profits without inducing entry. Consequently, most empirical studies
of the role of entry barriers have focused on the relationship between
profitability and structural variables considered to be entry barriers.
However, without a concomitant entry equation the profit-market structure
specification makes it difficult to form reliable conclusions concerning
the ability of "Bainian" or structural entry barriers to deter entry.
If high profits were found to coincide with high levels of an entry
barrier, this result could be interpreted as indicating that the tradi-
tional entry barrier does truly deter entry only if the analyst tacitly
accepts and relies on the limit pricing hypothesis. Indeed, profit-
ability-market structure studies, the most common empirical study of

entry barriers, have tacitly accepted and relied on this assumption,

32

therefore their conclusions on the role of entry barriers must be re-
garded with skepticism.

Although profit-market structure studies are unable to discern
specific entry barrier-entry relationships, this specification has pro-
vided insight into a number of peripheral questions. These include the
effect of intra-industry competition on profits, the combined entry de-
terring and entry independent effects of entry barriers on profitability,
the effects of industry growth on profit levels, and less satisfactorily,
the effect of new competition on industry performance.

A representative and comprehensive study of the profit-market
structure genre is that done by Comanor and Wilson.27 Seeking to
explain inter-industry variability in profit rates by differences in
seller concentration, market growth, and hypothesized entry barriers --
capital requirements, economies of scale, and advertising intensity --
the Comanor and Wilson study formed the basis of subsequent empirical
work.28 They found that increases in seller concentration and rapid
industry growth increased profit levels. Furthermore, they found each
of the entry barriers identified by Bain to be positively related to
profits. On the basis of this evidence, they concluded that each of the
hypothesized barriers prevented market penetration by new competitors.

Even though the positive correlation between the hypothesized
barriers and profitability seems to indicate that structural entry
barriers insulate established firms from entry and thus enables them to
increase their profits, questions still arise as to whether a strong
positive relationship between specific barriers and profits may exist

even in the absence of the barrier's ability to deter new competitors.

Take, for example, the direct association between advertising intensity

 

33

and profits found by Comanor and Wilson. Does this result indicate that
advertising is an effective barrier to entry, as hypothesized? Adver-
tising expenditures of established firms may result in a cost disadvan-
tage for potential entrants enabling the incumbents to earn economic
profits. Advertising may also enhance profits by expanding the demand

29 The

for the established product(s) through information dissemination.
first described effect of advertising is clearly that assigned to an
entry barrier, but the latter need not be. In Comanor and Wilson's
study, it may be that both roles for advertising are represented and
together amplify the importance of advertising. Consequently, one cannot
take for granted that direct correlations between profitability and an
assumed entry barrier result only from the entry deterring effects of
that barrier.3O

An alternative approach to capturing the entry barrier-entry rela-
tionship is to introduce entry barriers as direct determinants of entry.
By focusing on a single functional role of entry barriers, this direct
approach will permit more precise conclusions regarding the entry re—
tarding capabilities of specific entry barriers.

This study incorporates this direct approach into a model which
simultaneously considers the various elements explaining profitability.
Such a treatment permits conclusions about not only the question of
whether entry barriers deter entry but also the questions as to whether

static entry barriers have an entry independent effect on profits.31

Entry Barrier-Entry Studies

 

All studies to date that have introduced entry barriers as direct

determinants of entry are inadequate in that they omit strategic

34

barriers from the posited entry conditions and/or utilize defective
entry measures. Therefore the derived results of these studies must
also be considered inconclusive.

The purpose of this section is to review the empirical work
adopting an entry barrier-entry specification. Preceding this review,
however, a definition of entry is presented, one consistent with the
theoretical role ascribed to it. Using this definition as a point of
reference, it will be shown why previous measures of entry have been
unsatisfactory.

Theoretically, entry is an influx of capital into an industry in-
troduced by investors in response to high profits in the given industry.
As a result supply is increased, thereby promoting the socially desirable
objectives of lower prices, efficient resource use, and the elimination
of monopoly profits.

In this context entry involves an addition to industry capacity, by
a firm new to the industry, capable of altering industry behavior and
performance. Small scale capacity disturbances need not entail entry
as it is theoretically described. Only intrusions at a scale signif-
icant enough to reduce the demand experienced by each firm and conse-
quently their profit expectations pose a serious competitive challenge.32

A variety of forms of "entry" have been considered in the economic
literature: 1) new firm, 2) diversifying firm, 3) foreign firm, and
4) merged firm. Whether any or all of these are included in an economic
definition of entry depends on how the character of competition is
altered during the process of their appearance.

The conception of an entrant as a newly established firm dominates

33

economic thinking. Clearly the introduction of newly developed

35

facilities provides an addition to industry capacity and if the addition
is of significant size new firms must be considered entrants. Neverthe-
less, entrants seldom appear as newcomers and are relatively unimportant
as regulators of competition.34
Entrants, as a rule, are either adjacent firms extending their prod-
uct lines or foreign firms expanding their markets. In both cases
capacity additions may be made by creating entirely new facilities, or
by switching capacity from production of a different or discontinued
good to the manufacturing of the new product line. The added production
from these sources is capable of altering the demand conditions faced by
the incumbent firms. Therefore both represent pertinent cases of entry.
Well-established firms, such as these, may be the least disadvan—
taged entrants and thus have the most important influence on the per-
formance of an industry. The rate of entry is apt to be more rapid
than that made by a firm building from scratch. Established firms may
also be able to utilize the goodwill associated with the brand names of
their other products in the entered market. Similarly, past reputations
may enhance linkages to the capital market. Consequently, established
firms may be the wielders of countervailing power.35
Entry definitions occasionally and incorrectly include entry by
merger. A merger does not increase the number of entrepreneurs within
an industry. In fact any merger between firms in the same market is a

36 More importantly, "merger entry"

step toward increased concentration.
does not necessarily entail additions to industry capacity. Rather it
represents a loss of potential added capacity; the loss of a potential

entrant o 37

36

In defense of merger as a form of entry, it has been argued that
merging firms may grow significantly faster than firms expanding in-
ternally. Rapid expansion may occur through merger when: 1) small
acquired firms in toe-hold mergers no longer face critical capital and
product differentiation disadvantages, 2) synergistic unions are
attained allowing lower per unit costs or more rapid reductions in unit
costs, and 3) improved managerial talent and effort are introduced.38
Contrary to these suppositions, Goldberg found no statistically sig-
nificant tendency for mergers to increase the market share of the
acquired firm.39 The pre-merger growth rate of the acquired firms
appeared to be the determining factor in their post-merger market
shares. 0n the basis of this evidence, mergers do not appear to promote
competition.

Mergers can be motivated by elements independent of the condition
of entry. Incentives for acquisitions may result from aberrant stock
market behavior. At any point in time a firm may be undervalued rela-
tive to its potential earnings and therefore susceptible to take-overs.
Acquisitions may also be sought in order to create or enhance monopoly
power.40 These motives are quite divergent from those of a "tradi-
tional" entrant.41 Because mergers do not enhance industry competition
by augmenting industry productive capacity and because mergers are not
significantly influenced by entry conditions entry by merger should be
and is excluded from this analysis.

New firms, diversifying firms, and foreign firms that inject sub-

stantial new productive capacity into an industry of which they are a

new member are included in the economic definition.

37

The success of entry penetration is most appropriately measured by
the market share of entrants. Market share is the operative variable
in describing the current and prospective lowering of the leading firm's
residual demand curve. Furthermore, market share describes the relative
intensity of entry and therefore the degree to which firms are insulated
from entry. Finally and importantly, market share is independent of the
current market structure of an industry.

The statistical studies specifying structural entry barriers as
direct determinants of entry have utilized a number of different measures
of the rate of entry, all but one of which are significantly flawed.
What follows is a critical analysis of the rate of entry measures,
accompanied by a review of the published works incorporating each of
the respective computations.

The rate of entry calculated as a percentage change in the number
of independent firms operating in an industry suffers a series of short-
comings. The measure ignores the intensity of entry by ignoring the
size of entering and exiting firms. Such an omission may give perverse
results. If the capacity addition of a single large entrant exceeds
the loss of capacity due to the exit of a number of small firms, entry
would then be inversely related to the number of firms in the market,
reversing the usual belief. An additional problem which arises because
size is ignored is that a single entrant into a concentrated industry
signifies greater market penetration than that of a single entrant into
an unconcentrated industry.42 Therefore there is an inherent relation—
ship between the rate of entry and current industry structure regardless

of entry conditions.

38

A second shortcoming that arises when entry is measured using the
number of firms in an industry as given in the statistical data avail-
able from the Census or IRS is that one of the most prominent modes of
actual entry is excluded. Foreign entry is not included since the
Census and IRS data is only national in scope. Knowledge of factors
affecting actual entry cannot be furthered when these important cases
of entry are absent.

Finally, the Census and IRS data do not permit the effects of
mergers to be isolated when identifying the number of companies within
an industry at any point in time.

Using the percentage definition of entry, Mansfield engaged in one
of the first statistical analyses of the determinants of entry.43
Mansfield examined the rate of entry in four industries -— steel,
petroleum, tires, and autos -- hypothesizing that increases in industry
profits make entry more attractive and that increases in the capital
requirements of a MOS plant make entry more difficult. The regression
results confirmed the prior expectations; entry was positively corre-
lated with profitability and indirectly related to capital requirements.

Duetsch also employed a measure of entry defined as the percentage
change in the number of firms operating in a market in an investigation
of the determinants of entry into 4-digit Census industries for the pe-
riods 1958-1963 and 1963-1967.44 His model consisted of an equation
stating that the rate of entry was a function of industry profitability
while profitability was determined by industry growth, product promo-
tion, concentration, and diversification. The concentration variable
was used in an attempt to explain the probability of incumbent firm

reaction to entry; the greater the seller concentration the more likely

39

entry would elicit an established firm response. The role of the diver-
sification variable was essentially to capture the effect of the stra-
tegic deterrent, liquidity. Increases in diversification are expected
to increase the potential for predatory pricing. Duetsch found the rate
of entry to increase with rapid industry growth, decreased promotional
activity, and reductions in diversification, all as anticipated. How-
ever, contrary to his expectations, Duetsch found concentration to be
directly related to the rate of entry. Duetsch attributed this result
to a desire, on the part of potential entrants, to enter industries where
there were possibilities for collusion and consequently higher profits.

Studies of Canadian industries have used a measure of gross entry
calculated as the average number of new firms beyond a given size that
have appeared in a specific industry in a given period of time. Using
the absolute number of entrants instead of the percentage definition
eliminates the difficulty associated with the initial industry popula-
tion but retains all other shortcomings. The defects include the con-
siderable weight given to small firms, the exclusion of new foreign
rivals and the inability to distinguish entry by merger from new capacity
entry.

Orr, in a cross section study of entry into 71 Canadian manufac-
turing industries, specified a model of entry consisting of two equa-

tions.45

The first stated that the long run profit rate predicted for

an industry based on the level of entry barriers (a limit profit rate)

was a function of the size of a MOS plant, capital requirements, adver-
tising intensity, research and development intensity, industry risk,

and seller concentration. The second equation declared that the rate

of entry was related to the difference between the past industry profit

40

rate and the limit profit rate and to the industry growth rate of demand.
Orr found that capital requirements, advertising intensity, economies of
scale, and concentration were significant barriers to entry, whereas
growth provided an incentive for entry.

In utilizing the same measure of entry as Orr, Gorecki classified
entrants as specialists or diversifiers, domestic or foreign, in an
attempt to ascertain whether or not the entry retarding capability of
any specific entry barrier will depend upon the character of the poten-
tial entrants.46 As discussed earlier, certain prospective entrants may
be capable of enduring and overcoming many entry barriers. He found that
rapid growth in industry demand spurred entry by all classes of entrants.
His evidence also indicated that specialist and/or domestic firms were
deterred by all hypothesized barriers. On the other hand, no entry
barriers were able to thwart entry by either diversifying or foreign
firms.

Berry examined entry made by any of the 461 largest U.S. corpo-
rations into 4-digit SIC industries. The market share of entrants was
calculated "estimating total employment, times 2500, of all plants
operated in the 4-digit industries in Question in 1965, by firms re-
porting no plants operated in that industry in 1960, divided by the
value of shipments in the industry in 1963."47 A trio of problems sully
this definition: 1) mergers are included, 2) imports are excluded, and
3) a constant output per labor quotient is assumed for all industries.
The primary goal of Berry's study was to investigate the relationship
between this measure of entry, which primarily measures the degree of

diversification, and growth. As expected, increases in the industry

41

growth rate were found to enhance the rate of entry. An additional re-
sult of some consequence was that there was no significant relationship
between concentration and entry.

Only a single empirically constructed measure of entry has appeared
which is consistent with the economic definition of entry derived here.
Harris measured the market shares of large entrants, domestic and

foreign, over the years 1950-1966.48

His measure also separated
capacity entry from merger entry.49 By regressing the rate of entry on
entry barriers, Harris attempted to assess the degree of insulation
barriers to entry provided the firms in 27 4-digit SIC industries. He
found that capital requirements tended to increase the rate of entry
but that advertising intensity and economies of scale do, indeed, repel
incoming firms. However, the factor most powerful in explaining the
extent of entry was pre-entry profitability.

Although the Harris study overcomes the principal flaw found in
most entry studies, the mis-measurement of the degree of entry, his
study remains incomplete because it largely ignores the expected re-
actions of established firms to the appearance of new competitors.50
If potential entrants perceive established firms as holders of retal-
iatory capabilities and thus anticipate that these firms will react
aggressively toward entrants, price may exceed cost by more than the
structural barriers alone would seem to warrant. Consequently, by
omitting strategic barriers the previously posited relationships between
entry barriers and the rate of entry have been beset by a misspecified
entry condition.

To capture this conjectural feature facing potential entrants

requires the use of detailed and specific indices of retaliatory

42

capability. Being aware of the ammunition -- excess capacity and liquid
assets -- held by established firms, potential entrants can make a
judgement as to whether or not the use of the available ammunition is
sufficient to rule out any possibility of profitable investment, given
the industry's structural entry conditions.

A number of earlier studies have used seller concentration as a
proxy for the anticipated reactions of established firms. The hypothesis
behind the use of the concentration ratio is that optimal action by a
group of firms facing potential entry necessitates strong collusion be-
tween the dominant firms in the industry, and a prerequisite for suc-
cessful parallel behavior is a relatively concentrated industry. The
results arising from the use of the concentration ratio were mixed and,
at times, contradictory. One reason for the ambiquity in these results
may stem from the inability of this variable to indicate resources avail—
able for retaliatory purposes. Certainly some relatively high level of
concentration may be needed before interdependence between incumbents
and newcomers is realized, but a hostile reSponse directed towards a
newcomer is only feasible when the established firms have prepared in
advance for such a contingency.51 Therefore, direct measures of excess
capacity and liquidity obviate the need for using an indirect and general

measure such as concentration.

SUMMARY

A review of the limit pricing model has shown that to accurately
depict the condition of entry, strategic entry barriers must necessarily
be included. A direct examination of the entry barrier-entry relation-

ship is a valuable tool in determining the deterrent capabilities of

43

strategic as well as structural entry barriers. Once the impact of
entry barriers on entry is determined, it is then possible to conclude
whether previous positive correlations between profit and structural
entry barriers did result from the barriers' deterring effect as is
usually concluded, or whether these positive correlations arose from
effects independent of any entry barrier-entry relationship. In the
next chapter, a theoretical framework is constructed which, when
analyzed empirically, will provide additional evidence and insight into
both the entry-entry barrier relationship and the profitability-entry-

entry barrier relationship discussed above.

NOTES

CHAPTER TWO

1For example refer to: Nicholas Kaldor, "Market Imperfections
and Excess Capacity," Economica 2(February 1935): 33-50.

2Joe S. Bain, Barriers to New Competition (Cambridge, Mass.:
Harvard University Press, 1956).

3Mansfield, "Entry, Gibrat's Law, Innovation, and the Growth
of Firms," American Economic Review 52(December 1962): 1020-1030,
presents evidence that the rate of entry is inversely related to the
absolute cost of providing capital for a plant of minimum efficient
scale. “

4The assumption was developed by Bain, Barriers, and indepen4
dently by Paolo Sylos-Labini in Oligopoly and Technical Progress
(Cambridge, Mass.: Harvard University Press, 1962). Both Bain and
Sylos-Labini emphasize that this assumption represents the most
probable response of established firms. Bain further contends that
this assumption of output maintenance is justified because an "entrant
is likely to read the current policies of the established firms as some
sort of statement of future intentions regarding their policies after
entry has occurred." However, it is too much to expect that established
firms will lie idle as new firms reshape the economic environment within
which decisions by established firms are made. Consequently, no apriori
reason yet exists that declares Sylos Postulate to be a better assumption
than either agressive output increases or entry accommodating output
decreases.

 

 

 

5Frances Modigliani, "New Developments on the Oligopoly Front,"
Journal of Political Economy 66(June 1958): 215-232.

 

6This representation follows the presentation made by Modigliani.
For a more standard description as adopted from Modigliani's work see
footnote 11, this chapter.

7Suppose that a dominant firm faces constant unit production
costs and a decreasing average advertising cost function such that the
average total costs for the firm are equal to ACf. Independent of any
other barriers to entry, the entrant would presumably face the identical
average total cost curve. However, to overcome existing brand loyalties
and penetrate the market, an entrant must incur additional promotional
cost represented as PCe. PCe increases as the degree of penetration

44

45

needed for viable entry increases. Therefore the average total cost
curve facing an entrant is AC which is the sum of AC and PCe. Con-
sequently, the dominant firm can raise its price to FL and capture
economic profits without inducing entry.

AC

 

 

($) eAC
‘ f
PL :\ ——
\q’
‘ PC
e
I D entran D market
Q Output

Figure 2.4 Advertising and Limit Price

8For a synopsis see William Comanor and Thomas Wilson, "The
Effect of Advertising on Competition: A Survey," Journal of Economic
Literature 17(June 1979): 453-476.

 

9Federal Trade Commission, Staff Report of the Bureau of
Economics on Cigarette Advertising and Output (Washington, D.C.:
Government Printing Office, March 1964).

10FTC v. Proctor and Gamble Co., 386 U.S. 568 (1967). In a re-
cent study, "Estimating Advantages to Large-Scale Advertising," Review
of Economics and Statistics 55(August 1978), p. 434, Randall Brown
indicates that the required advertising costs per unit of sale is 482
higher for a new brand than an established brand.

11The results of this demonstration would be identical to that of
an entrant facing higher costs resulting from operating at a suboptimal
scale. The effect of minimum optimal scale is shown in Figure 2.5.
Potential entrants facing cost conditions AC will not enter if pre-
entry price is set at PL which is the price at which post-entry profits
will be normal, given Sylos postulate. This occurs when the entrant's
average cost function is just tangent to its residual demand curve.
Similarly, potential entrants facing costs AC2 will be deterred at

46

prices equal to and below P'. Therefore limit price is directly related
to the size of a minimum efficient plant.

($) AC AC

P
1‘ 1\“~‘ -dlI>

 

industry dema

 

Out t
Figure 2.5 Limit Price and Scale Economies pu

12Bain, Barriers, p. 18.

13Jagdish Bhagwati, "Oligopoly Theory, Entry Prevention, and
Growth," Oxford Economic Papers 22(November 1970): 297-310.

14

 

Bhagwati, "Oligopoly Theory," p. 310.

15Darius Gaskins, "Dynamic Limit Pricing: Optimal Pricing Under
the Threat of Entry," Journal of Economic Theogy 3(September 1971):
306-322.

 

16N.J. Ireland, "Concentration and the Growth of Market Demand,"
Journal of Economic Theo£y75(0ctober 1972): 303-305.

17Alex Jacquemin and Jacques Thisse, "Strategy of the Firm and
Market Structure," in Keith Cowling, ed., Market Structure and Corpprate
Behavior (London: Gray-Mills, 1972), p. 71.

18

 

Jacquemin and Thisse, pp. 71-73.

19Morton Kamien and Nancy Schwartz, "Limit Pricing and Uncertain
Entry," Econometrics 39(May 1971): 441-454.

 

20David Baron, "Limit Pricing, Potential Entry, and Barriers to
Entry," American Economic Review 63(September 1973): 666-674.

21Richard Caves and Michael Porter, "From Entry Barriers to
Mobility Barriers," Quarterly Journal of Economics 91(May 1977): p. 245.

22For discussions concerning the use and effectiveness of excess
capacity as a barrier to entry, see Caves and Porter, "From Entry
Barriers," and Michael Spence, "Entry, Capacity, Investment and

 

 

47

Oligopolistic Pricing," Bell Journal of Economics 8(Autumn 1977): 534-
544 and John Wenders, "Excess Capacity as a Barrier to Entry," Journal
of Industrial Economics 20(November 1971): 14-19.

23Retaliation with excess capacity may not necessitate higher costs
if the plant in use provides minimum costs at the expanded output level
or if the plant provides constant costs over a wide range of output.
See Wenders, "Excess Capacity."

 

 

24If an established firm "finds itself matching expenditures or
losses, dollar for dollar, with a substantially smaller firm, the length
of its purse assures it of victory ... The large company is in a position
to hurt without being hurt." For an extended analysis review Corwin
Edwards, "Conglomerate Bigness as a Source of Power," in Business Con-
centration and Price Policy (Princeton: Princeton University Press,
1955), pp. 351-359.

25

 

 

' espouses this view.

Spence, "Entry, Capacity,’

26This example was extended by James Brock in "The Photography
Industry," unpublished Ph.D. dissertation, Michigan State University,
1981, p. 90.

27William Comanor and Thomas Wilson, "Advertising, Market Structure,

and Performance," Review of Economics and Statistics 49(August 1967):
423-440.

28For a discussion of a number of subsequent profit-market structure
studies refer to: Leonard Weiss, "The Concentration-Profit Relationship
and Antitrust," in Industrial Concentration: The New Learnipg, H. Gold-
schmid, et.al. (eds.) (Boston: Little, Brown, 1974): 201-220.

29Information dessimination may be the prominent role of adver-
tising in certain industries. Refer to Kenneth D. Boyer, "Informative
and Goodwill Advertising," Review of Economics and Statistics 56(Novem-
ber 1974): 541-548.

 

 

 

30This is equally true for empirical work using qualitative
measures of entry barriers. Note Bain, Barriers to New Competition and
H. Michael Mann, "Seller Concentration, Barriers to New Entry, and Rates
of Return in Thirty Industries, 1950-1960," Review of Economics and
Statistics 48(August 1966): 296-307.

31In a system of equations explaining profitability, advertising,
and concentration, Martin attempts to answer a similar question. His
investigation attempts to determine whether entry barriers affect
profitability directly or indirectly through their effects on concen-
tration; he finds the latter to be true. Stephen Martin, "Entry
Barriers, Concentration, and Profits," Southern Economic Journal
46(October 1979): 471-488.

 

 

48

32Within the framework of the limit pricing model, the demand

experienced by each firm will be some portion of the total industry
demand with the slope of the demand curve facing each firm being
identical. Effective entry will cause a parallel inward shift in each
incumbent firm's demand curve. For a detailed exposition see France
Modigliani, "New Developments on the Oligopoly Front," Journal of
Political Economy 66(June 1958): 215-232.

33This concept was promoted by Joe S. Bain, Barriers to New Com-
petition (Cambridge, Mass.: Harvard University Press, 1956). Although
Bain later recognized the possibility of entry by a firm supplying other
markets his limit pricing model focused on the new establishment.

34

 

 

Iowa Beef Packers represent an exception to this rule.

35Analysis of diversification can be found in: Howard Hines,
"Effectiveness of Entry by Already Established Firms," Qparterly Journal
of Economics 71(February 1957): 132-150 and Charles H. Berry, Corporate
Growth and Diversification (Princeton: Princeton University Press,
1975). For a study of import competition see Peter C. Frederiksen,
"Prospects of Competition from Abroad in Major Manufacturing OligOp-
1ies," Antitrust Bulletin 20(Summer 1975): 339-371. A synthesis is
made by Frank Kottke, "Market Entry and the Character of Competition,"
Western Economic Journal 5(December 1966): 24-43.

36The merger of two or more firms within an industry may allow them
to compete more effectively if increased size enables the "new" firm
better access to capital and/or realization of scale economies. On the
other hand the smaller number of firms may make firms more aware of their
interdependence and lead to parallel behavior.

 

 

 

37The courts have reviewed the potential competition doctrine in
several cases. A pair of cases with divergent results are: 1) U.S. V.
Falstaff Brewing Corp., 410 U.S. 526 (1973) and 2) FTC v. Proctor and
Gamble Co., 386 U.S. 568 (1967).

38The efficiencies may occur but do not necessitate the acceptance
of mergers. See Scherer, op. cit., p. 136.

39Lawrence Goldberg, "The Effect of Conglomerate Mergers on Com-
petition," Journal of Law and Economics 16(April 1973): 137-158.

 

0Any merger in the same market is a step to decreasing competi-
tion. A vertical merger may exacerbate the capital problems facing a
new entrant. A conglomerate merger may destroy competitors through
cross-subsidization or deny access to sellers and markets thrOugh
reciprocal buying and selling arrangements.

1Substantiation of divergent motives for capacity entrants and
merged entrants is found when Michael Gort reports that mergers increase
as entry barriers increase. See "An Economic Disturbance Theory of
Mergers," Quarteglerournal of Economics 83(November 1969): 624-637.

49

42One entry divided by a few incumbent firms is greater than one
entry divided by several established firms.

43Edwin Mansfield, "Entry, Gibrat's Law, Innovation, and the
Growth of Firms," American Economic Review 52(December 1962): 1023-
1051. Refer to Appendix 2A for an exhaustive account of the entry
studies.

44Larry Duetsch, "Structure, Performance, and the Net Rate of
Entry into Manufacturing Industries," Southern Economic Journal 41
(January 1975): 450-465.

45Dale Orr, "The Determinants of Entry: A Study of Canadian Manu-
facturing Industries," Review of Economics and Statistics 56(February
1974): 58-66.

46See the following by Paul Gorecki, "The Determinants of Entry
by New and Diversifying Enterprises in the U.K. Manufacturing Sector,
1958-1963," Applied Economics 7(June 1975): 139-147 and "The Determi-
nants of Entry by Domestic and Foreign Enterprises in Canadian Manu-
facturing Industries," Review of Economics and Statistics 58(November
1976): 485-488.

47

 

 

 

 

 

Berry, Corporate Growth, p. 126.

 

48Maury Harris, "Entry and Barriers to Entry," Industrial Organ-
ization Review 4(NO. 3 1976): 165-174.

49This study, in an attempt to identify cases of entry, adopts
the entry search strategy used by Harris. This strategy involves the
scanning of trade journals and studies, general business publications,
and government documents. Although the strategies to locate entry are
the same, the entry measures of this study differ from those used by
Harris because different time frames are analyzed. For a detailed
account of the process used to construct the entry variable in this
analysis, see Chapter 4, and for a list of the most relevant sources,
refer to Appendix 4A. In addition to this difference, this study
deviates from the Harris study in the measurement of the structural
entry barriers. Furthermore, as will be discussed, Harris omits
strategic barriers.

501m good conscience, Harris cannot be held singularly accountable
for this critical omission, since all other previously cited entry-entry
barrier analyses also ignore the retaliatory capabilities of existing
firms.

 

 

1For example, assume there are two industries both of which have
identical high levels of concentration but industry one includes firms
with substantial productive and financial reserves whereas industry two
includes member firms which possess insignificant reserves. Further
assume that an analysis of structural conditions alone indicates that
both industries would permit identical rates of return to a new

50

entrant. Given this framework, it is predicted that potential entrants
would rather enter industry two where the threat to potential profits
is weaker or non-existent.

APPENDIX 2A

A Survey of Entry Studies

 

 

 

Homes
wcfiuaum
uHaHH no
show momma Ho
Hammad woa ma ANV Ivoa wean:
ouofi Adv mouaufiumnSm mumau fmwuumspafi meuaw
.A-V.m “A-Vomi-vu is “.m-avw . m .N -.ms sous. sameness do Hagan:
mauve mauv< “Auvas as .u .m we “was moss uswaeum as mwameo Askmsv
he pmuuowmm huuoo .< .«M .«mvm u am .H uuouwp Imcmfi Hm ousaomnm who
newsman camououuw
a mu assessafia Ross uawaeus
NO on: muuoaasm a Imoma mom
was m we sowumaou Hopes onu mo show mood gawNVIq
luoo uoouavoa was vooavou mommawumo houso Immafi «mg
.Alvn ~o+nqn+umn+ «0 momma mouuu mauam «0
"A-v< ux+vu ux+vu <~s+wsnos a m .N -Haumume unanaH Muses: as Askmsv
"he mouuommm amuse Hm+m m+o< u m .H oawsmxo msmnoo omaonu N sumuaon
knuco ou
mmoauom wouma as opposes huuco onu one cowu mofiuumsv msuam mo
coaumaoommm uoxmma saws soaumuuamosou aw Imuueoosou momfi noH mam Hones: ca Ammmav
I u moosvou huuao owcmnu onu voumaouuoo ea owcmnu Incmd lame and omameo N saxoswuz
knuco do
.vou Hmu mmaa asoaouuom
atom usage so -oama mso .muuas macaw mo
amuse mounpcfi m unused woa moumafiumo muuommo lance .uh .Hooum possum aw Amomfiv
hoods woodman M ouManm u m ousumoo oH wcwhum> .mou:< owameo N vaowwmamz
maqumm HZduHmHonm Ammo: mmomm=m . nonmm mmmm>HZD mmbm<mz , wnDHm
Hmazm

 

 

 

 

 

 

 

 

mofivsum mounm mo mm>u=m <

”<N Nanmmm<

51

.vodfiaoov huuao mo ammo any man vomoouo
no“ muoauuun .omou owuou osu mo oaam> onu m4 .muuco mo ammo osu vouoUchw canon omouo>oo one .~

 

52

 

 

 

 

 

 

 

 

.huuao spam
ufiumoaov noxpcm soc
hHao uuowmo mumauumn
.A+Vo xa+vgm space
"pagan awwouom oeumoaov mowuu maufim
.Alvo “Alvm was awaouom unavaH mo Hogan:
“A-V< “Auvse xA+Vsm cmosump moss cassaaao as «manna Asamsv
“macaw usumuaoc Askmsv use on Human emfiamafiumau Incas sawfluum as ouaaomnm axuuuou
.A+Vu “he muuao
wouuomum muuao spam spam we“ msuaw
wawow .Auvm mA|V< lumflxo can moan» mo amass:
mA+vu "he vouoommo 3o: cooeuon mama lanes“ aw owawnu Amsmgv
huuco spam so: o+oqm+<mm+m~m+Ho a m emaawsauuwv Iwnmfi .x.= an condoned axuouoc
mmwuumav
mauso uoow IaH mamaoo Auowuos
new so: was uswaeus mcsssso
mcowuavooo Inmaowuouon new mucosa
.A+vo “A-V>o sm+osm+>omm+ spouses“ moms uuou menu lam spasm Amkmav
"an vouoommm muuco oum+sm<am+oo I m ouwuumsaaa nomad such ace uoxums huuom
magbmmm Hz<onszHm Ammo: mwommbm aonmm mmmm>Hz= mm=m<mz renew
wmszm .

 

 

 

 

53

 

msuou nouns
o>auuommoulao

owumu omouo>ouu>o

Am<vmoalsm¢

A.mua£m aav

uaoaaoao>ov mam sensuous mo Ho>oatm

Awioalsx

mm: mofiwu muumsvcfi aw
muommo omouo>m Nuance:
I muaoaouasoou Houwaoolx

oauuu moaomxwaamuuuo>vml<

oauou soaumuuaouaou shamtelu

possum sq museum huuuovaalo

muawoua usaaau«m

 

 

 

 

 

 

onam huumnvaalmm Amvwoalam muamoua vo>wououum
museums xmauuu mmz mo unsmmoalm anaemia
wax mammHm<>
A%Huao
spam swam
snow was
.waaumw
muoauuon Ixo .30:
Sauna mo and: movoaoowv
mmoao>auuom no mama .muomuu
two ocflsuou Honuao ma loo ommfi
.Aav< lop ou ousm wouhaoso umon mo
“Anew ma+vs mx+vm Au .u -aua spasm coma scan» manna gas Aeneav
"he vouoouwo Amuse .< .M .m .mvx I m Hmouoo on: Iowa" nonvoa mu Inca coma menus:
magammm HZdUHmHonm Ammo: mmommsm ‘monmm mmMm>Hza mapmdmz wnaam
wmazm

 

 

 

 

 

 

 

 

CHAPTER THREE

A Framework for Evaluating the Importance
of Strategic Entry Barriers

The limit pricing models described in the previous chapter dis-
tinguished between two types of competition, actual competition between
established firms, and potential competition arising from potential
entry of firms outside the industry. Actual competition has great
importance as a regulator of business activity. Each firm recognizing
its interdependence with other established firms in the industry must
formulate policies acknowledging that any change in policy is likely to
elicit a response from its rivals. However, high oligopoly prices de-
pend on more than the ability to control competition among existing
firms. Potential competition is also a determinant in policy-making.
Unusual profits from oligopoly pricing provide incentives for investors
to enter the industry. Entry of new firms will reduce concentration
within the industry making coordinated actions among existing firms in-
creasingly difficult. Furthermore, entry will tend to force profits and
prices toward competitive levels. Therefore, there exists a double form
of interdependence —- interdependence between the established firms and
also interdependence between established firms and potential newcomers
to the industry.

Previous models have recognized portions of this interdependence
and have totally neglected other aspects. A major assumption to date

is that firms are constrained by given market structures and must do

54

55

the best for themselves within the context of that environment. As far
as structure itself is concerned the established firms are considered
largely passive. One must reject the assumption that firms are fixed
bodies of assets unable to take an active role in influencing their
economic environment and recognize that sellers are able to control
entry by various investment strategies, which in turn, preserve or alter
the industry structure to the long run benefit of the firms.

In this chapter, a conceptual model of the limit pricing paradigm
is constructed which enables analysis of the entry decision from both
the perspective of existing firms and potential entrants. As a basis
for the conceptual model, the Gaskins dynamic limit pricing model is
described and employed. The Gaskins model is then transformed to ex-
plicitly account for strategic barriers, including both the costs to the
incumbent firms and the benefits that accrue to those firms as a result
of the strategic barrier's ability to deter entry. Next, the extended
Gaskins model is generalized to make it Operational for empirical anal-
yses, resulting in a new and unique specification with which hypotheses
about entry, entry barriers, and profitability can be tested.

Specifically, tests are conducted to evaluate the following:

1) the potential entrants' perceptions of strategic barriers and con-
sequently, the impact of those barriers on the behaVior of entrants;

2) the effectiveness of the traditionally hypothesized entry barriers
in deterring actual entry; and

3) the influence of the traditional entry barriers on profits exclusive

of their role in entry deterrence.

56

In examining these main elements, a number of related hypotheses are able

to be tested. These include:

1) the differential impact of static and strategic barriers on different
classes of entrants; and

2) the competitive impact of entry, including the individual as well as

the composite effects of foreign and domestic entry.

GASKINS MODEL

 

Gaskins models the long-run profit maximizing behavior of an estab-
lished dominant firm/group when constrained by the possibility of entry.1
Gaskins' basic premise is that the rate of entry into a particular mar-
ket in an increasing function of the difference between the existing
price and the limit price. In order to solve for the time path of prices
yielding the highest long-run profits for the firm, he maximizes the
functional
(1) v = I [p(t)-c1[f(p(t))-x<c>,]e‘“dt

t=0
subject to the rate of entry condition

2 d - ‘—
( ) d_:, - kOIp(t)—p]

where
V = present value of dominant firm profits,
p(t) = product price at time t,
c = constant average total cost,
f(p(t)) = industry demand schedule at time t,
x(t) = level of rival sales at time t,
r = firm/group's discount rate,
kQ = entrant's response coefficient,

p limit price.

57

The Gaskins model captures the simultaneity inherent in the potential
entrant-established firm relationship through the interaction of the
variables x(t) and p(t). The value of entry output x(t) influences the
pricing policy of the dominant firm at any point in time while con-

comitantly the choice of a pricing policy affects the rate of entry

dx.
dt

vestment policies of the firm (strategic barriers) and the rate of

 

Even so, the model, as is, ignores the interactions between in-

entry.

Nonetheless, the Gaskins model of limit pricing can be adapted to
incorporate the costs and benefits of entry retarding investments. Let
c1(I) represent the costs incurred by investing in such strategic entry
barriers as excess capacity or a reserve of liquid assets. If invest-
ment in excess capacity entails the building of a plant either designed
to minimize production costs at the retaliatory output level or to pro-
vide constant average costs for a range of output, the cost penalty to
the incumbent firm amounts to the increased pre-entry per unit costs of
production. Moreover, these costs depend on the elasticity of demand
which describes the extent that additional production is required to
reduce pre-entry price to or below the potential entrant's breakeven
price. Added costs also accrue to a firm maintaining large liquid
reserves. Maintaining such liquid reserves may not be part of an
optimal portfolio. The essential cost of holding liquid assets is,
therefore, the opportunity cost incurred by not investing these par-
ticular funds in capital projects or longer term securities.

In a very simple model the costs of strategic barriers may be in-

troduced as a reduction in the present discounted profits such that

58

(3) V= {[pm—c][f(p(t))-x(t)]-c'1<I>}e'“dt.

"'58

t 0
Investments in strategic barriers may result in not only a direct
effect on the firm's profits, but also a preservation of the current
industry structure. As potential entrants come to anticipate the utili-
zation of retaliatory capability held by the established firm, the rate
of entry declines.2

(4) j—f = Iko<p<t)-i§>+k1<1)1

In equation (4), k1(I) represents the reduction in the rate of entry

given any pre-entry price where it is assumed that

(5) k{(1)<o; ki"(I)ZO.

The problem is then to choose a level of strategic investment (I)
which maximizes V, subject to (4), taking into account the effect of the
choice on both the rate of profit and the rate of entry. The problem
may be solved by applying the Pontryagin Maximum Principle to the fol-
lowing Hamiltonian:

(6) H = {lp(t)-cl[f(p)-XJ-CI(I)}e'rt+21ko(p-5)+k1(I)J
where z is the adjoint variable; the multiplier associated with the rate
of entry constraint. To obtain the necessary conditions for a maximum,
set the partial derivative of (6) with respect to I equal to zero:
(7) {up}. + z d_k_1 = 0

dr d1
which may be rewritten

(8) Z dkl = e-rt dCl
dI dI

59

The left hand side of (7) is the shadow value of a marginal decline in
the rate of entry resulting from investment in strategic barriers; the
right hand side of (7) is the marginal cost of strategic investments.3
It is profitable, therefore, to invest in retaliatory capability until
the marginal cost of such a program equals the marginal benefit of re-
duced entry.

In the previous chapter, it was hypothesized that strategic bar-
riers would enable incumbent firms to increase their entry forestalling
price and thereby increase their long-run profits (refer to Figure 2.7).
A comparison of the optimal price paths derived in Gaskins' original
model and the extended model presented here clarifies this point. The
differential equations depicting the respective optimal price trajec-

tories are:

(9) P = k0(D-C)+r[x-f(p)-(p-c)f'(p)] (Gaskins original)4
-2f'(p)-(p-C)f"(p)

(10) P = k0(P'C)-k1(I)+rIx-f(p)-(p-c)f'(p)] (Gaskins extended)
-2f'(P)-(p-C)f"(p)

The only difference in the two equations is the term -k1(I) which de-

 

 

scribes the reduction in the rate of entry resulting from strategic in-
vestments. If these investments are indeed effective, ki(I) 0, then
the dominant firm will charge higher prices at every point in time than

if the investments had not been made.

THE THEORETICAL MODEL

 

The theoretical model is a generalization of the extended Gaskins'
model of dynamic limit pricing. Preserving the primary implications of
the Gaskins model, it assumes that existing firms take into account the

entry inducing effects of their pricing decisions and it is also assumed

60

that potential entrants are conscious of an expected post-entry price
level below which they will not enter the industry. Accordingly, the
theoretical model is presented in two parts; the first examines the
factors influencing the entry decision of potential entrants, the second

studies the elements conditioning the behavior of established firms.

The Potential Entrant

 

In this section an equation is derived which focuses on the deter-
minants of entry. This expression permits direct investigation into the
relative importance of individual factors which may enhance or impede
the rate of entry and also allows investigation into how these factors
influence the entry decisions of different classes of potential entrants.
The expression is based on the extended rate of entry equation, pre-

viously presented.

(4) If . _[k0(p-p)+kl(l)l.

Consistent with the extended Gaskins model, the rate of entry (E)
is assumed to be directly related to the difference between pre-entry
profit levels (n) and the limit profit rate (wL), and the level of in-
vestment into strategic barriers (I).5 This equation may be generalized
to form the potential entry function
(11) Et+1 = EI(w-flL)t; It]°
where
= the rate of entry in period t+1,
= dominant firm profitability in period t,

"L the limit rate of return in period t,

I = strategic entry deterring investments
made in period t.

t+1
w

61

Given that the limit profit rate is unobservable, a complete under-
standing of the above relationship begins with the necessary examination
into the origin of the limit profit rate.

The limit profit rate is considered to correspond to the limit
price, as described by Bain, and is, therefore the profit margin which
established firms can persistently capture without attracting entry.
The level of the limit profit rate is dependent upon the condition of
entry.6 Focusing on the specific components of the condition of entry,
the limit profit rate may be presented as a function of the structural
barriers to entry and the industry growth rate of demand:

(12) "L = NLIS;C;At;Gt]

where

the limit rate of return in pre-entry period t,
the output of a minimum optimal size plant,

the capital requirements of a MOS plant,

= advertising intensity of established firm

in period t,
the growth rate of industry demand.

If
‘3
C
At

Gt

Each of the limit pricing models presented in the previous chapter
hypothesized that structural entry barriers would be positively related
to the limit profit rate. However, certain modes of entry may be able

to overcome or endure these alleged barriers. Therefore it is expected

that
63 —'
where B denotes both the particular entry barrier -- S, C, or A -- or an

amalgamation of the specific barriers.
Entry barriers through their effect on the limit profit rate are

expected to have a negative impact on the rate of entry. The more

62

extensive the barriers to entry, the higher the limit profit rate, and

the narrower the gap between the observed dominant firm profit rate and
the limit profit rate. Consequently, higher structural entry barriers

imply a reduced entry rate.

The limit profit rate is also a function of the growth rate of
demand. In the previously cited works of Ireland and Bhagwati, growth
was incorporated into a limit pricing framework and found to decrease
the entry forestalling price or, in this case, the limit profits.7 This
relationship may be expected to occur for a number of reasons. First,
when market demand is growing, the price decline associated with any
given entry size is smaller. Additionally, growth may imply the influx
of customers without established buying patterns who are therefore more
easily captured by an entrant. Finally, managers of certain potential
entrant firms may be motivated and attracted by growth prospects thereby
using rapid demand growth as a signal to enter an industry. It follows
that independent of any anticipated reaction, hostile or accommodating,

by established firms to entry, that

(14) BNL

SG— < 0.

The inverse relationship between growth and limit profits indicates
that rapid growth will be related to low limit profits thereby in-
creasing the difference between observed profit rates and limit profits.
Accordingly it is assumed that growth has a positive impact on the rate
of entry.

In addition to the impact of actual profits and limit profits on

the rate of entry, behavioral variables, representing the expected post-

entry reaction of established firms, could affect the potential entrant's

63

decision to enter. In particular, if the usual production capacity ex-
ceeds industry demand, the level of excess capacity in the industry can
indicate retaliatory capability. However, the mere existence of excess
capacity need not always be considered a pre-emptive investment. Excess
capacity may result from the accumulation of obsolete capital -- old and
inefficient machinery differing considerably from the more technolog-
ically advanced equipment used on a regular basis by the firm. An in-
teresting example of the role of excess capacity was discussed in the

Wall Street Journal.8

The chairperson of Marquette Cement Manufacturing
Company stated that a disclosed plan of the Mississippi River Fuel Cor-
poration to diversify into cement manufacturing "exhibited a colossal
ignorance of the facts of business." The statement was in reference to
the existing overcapacity in the cement industry. Countering, the chair-
person of the Mississippi River Fuel Corporation pointed out that "a very
substantial part of the cement capacity is obsolete and the industry
knows it. That is why it doesn't want anyone to come in." If available
capacity cannot, therefore, be used to threaten potential entrants, entry
may occur. If, on the other hand, production has grown steadily and sub-
stantially, the presence of excess capacity may be inferred to be a signal
of potential harm to entrants and part of a response existing firms have
made to the threat of possible entry.9 It is this study's contention
that potential entrants predominately perceive excess capacity as a
weapon to be used in retaliation and, therefore, exhibited increases in
excess capacity are expected to dampen the rate of entry.

Substantial levels of liquid assets, held by established firms, may

also present a threatening signal to potential entrants. These funds

64

are available for consumption if the need so arises. Specifically,
liquid assets may be consumed to prevent liquidation or bankruptcy
during a price squeeze designed to combat infiltration of the in-
dustry.10 Even in the absence of excess capacity entry may be thwarted
when these assets are used to finance an output surge made by a plant
previously operating at capacity. Furthermore, liquid assets may be
consumed in the financing of a promotional campaign aimed at an entrant;
the introduction of additional temporary and competitive brands that
correspond to the entrant's product, and also changes in an existing

product's quality to counter the entrant's "better" product.
larger the total fund the more prepared are incumbent firms for the
contingency of entry, the more likely entry will be deterred.
Summarizing the a priori expectations of this study, it is antici-
pated that the rate of entry will rise with increases in pre-entry
profits and the rate of growth of the industry demand. It is also
assumed that both the static barriers outlined by Bain and the dynamic
barriers will vary indirectly with the rate of entry. However, it will

be shown that certain entry barriers are ineffective against the most

prominent modes of actual entry, foreign and diversifying firms.

The Established Firm

 

In this section an attempt is made to explain the profitability of
an established firm by constructing a framework which includes the inter-
dependence between the established firm and both the existing and poten-
tial competition. The steady-state value of profits resulting from the
long-run pricing decision of the firm, as formulated in the extended

Gaskins model (equation 3), can be considered to be a function of the

65

current state of competition (CR), the nature of the entry condition
including the effects of structural and strategic barriers (E,CE), and
the price elasticity of demand (GR,A). The other structural barriers

are included to determine if they have an entry independent effect on

12

profitability. Accordingly, we have

(15) “t = nICRt, Gt’ CE, Et’ At, S, C]
where

"t = the actual profits of the dominant firm(s)
in period t,

CR = 4-firm concentration ratio of industry in
current period t,

Gt = growth in industry demand, period t,

CE = interaction term of entry and concentration,

Et = the rate of entry in current period t,

At = advertising intensity of the dominant firms

in period t,

output of a minimum optimal size plant (MOS)

relative to the market size,

C = capital requirements of a MOS plant.

(I)
II

The concentration ratio describes the current market structure and
indicates the probable degree of industry collusion. Oligopoly theory
suggests that the ability of firms to collude (tacitly or overtly) in
order to maintain prices above the long-run average cost of production
is greater in industries in which a few sellers dominate the market.13
Firms in industries characterized by low levels of concentration will
find it difficult to create and maintain a coordinated pricing policy.
It is therefore hypothesized that increases in concentration will de-
crease intra-industry rivalry and enhance industry profitability.

Empirical studies, in general, have substantiated the hypothesized
positive correlation between profitability and concentration. A notable

departure is the work of Brozen which attempts to prove that the direct

association is a phenomenon of disequilibrium.14 Nonetheless, the

66

expected result for this study, based on overwhelming theoretical and
empirical evidence, is that higher concentration levels will promote
higher profit rates.

Growth in industry demand, as has been described, will tend to
lower the limit profit rate. Potentially, therefore, profits established
within a fast growing industry may be reduced. However this indirect
effect of growth on profitability will be captured by the rate of entry
variable.

Profitability may also be directly affected by the rate of growth
in industry demand. In periods of rapid expansion, profits may rise
either through increases in output sold or through increases in price
depending on whether capacity either approximates or lags, respectively,
the rate of demand. Conversely, growth may destabilize the market and
upset the collusive pattern of activities, reducing the level of profits.
Empirical studies do not provide a clear indication as to which of these
two roles for growth will dominate.15 A recent theoretical model focus-
ing on the dual role of growth lends Support to the former role.l6
Accordingly, higher rates of growth will be expected to elicit higher
rates of profit.

The established firm equation also contains a variable reflecting
the interaction between market structure and the threat of entry. This
variable is designed to explore an implication of the limit pricing
theory, focusing on the premise that prospective entry should have a
larger impact in the more concentrated markets. For example, if a
market were perfectly competitive, potential entry would not be expected
to have any influence on price. In a highly concentrated market, how-

ever, the interdependence between the established firm and the potential

67

firm should be significant enough to constrain current pricing power.
The limit pricing hypothesis suggests that the interaction term and
profitability will be inversely related.

As described in the previous chapter, static entry barriers may
influence actual profits in two ways. First, the entry deterring ef-
fects of entry barriers will increase the level of profits that could be
earned while still forestalling entry. Second, entry barriers could af-
fect profitability directly, independent of their effects on entry;
e.g., advertising intensity can expand the demand for the product of an
established firm thereby increasing the firm's profits. The first of
these two roles, analyzed in the preceding section, explained that the
effectiveness of barriers in impeding entry would be reflected in the
rate of entry variable. An analysis of the direct impact of entry bar-
riers on profitability remains.

In particular, the direct impact of advertising on profitability
is unclear. Profitability may be increased as advertising expenditures
increase if those expenditures effectively induce more customers to
purchase the firm's product at every available price. On the other
hand, if the information provided by advertising extends the effective
range of product choice, price and profit may decrease as advertising
increases. For example, Benham found that in states which permitted
advertising of eyeglasses, the average price of eyeglasses was less than

17 Because

half the price exhibited in states restricting advertising.
there are no compelling theories regarding the direct impact of adver-

tising, insight into its primary function must be sought empirically.

68

The direct impact on profits of the other barriers employed in the
profitability equation; S-scale economies and C-capital requirements is
equally unclear. In fact, not a single hypothesis concerning the ef-
fects of these barriers independent of their entry impeding influences
was discovered. An empirical approach to this issue will attempt to
yield more satisfying conclusions.

The complete theoretical model, which describes both the behavior
of established and potential firms, may be summarized by restating the
principal equations and reviewing the anticipated effects of the various

components,

(15) wt = nlCRt. 6:. CE. E . A: S C] (The established firm)
(+) (+) (- ) (- t) (?) (?)(?)

(16) Et E["t-1’ Gt’ At’ 3. C: EXC, LIQ] (The potential firm)
(+) (+) (-) (-)(

-) (-) (-)
where

EXC = excess capacity,
LIQ = holdings of liquid assets.

NOTES

CHAPTER THREE

1Gaskins, "Dynamic Limit Pricing."

2Other modifications in the rate of entry equation may be used
which result in identical qualitative implications. For example, a
general formulation could find the response cpfifficient, k, to be a
function of I such that: dx=k(1)(P‘P7 where-§~<0. For other examples
of entry reducing policies formulated via Gaskins' model, see Stephen
Martin, "Gaskins Models of Entry Impedance," Econometrics Workshop
Paper #7902, Michigan State University, August 1979.

3This conclusion is based on the interpretation of 2 as the
shadow price of an additional unit of rival entry at any point in time.
Refer to Gaskins, "Dynamic Limit Pricing," p. 308. A probablistic
approach to the entry deterring effects of investment into product
differentiation which derived comparable conclusions may be seen in
Spence, "Entry, Capacity."

4Gaskins, "Dynamic Limit Pricing," p. 309.

5Even though the previous discussions of entry centered on limit
pricing, it is justifiable to substitute pre-entry profits for pre-entry
price and limit profits for limit price, since the qualitative effects
are very similar. In addition it has been argued that a limit rate of
profit is more appropriate than a limit price, Caves and Porter, "Mobil-
ity Barriers." Furthermore, this substitution appears to be an accepted
practice. Refer to Appendix 2A for substantiating evidence.

6The condition of entry, as defined by Bain, is the premium
accruing to established firms which charge a price, PL, higher than
the competitive price, PC, but not high enough to induce entry. This
profit margin may be analytically expressed as EC=(P -PC)/PC which may
then be transformed to an expression explaining the Iimit price:
PL=PC(1+EC), where EC is the condition of entry. It is important to
remember that the limit price (limit profit rate) so defined is derived
using "Sylos Postulate." The reactions to entry by established firms
are analyzed separately later in the text.

7Ireland, "Concentration and the Growth," and Bhagwati,
"Oligopoly Theory."

8Wall Street Journal, May 27, 1963, p. 7.

69

70

9Reference the previously cited Alcoa Case.

10Price squeezes may be more effective deterrents to entry than
an alternative strategy of buying up new entrants. Entrants in the
former case face possible ruin but in the latter case an entrant firm
remains an entity, only now it is under new ownership. The strategy
most apt to disable the entrant is the strategy most apt to deter the
entrant.

11As an example, refer to J.A. Menge, "Style Change Costs as a
Market Weapon," Qparterly Journal of Economics 76(November 1972):
632-647.

12Initial analysis of the model also included the strategic bar-
riers to entry in the profitability equation, testing whether these
variables had an entry independent influence on profitability. However,
the strategic barriers failed to exhibit statistically significant ex-
planatory power, and were subsequently excluded.

 

13Stigler, "A Theory of Oligopoly," and Weiss, "The Concentration-
Profits Relationship."

14Yale Brozen, "Concentration and Structural and Market Dis-
equilibrium," Antitrust Bulletin 16(Summer 1971): 241-248.

15For further explanation and empirical analysis of these hypo-
theses, see William Shepherd, "Trends of Concentration in American
Manufacturing Industries, 1947-1958," Review of Economics and Statistics
46(May 1964): 200-212 and Stephen Martin, "Entry Barriers, Concentration,
and Profits," Southern Economic Journal 46(October 1979): 471-488.

 

 

16Takeo Nakao, "Demand Growth, Profitability, and Entry,"
Quarterly Journal of Economics 94(March 1980): 397-411.

 

17Lee Benham, "The Effect of Advertising on the Price of Eye-
glasses," Journal of Law and Economics 15(October 1972): 337-352.

 

CHAPTER FOUR

Strategic Barriers To Entry

In this chapter an empirical specification is presented, stipu-
lating the interactions between prospective and existing producers as
conceptualized in the preceding chapter. Attention primarily centers
on how potential entrants react to specific investment strategies under-
taken by current producers. Do potential entrants limit their penetra-
tion into a market where established firms hold reserves of productive
capacity and/or liquid assets, perceiving these investments to be de-
vices for retaliation? Concurrently the chapter will focus on the role
static industry conditions play in the entry decisions of potential
entrants. Subsequent to this analysis, the varying influences that both
strategic and static entry barriers exert on specific classes of en-

trants will be scrutinized.

DESCRIPTION OF VARIABLES

 

.Eppgy: Following the earlier discussions of the economic definition
of entry, the rate of entry variable was measured as the estimated mar-
ket share of de novo, diversifying, and foreign entrants. Separate
calculations were made for domestic and foreign firms.

Data for domestic entry was derived primarily from the Fortune

publication Plant and Product Directopy.1 The Directogy consists of

 

survey data identifying the plants of the largest 1000 U.S. industrial

corporations, and the 5-digit SIC products of those plants. The

71

72

initial step in finding entrants is to compare operating plants from one
edition of the Directory to the next. If a new plant under a new com-
pany name appears in an industry, entry is concluded to have occurred.
These "entrants" are then screened against other sources to insure that
the new firm is indeed "new", not merged, and not a firm that has grown
so as to be included in the top 1000 firms.2 The extent of entry was
calculated by taking the employment size of the entrants' plant(s) from
the Directory and attributing to that plant the output of a comparably

sized Census plant.3

Dividing this output level by the total value of
shipments in the industry generates the percentage market share of the
entrants. A variety of other sources were consulted to identify other
significant cases of entry and to validate the year entry was actually
realized.4
Foreign entry market share was derived from Census tabulations
which display imports as a percentage of industry output.5 The specific
market share was calculated by subtracting the 1959 share of imports to
total industry output from the 1968 import share. Revisions in this
calculation were made when the 1968 value of import share was an
anomaly -- not persisting for more than a year. In these instances,
the last persistent level of import share was substituted for the 1968
value.6
Measuring foreign entry as the market share of imports suffers two
shortcomings. Imports need not emanate only from foreign firms. Many
times imports are transfers from an overseas subsidiary operation to its

U.S. based mother firm. This category of imports is an element of cur-

rent market structure and as such cannot be expected to neutralize

73

market power. Using the import share as the measure of foreign entry
also omits direct foreign investment, the building of new capacity in
the U.S. by a foreign firm. The output from these plants would be in-
cluded in domestic output causing the foreign entry variable to be
understated. Despite these omissions, the measurement of foreign entry
based on imports should capture the main source of foreign firm penetra-
tion into domestic markets.

Using the described procedures, new capacity entry data was amassed
for 40 4-digit manufacturing industries over the years 1959-1968. Since
the limit pricing model on which this study is based is an oligopoly
pricing model, the sample industries, in principle, should be oligopolies.
The overwhelming majority of industries selected do indeed fit into this
category.7 However, due to the limited availability of entry data, par-
ticular non-oligopolistic industries were added to the sample. These
additional industries are manufacturing industries previously analyzed
by Bain or Mann therefore permitting comparison of the results of this
study to well-known profit-structure studies.

The time frame, 1959-1968, is important for a number of reasons.

To capture the effects of the long-run strategy of limit pricing re-
quires the use of an extended period of time, of which ten years should
be appropriate. To lessen the likelihood that any relationships arise
due to a disequilibrium phenomenon, analysis must be done over a time
interval sufficient for appreciable structural changes to occur. The
particular years chosen were selected to minimize the problems associ-

ated with the merger wave of the late 1960's.8

74

Profitability: The profit rates used in this study are measured by

 

net profit after taxes as a percentage of stockholders equity. Although
subject to the vagaries and discretionary allocations permitted by ac-
counting practices, this measure is deemed appropriate for both theoreti-
cal and empirical reasons. Because stockholders have voting rights and
assume the main risk of the firm the return on equity represents the
profits to the true ownership of the firm. Empirical justification for
this measure was found by Hall and Weiss.9 Each rate is the average
dominant firm yearly profit -- the first denoted as PR was calculated

for the pre-entry period, the second denoted as P was calculated over

the entire 1959-1968 period.10

Profitability of dominant firms was used in order to capture the
important consequences of entry. If entry substantially effects indus-
try performance, that effect must be felt by the leading firms within
the industry since these are the firms most apt to hold and utilize
market power.

However, reported dominant firm profits may not be an accurate in-
dicator of potential returns to be reaped by entrants. The reported
profit rates may be influenced by the existence of technical or x-
inefficiences. These excess costs may reduce the reported profits but
not the profits an efficient producer may claim. In such cases reported
dominant firm profits may play a minute or "perverse" role in the entry
decisions of potential entrants.

Moreover the greater the dominant firm profitability, the greater
may be the risk of entry. As hypothesized by Stonebraker, "most entry

occurs on a small scale and entrepreneurs are likely to estimate the

75

risk of entering an industry on the basis of the performance of existing

"11 Using the percent of small firms with profit rates be-

small firms.
low the competitive rate and the extent to which these rates fall below
the competitive rate as a measure of small firm risk, Stonebraker finds
that small firm risk is indirectly related to dominant firm profit-
ability. In other words, the worse the experience of small firms, the
greater the risk of entry, and the higher the profits of dominant firms.
Consequently, the rate of entry may decline as dominant firm profit-

ability increases.

Demand Growth: The measure of an industry's growth rate of demand

 

is the average yearly percentage rate of growth in the industry's value

of shipments from 1958-1967.12

Concentration: Concentration ratios were taken from Shepherd's

13

 

adjusted 4-firm concentration ratios estimated for the year 1966.
Certain ratios did not necessitate altering, in which case the unad-
justed Census ratios for 1963 were used. In that the adjustments made
to the 1966 concentration ratios would apply equally well to the 1963
ratios, the utilization of ratios from different years is not incon-

sistent.

Static Barriers to Entry

 

Economies of Scale: The output of a minimum optimal size (MOS)

 

plant relative to the total industry output was used to explain the
extent of economies of scale existing within an industry.14
An alternative measure of scale economies, also utilized in this

study, arises from the interaction of the market share of a MOS plant

with a cost disadvantage ratio (CDR), calculating the CDR as the value

76

added per worker in the smallest plants producing 50% of industry out-
put divided by the value added per worker in the largest plants account-
ing for the remaining 50% of industry output. The interaction reflects
the fact that with modest cost disadvantages, suboptimal entry may be
rational, and scale barriers different than that implied by the MOS
variable alone. In this case, scale economies were defined as MOS when
CDR < .90; MOS = 0 when CDR: .90.15

Capital Requirements: Capital requirements were measured as the

 

capital investment needed to finance a MOS plant -- the output share of
a MOS plant multiplied by the book value of gross depreciable assets for
the industry.16
Although this measure is often used in empirical work, additional
costs may be part of the capital requirements facing a potential en-
trant. For example, if book value assets are less than replacement
costs, then the funds needed to build productive capacity are under-
stated. This calculation is further understated since it excluded the
funds an entrant may need to withstand a price or an advertising war.
Furthermore, the capital required to set up distribution channels was
also omitted from this measure. Consequently, any entry deterring ef-
fects that may be statistically attributed to the capital requirements

variable are probably understatements of its true impact.

Advertising: Advertising intensity is used as a proxy for the degree

 

of product differentiation in the industry. In this study, advertising
intensity was calculated as the average annual advertising sales-ratio
of the dominant firms in the industry as stated in the August issues of

Advertising Age for the appropriate years. When advertising figures for

 

77

the dominant firms were unavailable, the industry advertising-sales

ratio was used.17

Aggregate Entry Barrier Classifications: Although aggregate entry

 

barrier classifications are necessarily judgmental, they were utilized
as a point of departure from and expansion on earlier examinations of
entry conditions and their impact on industry performance. Industries
18

were classified according to demarcations drawn by Shepherd.

Trade Barriers: A dummy variable was used to categorize the extent

 

of trade restraint. Both the height of the nominal tariff rate and the
presence of non-tariff barriers were considered in constructing this
variable.19

The trade barrier was included to improve the specifications of the
empirical model by accounting for factors affecting foreign entrants.
Interestingly, trade barriers may also affect the rate of domestic entry.
If trade protection is provided for declining industries, protection may
perpetuate the use of inefficient capital resources in the specified in-
dustry. Consequently, trade barriers may promote entry by a producer
introducing new and efficient technology. Trade barriers may also pro—
tect infant domestic industries. In turn, as the domestic industry grows
and matures, domestic entry may be anticipated. Based on these asser-

tions, a positive correlation between trade barriers and domestic entry

is expected.

Strategic Barriers to Entry20

 

Liquidity: The holding of liquid assets by established firms was
described in two ways: 1) the ratio of liquid capital to total capital,

and 2) the absolute value of liquid capital. The first value is

78

computed as the average ratio of current assets to total assets for the
dominant firms in each industry. The second value of liquidity is the
dominant firms' average holdings of current assets. Both measures are

the average yearly values found during the years 1959-1968. The second
measure of liquidity most accurately reflects the level of funds avail-
able for consumption; however, it is concurrently related to absolute firm
size. The former normalized variable purges the liquidity measure of

this discrepancy.

Excess Capacity: One excess capacity variable was generated using

 

McGraw-Hill capacity data.21 Excess capacity was computed as the aver-
age annual difference between the preferred operating rate of plant
utilization and the current operating rate over the Span of three years,
1964-1966. These particular years were chosen because they occur at or
near the peak of a business cycle. Observations of excess capacity
during a boom period should indicate the policies of the established
firms in maintaining reserve productive capacity. Because the McGraw-
Hill data are available only for 3-digit industries and the industries
examined in this study are at the 4-digit level, it is assumed that the
excess capacity held by the 4-digit industries comprising each 3-digit
industry exhibit similar excess capacity holdings.

Another proxy for excess capacity was calculated as the ratio of
the rate of growth of fixed assets to the rate of growth of sales for
the dominant firms in each industry. This excess capacity proxy was
expressed as a dummy variable, taking the value one when asset growth
is greater than sales growth. This variable necessarily implies that

the output/asset ratio is constant in each industry. However, if

79

technological advances occurred during this time Span, the output/
asset ratio may decline and the measurement of excess capacity would
be understated. On the other hand, if investments were made into non-
productive assets, e.g. pollution abatement equipment, the excess capa-
city variable would overstate the actual extent of idle capacity. In
that the latter circumstances are deemed insignificant, the rate of
entry should be indirectly related to increases in the excess capcity
variable.

A list of the variables and their respective acronyms is contained
in Table 4.1. Table 4.2 gives a statistical description of the vari-

ables.

ENTRY AND STRATEGIC ENTRY BARRIERS

 

The purpose of this section is to emphasize that market structure can
no longer be considered an unalterable condition but must rather be per-
ceived as a set of conditions which are open to manipulation by the
firms in that market. Firms can be expected to actively participate in
programs, which entail incurring additional present costs, in attempt to
change or sustain conditions of its market such that long-run returns to
the firms are enhanced.

In this context, the empirical analysis of this study focuses on
those investments made by established firms which both signal and permit
aggressive action towards possible entrants. By ignoring this feature
in previous studies, Caves and Porter argue that

"a modicum of determinancy has been bought
at an exhorbitant price in foregone under-
standing of influences on newcomer's de-
cisions to enter and of the selection and

use by going firms of devices to discourage
entry." 2

80

 

TABLE 4.1: VARIABLE LIST

 

ED
EF

PR

SC

CC

EXC

LQ

LIQ
HB
SB
CR
TB

CE

market share of all entrants
market share of domestic entrants
market share of foreign entrants

the average pre-entry rate of return on equity for the
dominant firms in an industry

the average rate of return on equity for the dominant firms
in an industry

industry growth rate of demand
advertising-sales ratio of dominant firms in an industry
share of industry output of a MOS plant

share of industry output of a MOS plant adjusted by the
cost disadvantage ratio

absolute capital requirements of a MOS plant

absolute capital requirements compensating for the possible
operation of a suboptimal size plant

static excess capacity (McGraw-Hill) in the industry

dynamic excess capacity (output/asset ratio) of the
dominant firms

ratio of current assets to total assets held by the
dominant firms

quantity of current assets held by dominant firms
very high entry barrier category

substantial entry barrier category

4-firm concentration ratio

trade barrier dummy variable

concentration-entry interaction variable

 

81

 

 

 

 

TABLE 4.2: DESCRIPTION OF VARIABLES
Variable Mean Standard Deviation Minimum Maximum
E 2.34% 2.15 0 7.82
ED 1.14% 1.66 0 6.90
EF 1.20% 1.87 0 6.00
P 11.93% 3.80 5.70 19.79
PR- 11.49% 3.58 5.31 19.79
G 6.03% 19.40 -0.96 11.35
A 3.29% 3.98 0.23 17.00
S 5.79% 7.49 0.11 37.97
SC 2.87% 4.84 0 22.46
C $30.46m 37.77 0.43 127.38
CC $11.95m 21.20 0 66.02
EX 2.11% 1.43 -0.90 4.40
EXC 0.48 0.51 0 1
L0 0.44 0.18 0.15 0.86
LIQ $397.05m 558.00 24 2512
HB 0.38 0.49 0 1
SB 0.35 0.48 0 1
OR 73.7% 17.70 35 98
TB 0.40 0.50 0 1

 

82

The particular devices, or as named here strategic barriers, under
evaluation in this study are holdings of excess productive capacity and
holdings of liquid asset reserves. It is hypothesized that the main-
tenance of either or both of these strategic barriers portrays to a
potential entrant that the established firm is willing and prepared to
battle for the market share under attack. Consequently it is assumed
that increases in excess capacity and liquidity will be found to reduce

the rate of entry.

Estimation

 

The model employed for analyzing prospective and current producers'

behavior is a simultaneous equation model, described by equations (4.1)

and (4.2):
(4.1) Et = a0+a1Pr+a2Gt+a3(EB)t+a4(SB)t+a5(TB)+Ut
where
EB = a vector of static or structural entry barriers
(3. SC. C. CC. A);
SB = a vector of strategic entry barriers (EX, EXC,
LQ. LIQ);
Ut = disturbance term of the regression;
all other variables as previously defined.
(4.2) P = b0+b1Gt+b2(EB)t+b3(CR)t+b4(CE)t+b5Et+Vt
where

v = disturbance term of the regression;
all other variables as previously defined.

The particular nature of this model is such that the behavioral equa-
tions exhibit a unilateral cause and effect relationship. Entry in the
current period (Et) is dependent on the profitability of established
dominant firms in a prior period (Pr). The current level of penetration

by new firms (Et) influence the current profitability of the existing

83

firms (P). Clearly the causal chain runs from Pt_1(PR)-§Et-ith(P).
Due to the recursive nature of the model, ordinary least squares would
normally be used to estimate the parameters of each equation, those
estimates being unbiased and consistent. This result necessitates that
the system of equations have only exogenous independent variables.

In this instance, it may be argued that concentration and adver-
tising intensity are endogenous to a complete system of equations ex-

plaining industrial structure, conduct, and performance.23

By the same
reasoning, the strategic barriers should be considered endogenous since
these variables reflect the results of decisions which are made by firms
within the system; they are not given by technical production conditions.
This analysis, accordingly, considers each of these variables to be
endogenous except for concentration. Previous structural equations
utilized to explain concentration appear to be nothing more than a dis-
guised entry equation.24 These equations have expressed concentration
to be a function primarily of structural entry barriers and growth.
Consequently this study imposes the entry equation (4.1) as an improve-
ment over a concentration equation. For those who would argue that
entry should be listed as an independent variable in a concentration
equation to show the effects of entry in reducing concentration, this
study adopts an interaction term between concentration and entry (CE)
which captures that possible relationship. Even with concentration
considered exogenous, the existence of other endogenous variables in

the system requires that an instrumental variables technique be used

for consistent estimation.

84

The set of instruments include all exogenous and predetermined
variables from both the entry and profitability equation. In addition,
the squares of each of these non-dummy instruments are contained in the
utilized set of instrumental variables.25

Although the estimation of the structural equations by ordinary
least squares (OLS) will result in biased and inconsistent parameter'
estimates, it has been found that OLS is more robust against specifi-
cation errors than the instrumental variables technique.26 In view of
this, OLS estimates of the structural equations explaining entry and

profitability are reported in Appendix 43 and Appendix 5A, respectively,

for use as a standard of comparison.

EMPIRICAL RESULTS AND INTERPRETATION

 

The parameters of the rate of entry equation are estimated and
analyzed from two perspectives.27 First, the effect of entry conditions,
in particular Strategic barriers, on the total entry variable (E) are
presented. Alternative specifications of the Variables are then uti-
lized to test the reliability of the estimated coefficients. In a fol-
lowing section, regressions are performed which explain, separately,
the domestic (ED) and foreign (EF) rate of entry; indicating the via-

bility of treating entry as a homogeneous process.

Total Entry

 

Table 4.3 presents estimates obtained before strategic barriers are
introduced into the analysis. The independent variables most comparable
to previous studies are those reported in equations (1) and (2). With
strategic barriers excluded, these estimates are similar to those re—

ported by Harris, except that they 1) refer to oligopolistic industries

85

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 4.3: REGRESSION ANALYSIS OF TRADITIONAL FACTORS DETERMINING THE ENTRY CONDITION
Independent Regressions3
Variables (1) (2) (3) 151 (M) g (2A) (3A) (4A)
Constant 5.24 4.77 5.25 4.86 5.14 4.17 5.21 4.60
c 0e63 0029 0.70 0051 0025 -Oeoog Oe‘7 0.37
(4.22) 410.49) 44(1.28) (0.84) (0.§§) (0.21) (0.99) (0.73)
A -0.15* -0.17* -0.13* -0.10* -0.56*** -0.62**‘ -0.48*** -0.42***
(1.68) (1.64) (1.51) (1.50)yy7 (4.32) (4.93), (3.78) (2.97)
(1.22) (1.81) (1.36) (1.88)
SC -—- -- 40.06 -0.13** --- --- —0.04 -0.08*
(0.83) (1.40) (0.60) 4(1.35)
C 0.003 -0.0009 --- --- -0.006 -0.01** --- ---
(0.29) (0.10) (0.89) (1.74)
CC --- --- 0.004 0.01 —-- -- -0.007 0.002
(0.25) (0.84) (0.49) (L16)
TB -0.96* -1.10* -0.78 -0.64 -1.49*** -1.15*‘* -1.16** -0.92*
(1.42) (1.55) (1.17) (0.94) (2.58) (3.00) (1.97) (1.56)
PR -0.19 -0.09 -0.21** -0.19** 0.03 0.15* -0.06 -0.04
(1.26) (0.76) (2.08) (1.76) (0.29) (1.51) (0.66) (0.42)
N 40 27 40 27 40 27 40 27
1) The figures in parentheses are t-ststistics.
2) Based on one-tailed tests: * 102 significance; ** 52 significance; *** 1% significance.
3) Equations (1-4) estimated by ordinary least squares; equations (1A-4A) estimated by
instrumental variables technique.
4) The difference in number of observations results from incomplete excess data being

available from McGrav-Hill.

86

rather than a sample of industries covering the entire spectrum of in-
dustries from competitive to and including oligopolies, 2) include a
measure of trade barriers, and 3) indicate a statistically insignificant
role for pre-entry profitability.28

The introduction of a scale variable (SC), which captures the slope
of the entrant's prospective average total cost curve, results in a
negative correlation between entry and pre-entry profitability. However
in those equations which simultaneously use this refined scale variable
and account for the endogeneity of advertising (eqs. 3A and 4A), the
results are again quite comparable to those found by Harris.

Table 4.4 presents the first in a series of specifications in which
the various combinations of the strategic barriers (EX, EXC, LQ, LIQ)
are introduced. Because the discontinuous scale variable (SC) portrays
most completely the effect economies of scale have on potential entrants,
it is contended that equations utilizing this variable and the corre-
sponding capital requirements variable (CC) represent the most accurate
account of the entry condition and are, therefore, displayed first.

The main development of these regression results is that strategic
barriers control the rate of entry over and beyond the deterrent capa-
bilities of the traditional set of entry barriers. Indeed, the excess
capacity and liquidity variables exhibited significant negative coef-
ficients in every specification used, notwithstanding the variable LQ.
By using actual entry data, these results show that the strategic in-
vestment decisions of existing firms do affect actual entry decisions
of potential entrants. Accordingly, the theory of limit pricing can

no longer adopt Sylos Postulate and be considered correct. The model

87

 

TABLE 4.4: INSTRUMENTAL VARIABLES ESTIMATION OF TOTAL ENTRY
UTILIZING SPECIFIC ENTRY BARRIERS (SCALE VARIABLE

 

 

 

 

 

 

 

 

 

 

 

 

 

DISCONTINUOUS)
Independent Regressions
Variables (1) (2) (3) (4)
Constant 5.30 7.18 7.17 7.13
C -0.65 -0.54 -0.29 -0.56
(1.04) (1.03) (0.68) (1.24)
A -0.50*** -0.45*** -0.45*** -0.47***
(3.61) (3.93) (4.05) (4.84)
SC -0.11** -0.11** -0.08* -0.05
(1.85) (2.19) (1.33) (1.10)
CC 0.01 0.01 0.002 -0.006
(0.93) (1.19) (0.14) (0.49)
EX -1.01** -0.84** --- ---
(2.42) (2.51)
EXC --- --- -3.74*** -6.13***
(4.33) (4.59)
LQ 10.42* --- -3.17 ---
(1.56) (0.58)
TB -1.44** -1.36*** -1.44*** -1.68***
(2.45) (2.59) (2.83) (3.60)
PR -0.002 0.02 0.03 0.05
(0.02) (0.26) (0.39) (0.58)
N 27 27 40 40

 

1) The figures in parentheses are t-statistics.

2) Based on one-tail tests:

** 5% significance; *** 1% significance.

* indicates 10% significance;

88

must be expanded to explicitly account for other reactions by the estab-
lished firms facing the threat of potential entry.

Including the strategic barriers in the equation has not altered
the character of the coefficients of the other determinants of entry
found in the estimations omitting the strategic barriers. Each of the
significant variables (A, SC, TB) affect the rate of entry as suggested
by the limit pricing model.29

To examine the sensitivity of the estimated coefficients to alter-
native specifications, we proceed by considering a specification utiliz-

ing the more common continuous scale variable.30

The major problem with
this measurement of scale economies is that it ignores the possibility
that entry at a non-optimal scale may result in minimally increased pro-
duction costs. Consequently, the use of this scale variable will exag-
gerate the economies of scale barrier in industries with relatively flat
per unit cost curves. Table 4.5 displays the results using the respeci-
fication incorporating the scale variable (S) and the corresponding capi-
tal requirements variable (C).

Once again, excess capacity was found to reduce the rate of entry.
These results support the premise that current producers may not rely
solely on the traditional barriers to entry but are consciously invest-
ing resources into excess capacity. Acknowledging the potential of such
investments, prospective entrants have been turned back.

Although excess capacity variables persistently repel potential
entrants, liquidity variables portray a less distinct role. Absolute
quantities of current assets (LIQ) continues to deter entry, but less

significantly than in the initial specification. On the other hand,

89

 

TABLE 4.5: INSTRUMENTAL VARIABLES ESTIMATION OF TOTAL ENTRY
UTILIZING SPECIFIC ENTRY BARRIERS (SCALE VARIABLE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

CONTINUOUS)
Independent Regressions
Variables (1) (2) (3) (4)
Constant 3.24 8.79 6.21 6.87
G -0.65 -0.84* -0.51 -0.59
(1.25) (1.37) (1.28) (1.24)
A -0.72*** -0.38*** -0.45*** -O.44***
(5.98) (3.30) (3.79) (4.04)
S -0.10*** -0.14*** -0.09*** -0.06*
(3.11) (4.08) (2.31) (1.47)
C -0.005 +0.04** 0.02** 0.01
(0.61) (2.50) (2.03) (1.11)
EX -0.67** -1.23*** --- ---
(1.88) (3.13)
EXC --- --- -4.57*** -5.29***
(4.45) (3.69)
LQ 14.84* --- 2,49 ---
(1.65) (0.45)
LIQ --- -0.003*** --- -0.0008
(3.59) (0.71)
TB -1.89*** -1.32*** -1.42*** -1.54***
(3.97) (3.05) (3.06) (3.12)
PR 0.18** -0.02 0.04 0.04
(1.97) (0.18) (0.44) (0.46)
N 27 27 40 40

 

1) The figures in parentheses are t-statistics.

2) Based on one-tail tests:

** 5% significance; *** 1% significance.

* indicates 10% significance;

90

the ratio of current to total assets (LQ) appears to promote entry. This
may occur because liquidity could possibly be an indicator of prospective
profits since the profitable firms are most apt to be able to accumulate
liquid assets.31 More insight into the role of this Variable will be
realized once this study analyzes the separate effects of LQ on domestic
and foreign entry.

The specific entry barriers of advertising intensity, economies of
scale, and trade barriers consistently and significantly repel potential
entrants. The capital requirements barrier, however, is shown to en-
courage entry.32 This result confirms the anticipated measurement
problems associated with this variable.

Continuing to check the sensitivity of the estimated coefficients
to alternative variable specifications, estimates are made substituting
aggregate barrier classifications for the Specific barriers used in the

prior specifications.33

Table 4.6 presents those results.

Each of the strategic deterrents display negative coefficients but
are only significant in the larger sample of 40 industries. Neverthe-
less, these results lend support to the hypothesis that strategic invest-
ments are relevant entry barriers.

The coefficients on the entry barrier aggregates displayed the cor-
rect signs; but inconsistencies remain. In half of the regressions the
substantial barrier category restricted the rate of entry more than the
very high barrier class did. These judgmental categories have tended
to emerge as amalgams of specific influences which are not easy to formu-

late or measure. Correct composite variables are difficult to produce

since individual barriers are each relevant to certain industries and

91

 

TABLE 4.6: INSTRUMENTAL VARIABLES ESTIMATION OF TOTAL ENTRY
UTILIZING AGGREGATE BARRIER CLASSIFICATIONS

 

 

 

 

 

 

 

 

 

 

 

 

Independent Regressions
Variables (1) (2) (3) (4)
Constant 7.17 6.93 9.80 7.89
G 0.23 -0.08 0.14 -0.09
(0.26) (0.10) (0.28) (0.19)
HB -0.43 0.08 -0.92* -1.23**
(0.40) (0.08) (1.39) (1.72)
SB -0.95 -0.87 -1.04* -1.06*
(1.03) (0.95) (1.48) (1.52)
EX -0.40 -0.61 --- ---
(0.69) (1.12)
EXC --- --- -3.53*** -6.67***
(3.47) (3.95)
LQ -4.78 --- -13.16*** ---
(0.65) (2.51)
LIQ --- -0.0008 --- -0.003***
(0.98) (2.57)
TB -0.66 -0.77 -0.90* -1.08**
(0.77) (0.90) (1.53) (1.81)
PR -0.21** -0.19** -0.17** -0.17**
(2.04) (1.91) (1.89) (2.00)
N 27 27 40 40

 

1) The figures in parentheses are t-statistics.

2) Based on one-tail tests:

* indicates 10% significance;

** indicates 5% significance; *** indicates 1% significance.

92

it is difficult to judge which barrier(s) must be given the greatest
weight.

The influence of profitability on the rate of entry runs counter
to the theoretical role implied for it. Normally, one would expect
domestic firms to invest resources in the most lucrative markets avail-
able and respond positively to profit conditions. However, reported
empirical results indicate a negative correlation between the rate of
entry and profitability. This result must be interpreted with caution
because the simple correlation between profitability and the HB dummy
is .71.34 As a consequence, the negative coefficient on profitability
is quite sensitive to model and sample specification.

This section has considered the hypothesis that strategic invest-
ments of established firms, such as excess capacity and liquidity,
represent the tools used to retaliate against new competitors, and that
potential entrants recognizing this capability are indeed deterred. The
evidence assembled in this section is consistent with this hypothesis.
Statistical tests demonstrated that excess capacity and liquidity reduce
the rate of entry. These results imply that effective entry barriers
are not only given industry conditions but also devices or strategies
which attempt and are able to alter those conditions. Therefore, pre-
vious studies which have posited a static industry structure must
clearly be extended to include the active role established firms pursue

in order to shape an economic environment suitable to the firms' goals.

Domestic-Foreign Entry

 

Statistical studies investigating the determinants of entry have

35

usually assumed entry to be a homogeneous process. However, there

93

are reasons to believe that market characteristics and strategic deter-
rences will affect various types of entrants differently. If policy is
to be made to encourage entry, knowledge of entry inhibiting factors
must be gained. And, if different factors affect different types of
entrants, policy makers must take this into account and react
accordingly.

Using a unique approach to a determinants of entry study, this
section divides the extent of actual entry into two distinct classes:

1) actual domestic entry, and 2) actual foreign entry. In so doing,
this study is able to show that different factors are pertinent to the
entry decisions of the different classes of entrants.

The perennial focus of the entry literature is entry by a new firm.
Actual entry data indicates that new firm entrants play a very small
role in the process of entry. Rather, diversifying firms represent the
greatest number of cases. Firms previously established in other markets,
both foreign and domestic, may be less disadvantaged facing traditional
entry barriers and less vulnerable to strategic barriers than de novo
entrants. Existing firms may have ready-made productive facilities, and
in the case of foreign entrants, may only need an outlet for surplus pro-
duction rendering scale and capital requirement barriers inoperative.
Established firms may be able to transfer the goodwill of their other
brands into the entered market negating the advertising barrier. Final-
ly, established firms diversifying into new markets may be able to ef-
fectively finance and therefore compete in any price or advertising war.
Consequently, retaliatory capabilities of established firms become

impotent.

94

Not only may the effectiveness of any entry barrier or class of
barriers be different for established firm entrants than new firms en-
trants, they may also deviate from one type of established firm entrant
to another. A foreign firm may have a comparative advantage in the pro-
duction of a good and thus be less threatened by scale economies than a
domestic firm. On the other hand, a foreign entrant may lack expertise
in promotional methods in this a new and strange market and consequently
be more prone to be deterred by advertising intensity than the domestic
firm. In addition, the foreign firm faces barriers other than the con-
tingent faced by domestic firms. These include tariffs, quotas, and
other trade barriers. When entrants are separated into two groups,
domestic and foreign, contrasting relationships may be found. Conse-
quently, significant relationships may be amplified, masked, or diluted
when entry is considered to be a homegeneous process.

Independent domestic and foreign rate of entry equations are esti-
mated so that the factors affecting the entry decisions of each may be
discerned. Table 4.7 presents the results when the discontinuous scale
variable (SC) is used; Table 4.8 presents the estimated coefficients when
the less accurate scale variable (S) is used. As expected, the diversi-
fying entrants are insensitive to particular entry barriers. Moreover,
domestic and foreign entrants are able to overcome different barriers.

Foreign firm entrants are less likely to enter markets where pro-
ducts are highly promoted as opposed to markets with low advertising
expenditures. This result was robust; continuing in all Specifications
of the model. A problem inherent in the advertising market for a

foreign entity is the lack of knowledge of the best methods of promotion

 

£75

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 4.7: INSTRUMENTAL VARIABLES rsrrusrxos 0r DOMESTIC 4x0 roarxcs ENTRY
UTILIZING A 01sc0wrrnvous scar: VARIABLE
Independent Regressions and Dependent Variable
Variables 20(1) EF(1) r0(2)1 EF(2)_, 20(3) rr(3) ED(4) rr(4)
Constant 4.97 -0.25 4.70 2.48 7.66 -0.48 4.77 2.35
c 0.18 -0.31 40.92 0.39 -0.55 0.26 -o.95 0.38
(0.30) (0.38) (1.04) (0.44) (1.26) (0.49) (;,27) ((0.68)
A -0.13 -0.34*9 -0.28* -0.17** -0.02 -0.46*** -0.12* -0.35***
(0.78) (1.89) (1.39) (1.89) (0.20) 4(3.34) (1.33) (2.62)
50 -0.09* -0.11 -0.05** -0.06 ' -0.34 -0.04 -0.06* 0.03
(4.34) (1.25) (1.75) (0.75) (0.78) (0.58) (1.39) (0.45)
cc -0.03** 0.04** -0.003 0.02 -0.13* 0.02 -0.006 0.002
(1.86) (2.07) (0.16) (0.83) (1.34) (0.91) (0.72) (0.1;)
EX 0.25 -0.75* -0.68* -0.16 --- --- --- ---
(0.92) (1.37) (;,62) (1.27)
EXC --- -- --- --- -3.82*** -0.08 -6.73*** 0.60
(5.62) (0.07) (5.57) (0.32)
10 —14.76** 21.03“ --- --- -16.85*** 13.67** --- ---
(1.95) (2.41) (3.89) (2.04)
110 --- --- -0.001** 0.0002 --- —-- -0.0c2*** -0. 305
(2.23) (0.03) (3.04) (0.04)
TB -0.75 -0.45 -1.07« -0.28 -0.89** ~0.53 -1.10*** -0.59
(1.14) (0.58) (1.63) (0.33) (2.34) (0.89) (2.62) (0.89)
PR -0.03 0.006 0.07 -0.05 0.008 0.02 0.05 -0.007
(0.25)), (0.05) (0.68) (0.36) (0.13) (0.24) (0.74) (0.06)
n 27 27 27 27 40 40 40 4c

 

1) The figures in parentheses are t-statistics.

2) Based on one-tail tests: * indicates 101 significance; ** 5% significance; *** 1% significance.

9f)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 4.8: INSTRUMENTAL VARIABLES ESTIMATION OF DOMESTIC AND FOREIGN ENTRY
UTILIZING A CONTINUOUS SCALE VARIABLE
Independent Regressions and Dependent Variable
Variables E0(1) .sr(1) 80(2) £F(2) E0(3) EE(3) E0(4) £P(4)
c -0.52 -0.14 -l.26* 0.42 -o.85 0.34 -1.02 0.43
((0.73) (0.17) (1.67) (0.45) (1.15) (0.69) (1.27) (0.76)
A -0.32** -0.41*t -0.18 -0.21** 0.08 -0.53*** -0.07 -0.37***
(1.81) (2.01) (1.07) (1.84) (0.79) (3.54) (0.67) (2.49)
5 0.003 -0.11** -0.10** -0.04 -0.02* -0.079 -0.04* -o.02
(0.05) (1.92) (2.03) (0.55) (1.48) (1.52) (1.62) (0.38)
c -0.02** 0.01 0.03* 0.007 0.02** 0.003 0.01 -0.0008
(1.81) (1.22) (1.40) (0.23) (2.05) (0 26) (1 27) (0.05)
Ex -O.67* -0.61 -1.07** -o.16 --- --- --- ---
(1.38) (1.11) (1.90) (0.19)
Exc --- --- --- --- -5.05*** 0.47 -6.27*** 0.97
(5.92) (0.37) (4.69) (0.49)
LQ -6.86 21.71** --- --- -14.31*** l6.80** --— ---
(0.87) (2.35) (3.13) (2.42)
LIQ --- --- -0.003** -0.0001 --- --- -0.001* -0.0306
(2.4)); ((0.06) (1.34) (0.39)
TB -1.12* -0.77 -0.90* -0.42 -O.68** -o.75 -0.91** -o.63
(1.61) (0.98) (1.45) (0.46) (1.75) (1.27) (1.98) (0.93)
PR 0.15 0.02 0.03 0.04 -0.03 0.07 0.03 0.01
(1.12)) (0.15) (0.21) (0.22) (0.45) (0.65) (0.37) (0.09)
N 27 27 27 27 40 40 40 40

 

l) The figures in parentheses are t-ststistics.

2) Based on one-tail tests:

* indicates 10% significance; ** 5% significance; *** 1% significance.

97
in a new market. When this ignorance is coupled with the fact that
advertising is a risky investment apt to have little salvage value if
unsuccessful, the derived indirect relationship between advertising and
foreign entry seems quite reasonable.
No other entry barrier, structural or strategic, presented an in-

36 This is not an

surmountable obstacle for the entering foreign firm.
unexpected result given that foreign firms: 1) may have many markets
over which to spread their output thus aiding the firm in overcoming the
scale barrier; 2) may have access to financial markets in their homeland
which may provide provisions which enables the firm to counter the im-
perfections of the U.S. capital markets; and 3) may hold resources to
meet head on any retaliatory programs of established firms.37
One variable, the liquidity ratio (LQ), fosters foreign entry rather
than inhibiting it as expected. Earlier it was stated that a potential
entrant could respond positively to the holdings of liquid assets if the
entrant perceives these assets to be an optimistic sign of future profit
expectations rather than a source of retaliatory capability. The exist-
ing firms in the industry may utilize their market power to accumulate
rents in the form of liquid assets and as such can indicate to a pro-
spective entrant the possible returns from entering that industry as a
new firm accommodated by or accepted into the collusive group of estab-
lished firms. There are risks to this perception, eSpecially if the
assets are accumulated for purposes of retaliation. However, the risks

are less for the firm which is capable of countering such threats. An

established foreign firm entering a domestic market is such a firm.

98

Although diversifying domestic entrants are less disadvantaged and
more likely to become a viable new competitor relative to new firm do—
mestic entrants, the data reveal that diversifying domestic entrants are
not able to overcome entry barriers with the relative ease demonstrated
by foreign entrants.38 In contrast to the relative insensitivity of the
foreign entrants to entry barriers, domestic entrants were deterred by
both of the strategic barriers, and in every combination. It appears
domestic entrants were repelled by the liquidity variables due to the
retaliatory capabilities possessed by incumbent firms holding such as-
sets. These include price warfare, advertising battles, brand prolif-
eration, and so on. Excess capacity, representing potential precipitous
intense intra-industry price competition, also motivated potential do-
mestic entrants to withdraw or relocate their entry securing resources.
The other primary determinants of domestic entry include scale economies,
capital requirements, and trade barriers; all of which reduced the rate
of domestic entry. Scale economies and capital requirements were ex-
pected to exhibit negative coefficients but the latter variable was pre-
dicted to enhance the rate of domestic entry. The relationship between
domestic entry and trade barriers most likely can be attributed to the
use of nominal tariffs in the creation of the trade barrier dummy. Nom-
inal tariffs can, at times, be indicative of the protection afforded an
industry, but a more precise measure would be the effective rate of pro-
tection.39 However, the effective rate of protection was not available
for these 4-digit SIC industries.

The empirical results of this section show marked contrasts in how

foreign and domestic entrants respond to domestic entry conditions. To

99

analyze the statistical significance of these different behavioral pat-
terns of foreign and domestic entrants, a Chow test is performed.40
Utilizing the four pairs of equations in Table 4.7, it was found that
the effects of the entry condition on the two groups of entrants are
significantly different in the large sample regressions (3 and 4).41
In particular, this result indicates that foreign entrants are able to
overcome more of the entry restricting market conditions and strategies
than domestic entrants.42 Such a result is consistent with the hypoth-
esis that there exists a heterogeneous pool of potential entrants.
Moreover the evidence points out that the most favored potential en-
trants, in the pair of classes considered here, are the foreign entrants.
The determinants of the total rate of entry reflect the differing
relative importance of these determinants to the foreign and domestic
rate of entry. The robust deterrent capabilities of the strategic bar-
riers arises from their strong impact on domestic entrants. Similarly,
the negative coefficients of the scale economies barrier and the trade
barrier reflect, primarily, their role in determining domestic entry.
The appearance of advertising intensity as a major entry inhibitor is,

on the other hand, due to its influence in repelling the market penetra-

tion of foreign entrants.

SUMMARY

In this chapter, it was hypothesized that established firms would
not rely solely on current market conditions to repel potential entrants
but that they would react aggressively to the threat of entry by accumu-
lating resources which indicate and permit retaliation. Empirical evi-

dence points out that these strategic investments or barriers did indeed

100

reduce the rate of total entry beyond that which would have occurred due
to the traditional barriers alone. Consequently, it was concluded that

strategic barriers are important elements to be considered in the entry

decision-making process.

The chapter continued by hypothesizing that diversifying entrants,
being less disadvantaged than new entrants, would be able to overcome or
endure certain of the structural and strategic entry barriers. It was
further hypothesized that the impact of entry barriers is not only likely
to be weaker among established firm entrants but also differ in effec-
tiveness according to the type of established firm entrant. Therefore,
it was postulated that different classes of diversifying entrants,
specifically foreign and domestic, would be able to surmount different
entry barriers. Both hypotheses were supported by the empirical work
presented in this chapter. Foreign and domestic entrants responded dif-
ferently to various proposed entry deterrents. Domestic firm entrants
were able to overcome the advertising barrier; foreign firm entrants
were able to hurdle the scale and capital requirements barriers, and the
strategic barriers. A Chow test confirmed the striking difference in

entrant reactions to both structural and strategic entry barriers.

NOTES

CHAPTER FOUR

lFortune Plant and Product Directory of the Largest 1000 U.S.
Industrial Corporations (Fortune, 1961, 1964, 1967).

 

 

2Consulted were Standard and Poor's Register of Corporations,
Directors, and Executives and Moody's Industrial Manuals.

 

 

 

3Census of Manufactures, Bureau of the Census (washington, D.C.:
U.S. Government Printing Office, 1963).

4These include the financial press, trade publications, and
industry studies. For a complete list refer to Appendix 4A.

 

5U.S. CommoditygExports and Imports as Related to Output, 1959—
1968, Department of Commerce.

 

6Most previous studies of foreign entry have used the absolute
level of imports. However it is more appropriate to consider the supply
of imports with respect to changes in domestic prices. This elasticity
may not be represented by actual levels of imports. Rather a change in
imports, as used here, is a better approximation. Note Phillip Turner,
"Import Competition and the Profitability of United Kingdom Manufacturing
Industries," Journal of Industrial Economics 29(December 1980): 155-166.

 

7Reference the listing of oligopolies in David Kamerschen, "An
Empirical Test of Oligopoly Theories," Journal of Political Economy 76
(July/August 1968): 615-634. The appendix offers a list of the sample
industries and selected data for each industry.

8The more mergers that arise, the more involved and probably the
less precise is the search for entrants fitting the economic definition
of entry. Mbreover, as firms become more diversified and are classified
only into their primary industry category, aggregate industry data will
contain more and more irrelevant information for any specific industry.

9Marshall Hall and Leonard Weiss, "Firm Size and Profitability,"
Review of Economics and Statistics 49(November 1967): 310-319.

10The pre-entry period includes all years until the final case of
entry was realized. If no entry occurred P and PR were identical. Domi-
nant firms were chosen on the basis of their sales within the industry;
selected were the top four firms in each industry. The profitability
data are derived primarily from the report of the Federal Trade Commission

 

101

102

on Rates of Return in Selected Manufacturing Industries, (Washington,
D.C.: U.S. Government Printing Office, 1959-1968. Gaps are filled by
the rates of return published in Standard and Poor's Compustat annual
industrial tape (New York: Standard and Poor Corporation, 1979).

 

11R.J. Stonebraker, "Corporate Profits and Risk of Entry," Review
of Economics and Statistics 58(February 1976): 33-39.

 

12Shipments were culled from the Census of Manufactures, 1958 and

 

1967.

13William Shepherd, Market Power and Economic Welfare (New York:
Random House, 1970), appendix table 8, pp. 263-267.

 

14The output of a MOS plant was considered to be the average out-
put per establishment of the largest plants comprising at least 50% of
the total industry output as derived from the 1963 Census of Manufactures.

 

15R.E. Caves, J. Khalilzaden-Shiraz, and M.E. Porter in, "Scale
Economies in Statistical Analyses of Market Power," Review of Economics
and Statistics 57(May 1975): 133-140 introduced this measure to the
literature. Another advantage of using this variable is the reduction
in multi-collinearity between concentration and the static entry bar-
riers.

 

 

16Assets were derived from the Annual Survey of Manufactures,
Bureau of the Census (Washington, D.C.: U.S. Government Printing
Office, 1963).

 

17Sourcebook of Income, 1959-1968, Internal Revenue Service.

 

18Shepherd, Market Power, appendix table 13, pp. 274-281.

19If the nominal tariff rate in an industry was greater than 10%
and/or if the industry was protected by import quotas the dummy variable
took the value 1. The nominal tariff rate was found by dividing the
dollar value of duties collected by the dollar value of imports as re-
ported in Commodity Exports and Imports as Related to Output. Non-
tariff barriers were reported in Ingo Walter, "Non-Tariff Protection
Among Industrial Countries: Some Preliminary Evidence," Economia
Internazionale 25(May 1972): 333-354.

20The data for the strategic barriers is located on the Compustat
tape unless otherwise noted.

 

 

 

21These figures were made available by H. Michael Mann. Previous
use of the figures may be viewed in H. Michael Mann, James Meehan, and
G.A. Ramsey, "Market Structure and Excess Capacity: A Look at Theory
and Some Evidence," Review of Economics and Statistics 61(February 1978):
156-160.

22

 

Caves and Porter, "Mobility Barriers," p. 242.

103

23A complete model was analyzed by Stephen Martin, "Advertising,
Concentration, and Profitability: The Simultaneity Problem," The Bell
Journal of Economics 10(Autumn 1979): 639-647.

 

24Martin, "Advertising, Concentration, and Profitability," pp.
641-642.

25Harry Kelejian, "Two-Stage Least Squares and Econometric Models
Linear in Parameters but Non-Linear in the Endogenous Variables," Journal
of the American Statistical Association 65(June 1971): 373-374.

 

26Lawrence Klein, A Textbook of Econometrics 2nd edition (Englewood
Cliffs, N.J.: Prentice—Hall, Inc., 1974), p. 150.

27The analysis of the profitability equation (4.2) is deferred to
the following chapter.

28Harris used the set of industries studied by Mann and, in so
doing, found profitability to significantly and positively influence the
market share of entrants. Harris' study, "Entry and Barriers to Entry,"
is used as the primary source for comparison since his study is the only
previous work utilizing a correct measure of entry (note Chapter 2, p.
35). It should also be noted that Orr in "The Determinants of Entry,"
using a statistical facsimile for an entry variable, found profitability
to have an insignificant impact on domestic entry.

29The estimated coefficients found using the instrumental vari-
ables technique (IV) differ little from the estimates derived using
ordinary least squares (refer to Appendix 4B). This is somewhat sur-
prising since the first stage advertising equations explained only about
half of the variability of the actual advertising variable (R for ex-
cess capacity and liquidity first stage regressions ranged from .74 -
.83). The similarities between IV and OLS continue for subsequent
variable definitions as well.

 

30Each of the entry studies described in Chapter 2 which use scale
variables adopt an identical or very similar formulation.

31Deutsch, "Structure, Performance, and the Net Rate of Entry,"
p. 454 found mixed results for his measure of the "deep pocket" hypoth-
esis. His measure was a proxy for diversification, that proxy being the
primary product Specialization ratio.

32Harris, "Entry and Entry Barriers," p. 170 also found a direct
and statistically significant relationship to exist between actual entry
and the capital requirements barrier. However, this result conflicts
with the evidence reported by Orr, "The Determinants of Entry," and
Mansfield, "Entry, Gibrat's Law."

33Bain made a personal assessment of inhibiting effects on entry
in each particular industry, categorizing each industry as exhibiting

104

either very high barriers, substantial barriers, or moderate-to-low bar—
riers. Mann subsequently expanded Bain's set of industries. See Bain,
"Barriers to New Competition," and Mann, "Seller Concentration." Both
Mann and Bain showed that profits rose as entry barriers rose, and con-
cluded that higher entry barriers deter entry. However their evidence
was highly sensitive to the entry barrier group into which an industry
is classified.

34Farrar and Glauber suggest that to avoid serious multicol-
linearity no independent variable in an equation should have a R2 with
the other independent variables which exceeds the R of the equation.
D. Farrar and R. Glauber, "Multicollinearity in Regression Analysis,"
Review of Economics and Statistics 49(February 1967): 92-107.

 

35The two studies by Gorecki are the exceptions. See Appendix 2A.

36Similarities between IV and OLS estimations continue to exist
even when entrants are divided into the two distinct groups. One dif-
ference worth noting, however, is that excess capacity does not deter
foreign entry in the instrumental variable estimations but does deter
foreign entry in the OLS equations, although its effect is in the latter
case is barely significant.

37Louis Esposito and Frances Esposito, "Foreign Competition and
Domestic Industry Profitability," Review of Economics and Statistics
53(November 1971): 343-353.

 

38For a study which anticipates this result refer to Richard
Caves, "International Corporations: The Industrial Economics of Foreign
Investment," Economica 38(February 1971), especially pp. 13-14.

39One can refer to any standard textbook on International Economr
ics. For example, refer to: Mbrdechai Kreinin, International Economics
2nd edition (New York, New York: Harcourt Brace Janovich, Inc., 1975),
p. 329.

 

40F.M. Fisher, "Test of Equality Between Sets of Coefficients in
Two Linear Regressions," Econometrica 38(March 1970): 361-366.

 

41The F-test for equations 1-4 are, respectively: F9,35=1.27;
F9 3 =0.60; F9 61=3.08***; F9,61=1.75*. Although equations 1 and 2 did
not gisplay significant differences in all coefficients of the model,
certain individual barriers were found to affect foreign and domestic
entrants in significantly different ways. Equation 2 exhibited no dif-
ferences, but equation 1 showed significant differences in the effect
of capital requirements (t-stat=l.93**) and the liquidity ratio (t-stat
=2.04**). Both deterred domestic entry and enhanced foreign entry.

42Although he ignores strategic barriers and uses a faulty entry
measure, Gorecki finds foreign entrants to be less sensitive to the tra-
ditional set of entry barriers. Gorecki, "The Determinants of Entry by
Domestic and Foreign Enterprises," p. 487.

APPENDIX 4A

SELECTED DATA FOR SAMPLE INDUSTRIES

APPENDIX 4A - SELECTED DATA FOR SAMPLE INDUSTRIES

 

 

29 SIC Industry ED EF EX EXC LQ LIQ
2011 Meat Packing 2.48 1.00 2.4 1 0.30 155
2033 Canned Fruit & Veg. 0.07 2.00 1.0 1 0.44 206
2043 Cereals 0.00 1.00 4.4 1 0.47 127
2046 Wet Corn Milling 0.70 0.00 4.4 1 0.64 525
2051 Bread 1.56 0.00 3.8 0 0.41 24
2052 Biscuits, Cookies 2.20 0.00 3.8 1 0.45 190
2062 Sugar Refining 1.25 0.00 2.3 0 0.29 40
2073 Chewing Gum 0.00 1.00 0.1 0 0.62 128
2082 Beer 0.00 0.00 2.6 1 0.26 81
2085 Liquors 0.00 5.00 2.6 0 0.53 291
2086 Soft Drinks 1.00 0.00 2.6 0 0.38 104
2111 Cigarettes 0.00 0.00 2.4 1 0.66 594
2771 Greeting Cards 0.00 1.00 NA* 0 0.62 27
2834 Drugs 0.01 0.00 1.4 1 0.86 257
2841 Soups & Detergents 1.84 -3.00 N 0 0.46 452
2871 Fertilizers 0.30 0.00 NA 1 0.25 53
2911 Petroleum Refining 0.02 0.00 1.7 1 0.23 2512
3011 Tires, Tubes 0.00 2.00 —0.9 1 0.42 648
3141 Shoes 0.00 6.00 1.1 0 0.62 135
3211 Flat Class 5.00 2.00 NA 0 0.28 88
3221 Glass Containers 2.44 0.00 NA 0 0.35 150
3241 Cement 6.90 0.00 1.3 0 0.15 39
3275 Gypsum 3.58 0.00 4.0 1 0.28 110
3312 Steel 0.00 5.00 3.4 1 0.20 951
3331 Copper 1.00 0.00 NA 0 0.34 306
3334 Aluminum 4.82 3.00 NA 0 0.22 404
3411 Metal Cans 3.20 -1.00 1.0 0 0.38 370
3522 Farm Machinery 2.46 0.00 3.2 1 0.77 444
3534 Elevators 0.00 1.00 3.4 0 0.71 84
3562 Ball & Roller Bearings 0.00 3.00 N 0 0.55 78
3571 Computing & Related

Machines 2.19 0.00 0.1 1 0.30 609
3572 Typewriters 0.00 2.00 0.1 0 0.26 451
3641 Electric Lamps 0.00 2.00 1.8 1 0.34 1711
3651 Radio-TV Receiving Sets 0.05 4.00 0.8 1 0.48 354
3717 Autos 0.00 3.00 NA 0 0.50 2255
3722 Aircraft Engines 0.09 0.00 2.2 1 0.22 372
3741 Locomotives 0.00 0.00 NA 0 0.42 78
3861 Photographic Equipment 1.06 1.00 NA 0 0.76 358
3871 Watches & Clocks 0.90 2.00 NA 0 0.45 45
3982 Hard Surface Floor

Coverings 0.29 5.00 NA 1 0.62 76

 

NA - data not available for that particular industry

New York Times, various issues
Wall Street Journal, various issues
Business Week, various issues

Sources for Entry Data:

 

 

 

 

105

106

Sources for Entry Data (cont'd.)

 

Barrons, various issues
Annual Reports, various companies for various years
Industry Surveys, Standard and Poor
Pit and Quarry, Cement Maps, odd numbered years
Aluminum, Mineral Industry Survey, monthly reports
The Structure of American Industry, Walter Adams (ed.)
1. "The Steel Industry," Walter Adams;
2. "The Aluminum Industry," Robert Lanzillotti.
Federal Trade Commission Reports on:
1. "Mergers and Vertical Integration in the Cement Industry,"
April 1966;
2. "Manufacture and Distribution of Automotive Tires,"
March 1966;
"The Baking Industry," November 1967;
"Investigation of the Petroleum Industry," 1973;
"The U.S. Sugar Industry," July 1975;
"The U.S. Steel Industry and Its International Rivals,"
November 1977;
. "The Brewing Industry," December 1978;
"The Development and Structure of the U.S. Electric Lamp
Industry," February 1980.
Fortune Plant and Product Directory, 1961, 1964, 1967

 

 

 

O‘U‘IJ-‘w

WV

 

APPENDIX 4B

ORDINARY LEAST SQUARES ESTIMATION

APPENDIX 4B - ORDINARY LEAST SQUARES ESTIMATION

 

TABLE 4.9: ORDINARY LEAST SQUARES REGRESSION ANALYSIS OF
TOTAL ENTRY UTILIZING SPECIFIC ENTRY BARRIERS
(SCALE VARIABLE DISCONTINUOUS

 

 

 

 

 

 

 

 

 

 

 

 

 

Independent Regressions
Variables (1) (2) (3) (4)
Constant 6.93 6.48 5.81 5.02
G -0.20 -0.30 0.57 0.47
(0.44) (0.65) (0.93) (0.78)
A -0.09 -0.13 -0.16** -O.19**
(0.78) (1.26) (1.88) (2.24)
SC -0.22** -0.19** -0.10 -0.12*
(2.27) (1.88) (1.21) (1.58)
CC 0.04 0.03 0.01 0.02
(1.08) (0.64) (0.31) (0.88)
EX -0.34* -0.38* --- ---
(1.33) (1.45)
EXC --- -—- -0.86* -0.70
(1.37) (1.04)
LQ -1.22 -—- -4.04 ---
(0.44) (0.76)
LIQ --- -.0005 --— -.0009*
(0.62) (1.38)
TB -1.53* -1.14 -1.27* -0.94
(1.62) (1.31) (1.54) (1.18)
PR‘ -0.19** -0.17* -0.13 -0.11
(1.79) (1.57) (1.08) (0.92)
R2 0.46 0.46 0.28 0.31
N 27 27 40 4O

 

1) The figures in parentheses are t-statistics.

2) Based on one-tail tests:

* indicates 10% significance;
** 5% significance; *** 1% significance.

107

108

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 4.10: ORDINARY LEAST SQUARES REGRESSION ANALYSIS OF TOTAL
ENTRY UTILIZING SPECIFIC ENTRY BARRIERS (SCALE
VARIABLE CONTINUOUS)
Independent Regressions
Variables (1) (2) (3) (4)
Constant 6.74 6.52 5.39 5.37
C 0.98 0.88* 0.48 0.40
(1.32) (1.34) (Q;85)> (0.78)
A —0.10 -O.14* -O.17** -O.19**
(0.85) (1.34) (2.00) (2.43)
S -0.10** -0.15*** -0.08* -0.14***
(1.95) (3.10) (1.50) (2.65)
C 0.01 0.03** 0.01 0.03**
(0.94) (2.26) (0.86) (2.27)
EX -0.11 —0.32 ——- ---
(0.35) (1.20)
EXC ——— ——— -1.01* -1.03*
(1.45) (1.56)
LQ -4.22 --- -2.62 —--
(0.69) (0.50)
LIQ ——— -.0018** --- -.0016**
(2.05) (2.36)
TB -1.82** -1.46* -1.16* -0.99*
(1.86) (1.73) (1.54) (1.45)
PR -0.24* -0.22* -0.11 -0.11
(1.71) (1.72) (0.92) (1.06)
R2 0.43 0.53 0.30 0.39
N 27 27 40 40

 

1) The figures in parentheses are t-statistics.

2) Based on one-tail tests:

* indicates 10% significance;
** 5% significance; *** 1% significance.

109

 

TABLE 4.11: ORDINARY LEAST SQUARES REGRESSION ANALYSIS OF TOTAL
ENTRY UTILIZING AGGREGATE BARRIER CLASSIFICATIONS

 

 

 

 

 

 

 

 

 

 

 

 

 

Independent Regressions
Variables (1) (2) (3) (4)
Constant 7.50 5.96 7.26 5.27
G 0.86 0.51 0.74 0.72
(1.04) (_(0.61) ~_(1.22) (1.14)
_(0.47) (0.28)( (1.76) (1.27)
SB -1.05 -1.16* -1.29* -1.11
(1.21), (1.35) (1.53) (1.26)
EX -0.07 -0.21 --- ---
(9.22) (0.72)
EXC --- --- -1.01* -0.87
(1.51) (1.22)
LQ -0.08 --- -8.11* ---
(1.23) (1.60)
LIQ --- —.0001 ——— -.0002
(0.16) (0.34)
TB -1.24 -0.59 -0.86 -0.43
(1.29) (0.62) (1.13) (0.55)
PR -0.27** -0.24** -0.17* -0.18*
(2.31) (1.93) (1.53) (1.50)
R2 0.34 0.29 0.24 0.18
N 27 27 40 40

 

1) The figures in parentheses are t-statistics.

2) Based on one-tailed tests:

* indicates 10% significance;
** 5% significance; *** 1% significance.

110

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 4.12: ORDINARY LEAST SQUARES REGRESSION ANALYSIS DOMESTIC AND FOREIGN ENTRY
UTILIZING A DISCONTINUOUS SCALE VARIABLE
Independent Regressions and Dependent Variable
Variables ED(1) EF(1) ED(2) EF(2) ED(3) EF(3) ED(4) £F(4)
Constant 6.93 1.58 3.34 3.39 5.46 0.35 3.48 1.53
C 1.40** 0.09 0.45 0.55 0.45 0.13 0.36 0.11
((1.93) ((0.09) ’ (0.58) (0.61)( (0.98) (0.22) ((0.74) (0.18)
A 0.04 -0.08 -0.07 -0.02 -0.01 -O.15** -0.05 -O.15**
(0444) (0.63) (0.63) (0.20) (0.19) (1.84) (0.66) (1.75)
SC -O.1l* -O.12 -O.11* -0.11 -0.08* -0.02 -0.12** -0.01
(1.54) (1.27) ((.33) (1.24) (1.35) (0.29) (1.81) (0.11)
CC —0.03* 0.03 -.004 0.03* -0.02 0.02 .002 0.01
(1,60) (((1.81) (0:23) (1.39), ((0.85) ((1.00) ((0.12) (0.80)
EX 0.40* -0.48* 0.05 -0.30 --- --- --- --
(j.51) (1.36) (0.19) (0.94)
EXC --— --- ..- --- -O.54 -O.32* -0.43 -O.27*
(1.07) (1.50) (0.79) (1.41)
LQ -14.91*** 7.70 --- --- -9.44** 5.40 --- -..
(2.60) (1.03) (2.40) (1.08)
LIQ —-— --— -.0006** .0002 ——— -—- —.0005 -.0003
(1.86) (0.26) (1.02) (0.57)
TB -2.38*** -O.24 -0.95 -0.92 -1.54*** 0.27 -1.09* 0.15
(2.64) (0.21) (0.95) (0.80) (2.52) (0.35) (1.69) (0.20)
PR -O.28** -0.07 -0.13 -O.l4 -O.12* -0.01 -O.12 0.01
(2.35) (0.45) (0.93) (0.84) (1.44) (0.01) ((1.25) (0.09)
R2 0.44 0.26 0.26 0.22 0.33 0.15 0.23 0.13
N 27 27 27 27 40 40 40 4O

 

1) The figures in parentheses are t-statistics.

2) Based on one-tail tests:

* indicates 102 significance; ** 5% significance; *** 1% significance.

111.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 4.13: ORDINARY LEAST SQUARES REGRESSION ANALYSIS DOMESTIC AND FOREIGN ENTRY
UTILIZING A CONTINUOUS SCALE VARIABLE
Independent Regressions and Dependent Variable
Variables ED(1) EF(1) ED(2) 50(2) ED(3) EE(3) 80(4) EF(4)
Constant 4.09 2.65 2.694, .3182 4.69 0.70 3.40 1.97
c 0.39 0.59 0.12 0.75 0.16 0.31 '0.14 0.26
(0.54) (0.68) (0.17) (0.89) (0.37) (0.59) (0.33) (0.48)
A -0.07 -0.02 —0.12 -0.01 -0.03 -0.14** -0.06 -0.14*
(0.67) (0.18) (1.22) (0.10) (0.47) (1.70) (0.84) (1.63)
S -0.02 -0.08 -0.07* -0.08 -0.04 -0.04 -0.10** -0.04
(0.48) (1.28) (1.35) (1.28) (0.99) (0.76) (2.26) (0.69)
C -0.01 0.02* 0.01 0.02* .001 .008 0.02* 0.01
(0.68) (1.37) (0.50) (1.34) (0.17) (0.75) (1.60) (0.86)
EX 0.27 -0.38 0.03 -0.34 —-- --- --- ---
(0.95) (1.08) (0.10) (1.01)
EXC —-— --- --- --- -O.67* -O.34 -O.68 -0.32
(1.34) (0.50) (1.25) (0.48)
LQ -9.06* -4.83 --- --- -8.04** -5.42 --- ---
(1.53) (0.67) (1.96) (1.07)
LIQ --- —-— -.001 -.0007 --— --- -.001* -.0006
(1.18) (0.60) (1.75) (0.83)
TB -1.15 -0.67 -0.53 -0.87 -1.17** 0.02 -0.91* 0.72
(1.23) (0.59) (0.64) (0.81) (2.00) (0.02) (1.56) (0.12)
PR -0.08 -0.17 —0.04 -0.18 -0.08 -0.03 -0.09 0.11
(0.55), (1.01) (0.29) (1.10) (0.86) (0.26) (ALOQ) (0.19)
R2 0.29 0.19 0.26 0.19 0.28 0.14 0.27 0.13
N 27 27 27 27 40 40 40 40

 

1) The figures in parentheses are t-statistics.

2) Based on one-tail tests:

* indicates 10% significance; ** 5% significance; *** 1% significance.

CHAPTER FIVE

Profitability, Entry Barriers, and Entry

Whereas the preceding chapter was devoted to explaining how in—
dustry conditions affects the rate of entry, this chapter considers the
relationship between rates of entry, industry conditions and the per-
formance of dominant firms. The measure of performance used is profit-
ability. Although profitability and the elements determining profit-
ability have received extensive attention from economists, certain
methodological gaps remain. An important extension of this analysis is
the introduction of an entry variable in the profitability equation.
Such an extension is essential to capturing the dynamic component of in-
dustrial structure.

The influence of entry on dominant firm profitability will be ex-
amined with regard to the extent of entry and the type of entrant in-
volved. As was the case in Chapter 4, the treatment of specific classes
of entrants may provide insights into the profit-entry relationship that
would be hidden when only an aggregate measure of entry is used.

This chapter will also unravel precise relationships between speci-
fic entry barriers and profitability. Specifically, this analysis will
dissect the influences of entry barriers into 1) the indirect effects of
entry deterrence encountered through the rate of entry variable, and 2)
the direct effects realized independent of any impact on the rate of
entry. Separating these influences in this manner will resolve some of
the conflicting theories concerning the role of entry barriers.

112

113

DOMINANT FIRM PROFITABILITY AND ENTRY

 

Microeconomic theory has stressed the importance of entry as the
enforcer of efficient allocation of resources. This section is an
attempt to ascertain whether actual entry does alter behavior and per-
formance of the dominant firms in the way that theory predicts.

The effectiveness of entry in encouraging good performance depends
on the character of entry. In turn, the character of entry is a function
of entry barriers. Given the evidence described in Chapter 4, entrants
are deterred due to a lack of knowledge of technology and of the market,
especially when advertising intensity and consequently brand loyalty is
strong.1 It follows that entry involves the introduction of the right
product, promoted in the right way. Thus, many cases of entry involve
established firms finding small niches, qualitative or geographic, with-
in an industry.2

The character of entry is also influenced by the existence of stra-
tegic barriers. If potential entrants are able to capture all available
economies of scale and add a substantial increment of output to an in-
dustry, dominant firms could be expected to counter by utilizing excess
capacity and liquidity to generate intra-industry warfare. Cognizant of
this, potential entrants adopt an entry strategy that best compromises
the advantages of scale economies and the risks of aggressive reaction
by incumbent firms.3

The influence of structural and strategic barriers is to decrease
the penetration levels undertaken by new competitors. Consistent with

this possibility is the entry data accumulated for this study; in the

chosen oligopolistic industries actual entry occurs in small increments.

114

As a result, the competitive constraint inherent in the entry prOcess
may only become effective after small firm entrants have had a number
of years over which to expand their operations. Market power would then
be eroded, but the process of erosion would be slow.4
The analysis of total entry in Chapter 4 pointed out that both
structural and strategic barriers did significantly deter entry.
Furthermore, it is probable that these barriers reduced the size of the
viable entrants, as just discussed. Combining these two effects of
entry barriers on the extent of entry, it would appear that barriers may
not only diminish but possibly render impotent the efficiency enforcing
power of entry. Moreover it appears the competitive constraint exerted

by entry, as brought out in Chapter 4, is most likely to originate in

the measure of foreign entry.

Estimation

 

In an attempt to capture the relationship between market performance
and entry, this study utilizes the profit equation constructed in Chapter
3 and stated in its estimating form in Chapter 4. Restating that profit-
ability equation yields

(4.2) p = bo+bIGt+b2(EB)t+b3(CR)t+b4(CE)+b5(E)t+Vt

This equation is unique to the study of profit-market structure studies
because it considers the effect of entry directly.5

The profitability equation is but one equation of a system of equa—
tions explaining the relationships between industry structure, conduct,

and performance. Within a complete model of industrial organization,

115

advertising, excess capacity, and liquidity -— elements of market con-
duct -- and entry should be considered endogenous.6 Therefore, consis—
tent estimation requires an instrumental variables technique.

To purge the estimation of the correlation between the explanatory
entry variable and the residual of the equation, the estimated values of
the entry variable (B) were substituted for the true values in the
structural profitability equation. In this regard, the specific values
for the rate of total entry, domestic entry, and foreign entry were de-
rived using equations 2 and 4 from Tables 4.4, 4.5, 4.7, and 4.8.

After invoking E as an instrument for E, the profitability equation
was estimated using instrumental variables. Included in the set of in-
struments were all exogenous and predetermined variables in the profit-
ability equation plus the trade barrier dummy (TB) and pre-entry rate
(Pr). Additional instruments were constructed using the squares of all

non-dummy instruments.7

Empirical Results and Interpretation

 

Table 5.1 presents the multiple regression equations relating domi-
nant firm profits to the trio of entry measures, seller concentration,
and various combinations of other explanatory variables.8

The regression coefficients for the aggregate entry variable con-
sistently displayed the expected negative sign. However, the degree to
which entry depressed dominant firm profits was modest relative to the
radical changes seen by some.9 Indeed, the total rate of entry variable
was slightly significant in only two specifications of the profitability
equation. While this modest relationship runs counter to certain previous

studies, the result is not unexpected. In the sample of oligopolistic

11(5

.oocuouuucwwo N_ «as unaccoueqcaun mm «a “ouccouuueuuo uo— sousowven a

«IUQOH dun—also 60 9.08 an

.ouuuauueuouu one consensus-a ea sausuwu ugh a~

.soaash also .Acv souussco
Iona vo>uuoo sun Aw~1~V occuamoawou “c.q use n.< .m.c .c.¢ nuance .ANV coauaavo noun vo>ausv one Act-v neoasssnnox A.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

as as cs. oq as as a~ “N AN a" an «N z
-z.oe ~ns.oe Ass.oe - .~mw«aa
nn.¢- _~.o- an.o- “9.: as
Aa~.°3 ~n_._v (.1. ~s~.oe()) aan.oe ..
c~.c1 is.o- an.o n_.c- an
Aen._v .1 Amn._e “no.—v Aen.ov
44s.“- .ms.a- an..- .a.cu u
(Add..v “no.2L Aon.oq, -a.ae Azo.ou “so.oe Ann.ov .-~.av Asunmv Anonmv Aou..«) .1amanma
«no.0. «oc.o1 .c.o anaoo.c- pac.c- ~oo.o ~oo.o- aoo.c1 «a.o snoo.o1 ooo.o- noo.o mo
aoc._c Aac.~e Ame.ov «wa.oe Aoa.oa .«oa.ov
~o.o- no.0- .o.o- ~o.o- ~o.o- ~o.c- so
Ann..e ..~n~._a A».._e ....s x-._e xa~._al ~a~.ae
.~o.o ~c.o ~o.o sno.= «no.o ~o.o o
Asm._e Amwnww Ana.o. (~MD.~V x~u.mu. Amw.au
.s_.o- 4~_.o- s_.a- 44n~.o1 n~.o- no~.o- om
«sense. Ano.ov An_.oe. . (.Amo.ne An_.oe x-.o5
goo.o ~oo.o so.c noo.=- .o.o- ~o.o m
Ana.aa ~m5.4e .Nan..v Nso._v Asa.ze Ace..v :Nwo.~e Aka.~v .¢_.~e Amw.~v Aca.~u .Awonme
as-.o ae-.o scun.o s¢-.o c¢n~.o ¢n~.c cann.o ecsn.o saun.o econ.o scon.o ae_n.o <
as..~e A~n44e “wo.oV xo~.~e -n.~V ion..e .1wa.oe Amwnoc .os.me ~5_.~w1 .sn.~v «an.~m
«0.0 eco.a no.cu secma.: assoc.o aeoo.: ¢=:.ct mc.o nc.c «soc.o asao.o eeoo.o do
:98 .flée ate «Sis mfifl!AjQEEfiWQ1AS;. :03 ARdw1dic Jfidq
ascon.u scann.~ secon.~ asa~o.~ acano.~ cacno.~ «casc.m «cemc.n «canm.~ scaoo.n seemn.~ ascan.~ u
an.n os.s 4n.a as.n “3.4 mwunzuu--:w~.e mwregu. s:.o~ no.4 °~.s .n.o scannnno
a~ae A__L Aces Ame .Nae Ass Ase -..xwg-)z Aae Ame .NNV NAM nnqnnsnns
ocsummonxom unevenness—

 

 

01. I. ll' 0. All

 

 

d

I I1: 1'1"- 0"

NHHAHQ<HHhoxm zx~m Pz<znzoa zo mmm~d¢<m rxkzm 22¢ >xezm mo HU<mZA mzb ac zouk<2~hmm mmgc<um<> A<hzmx=zhmzH

 

"_.n u4n<h

 

117

industries considered, it was shown that structural and strategic bar-
riers effectively deterred total entry. This impact is now making its
presence felt by limiting the ability of entry to reduce the market power
of dominant firms.10 On the other hand, this result does not imply that
total entry need be impotent. Rather, total entry can effect the be-
havior and performance of the smaller firms within the industry -- a bat-
tle between the helpless and the hapless, relative to the secure position
of the dominant firms in the industry. In conclusion, it can be said
that total entry conditions behavior to some extent but that the struc-
tural and strategic elements of the economic environment are most im-
portant in delineating the profit opportunities of an industry's dominant
firms.

The diluted competitive constraint exerted by total entry on dominant
firm profitability, while most likely to originate from the measure of
foreign entry, derives its nature from both foreign and domestic entry.
It was found that foreign and domestic entry have identical direct ef-

11 Although this appears to conflict with the

fects on profitability.
evidence presented in Chapter 4, foreign entry is more effective than
domestic entry in reducing dominant firm profitability through its ef-

fect on concentration. In the most accurate Specifications of the model --
those using the discontinuous scale variable (SC) -- the interaction be-
tween entry and concentration (CE) achieves a significant negative co-
efficient. Therefore, the first derivative of dominant firm profits

with respect to foreign entry is negative in Spite of the insignificance

of the foreign entry variable alone.

118

PROFITABILITY AND ENTRY BARRIERS

 

As was discussed in Chapter 2, previous analyses of the profit-
ability-entry relationship assumed that if profitability and Static en-
try barriers displayed a Significant positive correlation then the
hypothesized static entry barriers were in fact deterring entry. How-
ever, it was further shown that direct relationships between profits and
entry barriers could arise if the hypothesized barrier did not reduce the
rate of entry but rather had an entry independent effect which increased
profit rates. Therefore, profit-market structure studies to date have
been unable to discern the precise role of these structural entry
barriers.

By considering the interdependence between entry and profits ex-
plicitly, this analysis is able to identify each of the dual effects
Structural entry barriers may impose on profitability. The specific
formulation needed to accomplish this task is the inclusion of the entry
variable in the profit equation along with the structural barriers, as
follows

(4.2) P = b +b1Gt+b2(EB)t+b3(CR)t+b4(CE)+b5(E)t+Vt

O

The entry deterring capabilities of entry barriers are expressed through
the entry variable. In Chapter 4, the ability of individual barriers to
deter entry was discovered by estimating the entry equation (4.1) and

analyzing the coefficients of the equation. That estimated equation can

be written as

A
A

(4.1') E= a0+a1PR+a2Gt+a3(EB)t+a4(SB)t+a5(TB)+Ut

119

where
“ = an estimated coefficient value.
Substituting (4.1') into the estimated profitability equation yields the
following relationships:
(5,11) 8 = bo+blct+62(EB)t+b3(CR)t+b4(CE)+65 50+;1pn+;20t+;,(rn)t

+a4(SB)t+as(TB)+Ut +vt

The total effect of entry barriers on profitability is, therefore, equal

to12

such that
62 = the entry independent effects of entry barriers
A A on dominant firm profitability, and
b5a3 = the entry deterring effects of entry barriers on

dominant firm profitability.

A summary of the effects of specific structural entry barriers on
dominant firm profitability is presented in Table 5.2.
Profitability is, to varying degrees, reduced by entry and entry,

13 It

in turn, is reduced by advertising intensity and scale economies.
follows that the entry deterring effects of these two Structural entry
barriers result in increased profit opportunities for the dominant firms --
(BS-£3)>o. However, the same cannot be said for the capital require-
ments barriers. This variable did not deter entry. Thus, any positive
relationship reported between capital requirements and profitability does
not stem from this hypothesized barrier's entry retarding capabilities.14
The empirical results contained in Table 5.1 also point to the pos-

sibility that certain of these barriers have an entry independent effect

on profitability. Advertising intensity displays an overwhelming

120

 

ENTRY BARRIER

TABLE 5.2: THE EFFECT or ENTRY BARRIERS 0N PROFITABILITY
(I) . (2) (31
b .1

“ ‘ 3
2 b5 a3 _. b2+b5a3._

 

ADVERTISING INTENSITY (A') 0.23** -O.65* —0,45*** 0.52***

(A) o.22** -1.34* -o.45*** 0.82***
SCALE ECONOMIES (SC) —o.14 -0.65* -0.08* -o.o9

(S) 0.01 -1.34* -o.o9** o.13*
CAPITAL REQUIREMENTS (CC) -o.01 -0.65* 0.002 -0.011

(C) 0.02 -1.34* 0.02** 0.007

 

1)

2)

3)

4)

The figures for 62 are culled from equations (7) and (10) -- Table 5.1.
A' corresponds to equations using discontinuous scale variables.

The figures for b are derived from the same source as above; figures
for a3 are derives from equation (3), Tables 4.4 and 4.5.

The respective t-statistics are: 3.71; 3.84; 1.25; 1.34; 0.41; 0.53.
The procedures for deriving asymptotic variances for functions of non-
linear random variables can be seen in Arthur Goldberger, Econometric
Theory (New York: John Wiley and Sons, 1964), pp. 122—126.

Based on one-tail t—tests: * indicates 10% significance; ** 5% sig-
nificance; *** 1% significance.

121

propensity for increasing profitability directly. Profits increase as
increases in advertising intensity expand the quantity of information
disseminated to previous and prospective customers.15 Conversely, the
scale economies barrier has a negative entry independent effect on
profits. The scale economies variable may be acting as an agent for
firm size, noting that as firm size increases, management problems and
inefficiencies may set in reducing the level of profits.

In the equations estimated using the continuous scale variable (8),
profitability is directly affected by capital requirements. The coef-
ficient for capital requirements is positive and significant. In the
same equations, the concentration variable is consistently insignificant.
The behavior of these two variables appears to be due to the multicol-
linearity among these two variables and the scale variable.16 When profit-
ability equations were estimated using structural variables less collinear
(SC, CC), both the concentration and capital requirements take on their
expected roles. Concentration exhibits a significant positive coeffi-
ceint; capital requirements becomes insignificant.

These results lend insight into the assumption, unabashedly used by
analysts of profit-market structure Studies, that a significant positive
relationship between a hypothesized structural entry barrier and profit-
ability is a Sign that structural entry barriers do deter entry. This
assumption ignores the possibility of these entry barriers having an
entry independent and direct impact on profitability. In the case of
the scale barrier—profitability relationship, positive coefficients do
indicate the entry deterring propensities of the scale barrier since the

barrier's entry independent effect on profits is negative. In contrast,

122

advertising intensity displayed a robust ability to increase profit-
ability independent of its entry deterring capabilities. AS a result,

a positive relationship between advertising and profitability overstates
the potential of advertising to deter entry.

In column 3 of Table 5.2, the entry deterring and the entry inde—
pendent effects of the traditional entry barriers were combined to de—
termine the total impact of each barrier on dominant firm profitability.
The aggregate effect of advertising was to increase profitability; scale
economies also significantly increased profitability in one of the cases
considered; capital requirements never achieved a Significant impact on

profitability.

SUMMARY

Entry, or more accurately the probability of entry, has been relied
on as a regulator of efficient economic performance, enforcing allocative
and productive efficiency, encouraging innovation, and eliminating excess
profits. The evidence presented in this study has cast some doubt as to
the competitive constraint of entry. It has been Shown that the profit~
ability of dominant firms was not significantly reduced by entry; in
fact, although foreign firms appeared to be in the most favorable
position to enter an industry, neither foreign nor domestic firms were
able to overcome most strategic and structural entry barriers. Strategic
barriers, being effective in reducing the rate of entry and probably the
size of the viable entrant, increased profitability. The empirical evi-
dence presented in this study, also showed that Structural barriers de-
ter entry. Consequently, the traditionally assumed entry barrier-profit—
ability relationship has been empirically justified. However, one Sig-

nificant qualification must be noted. Advertising intensity has a

123

prominent and direct effect on profitability independent of its entry

deterring capabilities.

NOTES

CHAPTER FIVE

1See especially Tables 4.4 and 4.5.

2For example refer to L.A. Guth, "Advertising and Market Struc-
ture Revisited," Journal of Industrial Economics 19(April 1971): 179-
198. This discussion implies that entry barriers dampen the effective-
ness of entry. However, studies have Shown that entry can exert a pro-
competitive effect despite entry barriers. For such a Study see Maury
Harris, "Entry and Long-Term Trends in Industry Performance," Antitrust
Bulletin 21(Summer 1976): 295—315. Harris also shows that industries
utilizing extensive advertising are more insulated from the effects of
entry than are industries with relatively low advertising levels.

 

3For a further discussion on the impetus of such compromises,
refer to Scherer, Industrial Market Structure 2nd edition, pp. 248-249.
Of course, such a threat dominates the thinking of domestic entrants
more than foreign entrants, as described in the previous chapter.

 

4A comparison of two Studies analyzing the reduction in market
power over different lengths of time highlights this point. Harris,
"Entry and Long-Term Trends," utilized a 16-year period and found that
entry significantly reduced dominant firm profitability. In a related
study which examined the market share stability of 15 leading firms in
205 industries over the eight years from 1947-1954, Cort discovered
that 3/4 of the industries had firms which exhibited significant Sta-
bility. In particular, 20 industries, including some of the best known
oligopolies, had dominant firms which exhibited what could be termed
total Stability. Michael Gort, "Analysis of Stability and Change in
Market Shares," Journal of Political Economy_71(February 1963): 51-63;
especially Appendix Tables 9 and 13.

 

5Preliminary Specifications included the strategic barriers in
the profitability equation. However, these variables never exhibited
statistical significance and were omitted from subsequent estimations.

6See discussion on pp. 69-70.

7Kelejian, "Two-Stage Least Squares."

8The results generated using ordinary least squares are very
Similar to those found using instrumental variables (refer to Appendix
5A). The discussion in this chapter is, therefore, aimed simultaneously
towards both sets of results.

124

125

Representative examples are: Esposito and Esposito, "Foreign

Competition," and Harris, "Entry and Long-Term Trend."

10Berry, in an analysis of the effect of cross-industry entry
(diversification), noticed that entry was weakest in tight oligopolies.
Berry's source for entry data was identical to the primary source for
the entry data accumulated in this study. Charles Berry, "Corporate
Bigness and Diversification in Manufacturing," Ohio State University
Law Journal 28(Summer 1967): 402-426.

11Imposing the restriction that foreign and domestic entrants af-
fect dominant firm profitability in identical ways was found to be statis—
tically valid. The highest F-statistic calculated for the equations in
Table 5.1 was F1’31=1.97 but F 9022.88.

12These coefficients may be interpreted as the separate effects
of each barrier -- scale economies, capital requirements, and advertis-
ing -- and/or the collective effect of the set of barriers of which
these individual barriers are elements.

13See Table 4.4, p. 74. Strategic barriers also deterred entry
but because initial estimations found these barriers never to have a
statistically Significant entry independent effect, they were subse-
quently omitted.

14Comanor and Wilson, "Advertising, Market Structure, and Per-
formance," p. 435 imply that a positive relationship between profit
rates and capital requirements arise because capital requirements is
an effective barrier to entry.

 

15An example of this argument is contained in Phillip Nelson,
"Advertising as Information," Journal of Political Economy 82(July/
August 1974): 729-754.

16When a collinear variable is introduced into an equation, one
or more of the collinear variables may decrease in significance. This
appears to be the case for the concentration variable. When the con-
tinuous scale (S) and capital requirements (C) variables are introduced,
in contrast to the use of SC and CC, the regression coefficient on the
concentration variable falls, its Standard error increases, resulting
in a drastic reduction in the t-Statistic. Under certain circumstances,
the insertion of a collinear variable may raise the absolute value of
the regression coefficient of a related variable. If this effect is
enough to offset the increase in the standard error, the t-statistic
increases. This appears to be the situation for the capital require-
ments variable. G.S. Maddala, Econometrics (New York, New York: McGraw-
Hill Book Co., 1977), p. 185 and H. Theil, Economic Forecasts and Poligy
(Amsterdam: North-Holland Publishing Co., 1961), p. 327.

 

APPENDIX 5A

ORDINARY LEAST SQUARES ESTIMATION

1226

ounvoooun uvswactoaowvuou ozu .uo>uao=

.auaouuonvounoao; no nunozuoaa; 0:» Downs» ou voauau
.auunnoca ago «0 sumo use on concave zagwoqaau one: noouuaavu any mo ousavuoou any An

.ooccoauwcwfio NON a Amocnuauucmfim Nn as "socwuauwcwuu Nu nouoouvcfi «es

Nounou Huouloco no canon Au

.uoauowueuuuu one nomosucouwa e« oouauuu o:k A.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

as as 04 ca as as AN AN AN AN AN AN 2
nm.o mm.o mm.o sn.o em.o sm.c No.9 Ne.o as.o so.c oo.o so.o Ne
Awa.oe Aim.ov Asn.ov Ans.ov
so.o cm.o- _N.o- A_.o AN
Asn.ov AA_.SV Amn.ov Ann.ma
N~.c- Ns.o- AN.o AN.o- am
ASN.0v AAs.ov ANS.oM Am~.ov
SN.o- am.c- os.ou .N.o- N
Ane.oe And.cv An_.ov Aes._e Nan.oe Amz.oe AAo._e Amm.ov Aso.oV Acn.ov AA_.oV Awo.ov
Noo.o. ocoo.o- Noo.o nmoo.o- ~oo.o- Noo.o Aoo.o soo.o ~o.o Noo.o- Aooo.o Noo.o no
Ann.ov NAN.ov AaN.ov .Non.ov Asn.oa NNo.qw
moo.o1 oco.o- noo.o- Noo.o- No.o- noe.cu on
NAS.AL ANN.AV Aom.av Ano.NV Aao.NV AAA.AV
«Ne.o cano.o ssno.o semo.o semo.o sync.o o
AAs.Av ANA.AD RAN.LV Aao._v AAm._e Aeo.NM
«s2.o- .s_.o- s_.o- «ASN.c1 anoN.ou ¢NN.o- um
(No~.oe NN_.oV AoN.oV ANN.ov ASN.oV A_a.ov
Aoo.o Noo.o _o.o No.o No.o no.o m
ANA.NL ANA.NV NAN.NV ANN.ND Aim.NV Aq_.Ne ANA.NV AAA.NM, Amm.mv ASN.NV Amm.Nm Anm.Nv
«sewn.o saaoo.o seamo.o assen.o «nann.o acmq.o sseam.o gee—c._ «esoa.o «Neow.o sssoa.o «sacn.o <
Asm.ov ANo.av Ans.ou Ams.Nw1 ASS.NV AON.AV Ano.ov A_~.ov An_.ov Noo.Nv AA_.NV Astwm
No.o No.o Na.o «nmo.c namo.o so.o Noo.o- S=O.o _o.o- ..no.c 4.no.o Ao.o Nu
)NNn.Ne ASS.NM, Awn.nw «08.nm. Ans.Ne AoA.AV AAA.NV ASA.NV AoA.Nv .NNS.NV Aon.No NAA.NM.
asen~.~ saco~.~ sesn~.N asewq.~ seeme.~ cae~n.~ s44mm.~ «sewn.~ aasnu.~ asson.N asa-.~ «cann.~ o
as.“ ms.4 A5.“ sq.N 4N.N no.4 no.4 oo.c Ns.m Ns.a mo.N NA.N unannnnu
AN_V A~_V Nose Ame Ame AAV Ase Ame ASL AND ANV Ame noNanNnns
ucovcoaovou
NNNaAa<ANAQNN :NNN Nz<zezoa zo mmmNNN<N Amaze nz< Amaze no ao<nzA may no mAwya<z< zonmNNumm mme<aam enema NN<anNo "N.m man<a

 

chA<zAAmN mme<=om Amena Ne<zaaeo . <n xaazean<

CHAPTER SIX

Conclusions and Policy Implications

The primary objective of this study was to extend the concept of
entry barrier to include strategic barriers and in so doing empirically
test if these investments made by established firms facing the threat of
entry had the ability to influence entrants' expectations regarding their
post-entry profit opportunities. Previous entry-entry barrier studies
had considered only structural market conditions to be barriers to entry
and consequently, implied that a given industry structure was fixed and
that that structure determined the behavior and performance of the firms
in the industry. Employing the concept strategic barriers in an analysis
of entry conditions allows for the unique possibility that conduct may
have a feedback effect on structure.

In order to determine whether strategic barriers did reduce the rate
of entry, measures of excess capacity and liquidity were incorporated into
equations seeking to explain inter-industry differences in the rate of
entry for a set of 40 oligopolies over the period 1959-1968. By utilizing
indices of excess capacity and liquidity as direct determinants of actual
entry, the study was able to point out that Strategic barriers did indeed
reduce the rate of entry beyond that which would have occurred due to
Structural barriers alone. This evidence remained in spite of alternative

(definitions of the variables and sample.

127

128

Consequently, firms can no longer be considered to be passive agents
within a given economic environment. Rather they can be expected to
undertake an active role in an attempt to mold their environment so as
to be more conducive to their particular long—run goals. In this case
established firms did not remain idle depending on structural barriers
to insulate them from entrants but instead appeared to take action, or
were prepared to take action, against new competitors in order to preserve
their market shares.

In the process of ascertaining the effectiveness of Strategic bar-
riers, this study also sought to identify whether or not different
classes of entrants responded differently to both Strategic and Structural
entry barriers. The Statistical evidence showed conclusively that foreign
and domestic entrants do engage in dissimilar reactions to the various
barriers. Furthermore, foreign entrants appeared to be less disadvantaged
vis-a-vis established firms than were the domestic entrants. This result
signifies that there is a heterogenous supply of potential entrants, with
foreign potential entrants less likely to be deterred than domestic po-
tential entrants given a particular blend of strategic and structural
barrier in the industry. Therefore both the extent of entry barriers and
how they deviate in effectiveness with respect to different potential en-
trants are important in the formulation of an entry strategy by established
firms.

A second major objective of this study was to determine whether the
traditional entry barriers impact: 1) indirectly on profits through their
direct effects on entry; 2) directly on profitability through an entry in-

dependent effect; or 3) both. By considering explicitly the

129

interdependence between entry and profitability, this study revealed

that advertising intensity and scale economies had a direct effect on
profitability in addition to their ability to deter entry. Noting the
direction of the direct effects, this study concludes that previous
positive correlations between profitability and scale economies under-
state the capacity of scale economies to deter entry whereas such positive
relationships between advertising intensity and profits overstate the
negative impact of advertising on entry. Therefore, conclusions arising
from profit-market Structure analyses concerning the entry retarding
capabilities of specific entry barriers must be regarded with caution.

In accomplishing the objectives described above, this study was able
to focus on actual entry and its ability to alter the behavior and per-
formance of the dominant firms in a set of oligopolistic industries.

This Study provides evidence which suggests that entry offers little
prospect for constraining oligopolistic behavior. Even the most favored
class of entrants, foreign entrants, were unable to overcome strategic
and structural barriers to a degree which enabled them to radically alter

the existing industry performance.

Public Policy

 

A series of public policy recommendations can be derived from the
results contained in this study. All of the policy implications are con-
cerned with the complex relationships between entry, entry barriers, and
industry performance.

In order to encourage entry and capture the pro-competitive effects
of entry on industry performance, antitrust enforcement agencies Should

monitor the existence and use of strategic barriers, especially excess

130'

capacity and liquidity, which were proven to be significant entry deter-
rents. Investments in excess capacity and liquid asset reserves by estab—
lished firms represent non-Structural, and hence unnecessary, limitations
imposed on potential competitors. To determine whether Strategic barriers
are being erected in particular industries, regulatory agencies must con-
sider the growth of the industry and the entrance of new firms into the
industry. If there exists excess productive capacity and/or substantial
liquid reserves, and if the number of firms in the industry remain un-
changed for long periods despite the growth of demand, one might suspect
that firms within the industry are guilty of utilizing their productive
and liquid asset reserves to discourage market penetration by potential
entrants.1 Industries displaying these characteristics should be the
primary targets for intensive scrutiny.

The influence of liquidity in deterring entry suggests that antitrust
enforcement agencies should not only monitor the existence and use of
liquidity but should also disapprove, or be hesitant in approving, mergers
involving the acquisition of relatively non-liquid firms in currently
competitive industries by financially strong and highly liquid companies.
Such acquisitions would enable the relatively rich firm to devote portions
of its budget to the creation of liquid barriers in this new industry. As
a direct result, potential entrants would be more reluctant to enter the
industry than they would have been prior to the merger.2 Consequently,
the competitive nature of the industry would be reduced.

Moreover, policy Should be aimed at reducing or eliminating excess
capacity and liquidity independent of their anti-competitive features.

As firms attempt to create, protect, or extend their monopoly power by

131

investing in excess capacity and liquidity, society's resources are being
wasted. Because no social utility is gained by these expenditures, re—
gardless of whether Strategic barriers are or are not effective in deter—
ring entry, such investments represent a loss to society. This loss is
over and above the traditional deadweight loss associated with non-
competitive markets.

A recent policy trend directed at improving industry performance is
the policy of deregulation. The premise behind the application of a de—
regulation policy is that entry can substitute for regulation in encour-
aging and enforcing competitive behavior. For entry to fulfill this
critical role, entry must bring about dramatic changes in industry organ-
ization. This study has empirically shown that entry in unregulated
oligopolies was unable to affect the performance of the dominant firms
in those industries.

It is the opinion of this study, however, that the role of entry
could be improved by changes in current public policy. These policy
alterations should include the directives of this study, namely that
there should be a strict monitoring, if not elimination, of excess capa-
city and liquidity, and that there should be a denial of mergers in-
volving acquisition by a highly liquid firm. In addition, these public
policies may be supplemented by other policies including tax subsidiza-
tion and purchase guarantees.4 Furthermore, entry promoting policies
must recognize the heterogeneous nature of potential entrants, as was
brought to light in this study. In aggregate, these factors could
change not only entry conditions but could alter also the character of

entry. Although these policies may increase the power of entry as an

132

enforcer of efficient economic performance, entry should be viewed as
skeptically as present regulatory procedures in determining whether

deregulation is the specific policy most apt to enhance social welfare.

NOTES

CHAPTER SIX

1Such an interpretation is not without precedent. In U.S. v.
Aluminum Co. of America, 148 F.2d 430 431 (1945), Judge Learned Hand
stated that "we can think of no more effective exclusion" program
"than progressively to embrace each new opportunity as it Opened,
and to face every newcomer with new capacity already geared into a
great organization..." Using excess capacity in this fashion pointed
out that "Alcoa meant to keep, and did keep, that complete and exclusive
hold upon the ingot market with which it started."

2The Proctor and Gamble case, which involved the attempted mer-
ger of the financially secure firm of Proctor and Gamble with the rela-
tively small bleach firm Clorox, indicates that the courts might be
receptive to such an argument.

3A generalization of this notion can be seen in Gordon Tullock,
"The Welfare Costs of Tariffs, Monopolies, and Theft," pp. 224-232. See
also Spence, "Entry Capacity, Investment, and Oligopolistic Pricing,"
pp. 543-544.

4Such policies were instrumental in maintaining the viability of
entrants in the aluminum industry and the aircraft engine industry,
respectively. It is important to note that the competitive benefits
occuring from any entry promotional campaign must always be weighed
against the costs incurred implementing such policies, and only if the
benefits exceed the costs should the program be adopted.

133

BIBLIOGRAPHY

BIBLIOGRAPHY

Books and Articles

 

Adams, Walter. "The Steel Industry." In The Structure of American
Industry. Ed. Walter Adams. New York: MacMillan Publishing
Co., Inc., 1977, pp/ 86-129

 

Bain, Joe S. Barriers to New Competition. Cambridge, Mass.: Harvard
University Press, 1956.

 

Baron, David. "Limit Pricing, Potential Entry, and Barriers to Entry."
American Economic Review, 63(September 1973), 666-674.

 

Baumol, William, Elizabeth Bailey, and R.D. Willig. "Weak Invisible
Hand Theorems on Pricing and Entry in a Multiproduct Natural
Monopoly." American Economic Review, 67(June 1977), 350-365.

 

Benham, Lee. "The Effect of Advertising on the Price of Eyeglasses."
Journal of Law and Economics, 15(October 1972), 337-352.

 

Bergson, Abram. "On Monopoly Welfare Losses."

63(December 1973), 853-870.

American Economic Review,

Berry, Charles. "Corporate Bigness and Diversification in Manufacturing."
Ohio State University Law Journal, 28(Summer 1967), 402-426.

 

---------- . Corporate Growth and Diversification. Princeton: Princeton
University Press, 1975.

 

Bhagwati, Jagdish. "Oligopoly Theory, Entry Prevention, and Growth."
Oxford Economic Papers, 22(November 1970), 297-310.

 

Bishop, R.L. "Duopoly: Collusion or Warfare?" American Economic
Review, 50(December 1960), 933-961.

 

Boyer, Kenneth D. "Informative and Goodwill Advertising." Review of
Economics and Statistics, 56(November 1974), 541-548.

 

Brown, Randall. "Estimating Advantages to Large-Scale Advertising."
Review of Economics and Statistics, 55(August 1978), 428-437.

Brozen, Yale. "Concentration and Structural and Market Disequilibria."
Antitrust Bulletin, 16(Summer 1971), 241-248.

 

134

135

Caves, Richard, and Michael Porter. "From Entry Barriers to MObility
Barriers." anrterly Journal of Economics, 91(May 1977), 241-261.

 

Caves, Richard. "International Corporations: The Industrial Economics
of Foreign Investment." Economica, 38(February 1971), 1-27.

---------- , J. Khalilzaden-Shiraz, and M.E. Porter. "Scale Economies in
Statistical Analyses of Market Power." Review of Economics and
Statistics, 57(May 1975), 133-140.

 

 

Clark, J.B. Essentials of Economic Theory. New York: MacMillan, 1907.

 

Comanor, William, and Thomas Wilson. "Advertising, Market Structure,
and Performance." Review of Economics and Statistics, 49(August
1967), 423-440.

 

---------- . "The Effect of Advertising on Competition: A Survey."
Journal of Economic Literature, 17(June 1979), 453-476.

 

Comanor, William, and Robert Smiley. "Mbnopoly and the Distribution of
Wealth." Quarterly Journal of Economics, 89(May 1971), 327-336.

 

Cowling, Keith, and Dennis Mueller. "The Social Costs of Mbnopoly Power."
Economic Journal, 88(December 1978), 724-748.

 

DeBondt, Raymond. "0n the Effects of Retarded Entry." Europgan Economic
Review, 8(August 1977), 361-371.

 

---------- . "Limit Pricing, Uncertain Entry, and Entry Lag."
Econometrica, 44(September 1976), 939-946.

 

Deutsch, Larry. "Structure, Performance, and the Net Rate of Entry into
Manufacturing Industries." Southern Economic Journal, 41(January
1975), 450-465.

 

Edwards, Corwin. "Conglomerate Bigness as a Source of Power." In
Business Concentration and Price Poligy. National Bureau of
Economic Research Conference Report. Princeton: Princeton
University Press, 1955, pp. 351-359.

 

Esposito, Louis, and Frances Ferguson Esposito. "Foreign Competition and
Domestic Industry Profitability." Review of Economics and Statistics,
53(November 1971), 343-353.

 

Farrar, D., and R. Glauber. "Multicollinearity in Regression Analysis."
Review of Economics and Statistics, 49(February 1967), 92-107.

 

Fellner, William. "The Influence of Market Structure on Technological
Progress." anrterly Journal of Economics, 65(November 1951),

 

136

Fisher, F.M. "Test of Equality Between Sets of Coefficients in Two
Linear Regressions." Econometrica, 38(March 1970), 361-366.

 

Frederiksen, Peter C. "Prospects of Competition from Abroad in Major
Manufacturing Oligopolies." Antitrust Bulletin, 20(Summer 1975),
339-371.

 

Galbraith, John Kenneth. Economics and the Public Purpose. Boston:
Houghton Mifflin Co., 1973.

 

Gaskins, Darius. "Dynamic Limit Pricing: Optimal Pricing Under the
Threat of Entry." Journal of Economic Theory, 3(September 1971),
306-322.

 

Goldberg, Lawrence. "The Effect of Conglomerate Mergers on Competition."

Journal of Law and Economics, 16(April 1973), 137-158.

 

Gorecki, Paul. "The Determinants of Entry by Domestic and Foreign
Enterprises in Canadian Manufacturing Industries." Review of
Economics and Statistics, 58(November 1976), 485-488.

 

---------- . "The Determinants of Entry by New and Diversifying Enter-
prises in the U.K. Manufacturing Sector, 1958-1963." Applied
Economics, 7(June 1975), 139-147.

Gort, Michael. "Analysis of Stability and Change in Market Shares."
Journal of Political Economy, 71(February 1963), 51-63.

 

---------- . "An Economic Disturbance Theory of Mergers." Quarterly
Journal of Economics, 83(November 1969), 624-637.

 

Guth, L.A. "Advertising and Market Structure Revisited." Journal of
Industrial Economics, 19(April 1971), 179-198.

 

 

Hall, Marshall, and Leonard Weiss. "Firm Size and Profitability."
Review of Economics and Statistics, 49(November 1967), 310-319.

 

Harberger, Arnold. "Manopoly and Resource Allocation." American
Economic Review, 44(May 1954), 77-87.

 

Harris, Maury. "Entry and Barriers to Entry." Industrial Organization
Review, 4, No. 3 (1976), 165-174.

 

- --------- . "Entry and Long-Term Trends in Industry Performance."
Antitrust Bulletin, 21(Summer 1976), 295-315.

 

Hines, Howard. "Effectiveness of Entry by Already Established Firms."
Quarterly Journal of Economics, 71(February 1957), 132-150.

 

Ireland, N.J. "Concentration and the Growth of Market Demand." Journal
of Economic Theory, 5(0ctober 1972), 303-305.

 

137

Jacquemin, Alex, and Jacques Thisse. "Strategy of the Firm and Market
Structure." In Market Structure and Corporate Behavior. Ed. Keith
Cowling. London: Gray-Mills, 1972, pp. 61-84.

 

Kaldor, Nicholas. "Market Imperfections and Excess Capacity." Economica,
2(February 1935), 33-50.

Kamerschen, David. "An Empirical Test of Oligopoly Theories." Journal
of Political Economy, 76(July/August 1968), 615-634.

 

Kamien, Merton, and Nancy Schwartz. "Limit Pricing and Uncertain Entry."
Econometrica, 39(May 1971), 441-454.

 

---------- . "Potential Rivalry, Mbnopoly Profits, and the Pace of Inven-
tive Activity." Review of Economics and Statistics, 45(October 1978),

 

Kelejian, Harry. "Two-Stage Least Squares and Econometric Mbdels Linear
in Parameters but Non-Linear in the Endogenous Variables." Journal
of the American Statistical Association, 65(June 1971), 373-374.

 

Klein, Lawrence. A Textbook of Econometrics. Englewood Cliffs, N.J.:
Prentice-Hall, Inc., 1974.

 

Kottke, Frank. "Market Entry and the Character of Competition." Western
Economic Journal, 5(December 1966), 24-43.

 

Kreinin, Mardechai. International Economics. New York: Harcourt Brace
Janovich, Inc., 1975.

 

Lanzillotti, Robert. "The Aluminum Industry." In The Structure of
American Industry. Ed. Walter Adams. New York: MacMillan
Publishing Co., Inc., 1961, pp. 185-232.

 

Leibenstein, Harvey. "Allocative Efficiency vs. X—Efficiency." American
Economic Review, 56(June 1966), 372-415.

 

---------- . "Aspects of the X-Efficiency Theory of the Firm." The Bell
Journal of Economics, 6(Autumm 1975), 580-606.

 

---------- . Beyond Economic Man. Cambridge, Mass.: Harvard University
Press, 1976.

 

Loury, Glenn. "Market Structure and Innovation." Iguarterly Journal of
Economics, 93(August 1979), 395-410.

 

Maddala, G.S. Econometrics. New York: McGraw—Hill Book Co., 1977.

 

Mann, H. Michael, James Meehan, and G.A. Ramsey. "Market Structure and
Excess Capacity: A Look at Theory and Some Evidence." Review of
Economics and Statistics, 61(February 1978), 156-160.

 

138

Mann, H. Michael. "Seller Concentration, Barriers to New Entry, and
Rates of Return in Thirty Industries, 1950-1960." Review of
Economics and Statistics, 48(August 1966), 296-307.

 

Mansfield, Edwin. "Entry, Gibrat's Law, Innovation, and the Growth of
Firms." American Economic Review, 52(December 1962), 1023-1051.

 

---------- . "Technological Change and the Rate of Imitation." Econo-
metrica, 29(October 1961), 741-766.

Martin, Stephen. "Advertising, Concentration, and Profitability: The
Simultaneity Problem." The Bell Journal of Economics, 10(Autumn
1979), 639-647.

 

---------- . "Entry Barriers, Concentration, and Profits." Southern
Economic Journal, 46(October 1979), 471-488.

 

Menge, J.A. "Style Change Costs as a Market Weapon." Quarterly Journal
of Economics, 76(November 1972), 632-647.

 

 

Modigliani, Franco. "New Developments on the Oligopoly Front." Journal
of Political Economy, 66(June 1958), 215-232.

 

Nakao, Takeo. "Demand Growth, Profitability, and Entry." Quarterly
Journal of Economics, 94(March 1980), 397-411.

 

Nelson, Phillip. "Advertising as Information." Journal of Political
Economy, 82(July/August 1974), 729-754.

 

Orr, Dale. "The Determinants of Entry: A Study of Canadian Manufacturing
Industries." Review of Economics and Statistics, 56(February 1974),
58-66.

Phillips, Joseph. "Estimating the Economic Surplus." In Mbnopoly
Capital. Paul Baran and Paul Sweezy. New York: Mbnthly Review
Press, 1966, pp. 369-389.

Posner, Richard. "The Social Costs of MOnopoly and Regulation." Journal
of Political Economy, 83(August 1975), 807-827.

Primeaux, Walter J. "An Assessment of X-Efficiency Gained Through Com-
petition." Review of Economics and Statistics, 59(February 1977),

Romeo, Anthony. "Interindustry and Interfirm Differences in the Rate of
Diffusion of an Innovation." Review of Economics and Statistics,
57(August 1975), 311-319.

 

Scherer, F.M. Industrial Market Structure and Economic Performance.
Chicago: Rand McNally College Publishing Co., 1980.

139

---------- . "Research and Development Resource Allocation Under
Rivalry." (Quarterly Journal of Economics, 81(August 1967), 359-374.

 

Schumpeter, Joseph. Capitalism, Socialismg and Democracy. New York:
Harper, 1942.

 

Shepherd, William. The Economics of Industrial Organization. Englewood
Cliffs, N.J.: Prentice-Hall, 1979.

 

---------- . Market Power and Economic Welfare. New York: Random
House, 1970.

 

---------- . "Trends of Concentration in American Manufacturing Indus-
tries, 1947-1958." Review of Economics and Statistics, 46(May
1964), 200-212.

 

Siegfried, John, and Thomas Tiemann. "The Welfare Cost of Mbnopoly: An
Inter-Industry Analysis." Economic Inquiry, 12(June 1974), 190-202.

 

Spence, Michael. "Entry, Capacity, Investment and Oligopolistic Pricing.‘
Bell Journal of Economics, 8(Autumn 1977), 535-544.

 

Stigler, George. "The Statistics of Mbnopoly and Merger." Journal of
Political Economy, 4(February 1956), 33-35.

 

 

---------- . "A Theory of Oligopoly." Journal of Political Economy,
72(February 1967), 44-61.

Stonebraker, R.J. "Corporate Profits and Risk of Entry." Review of
Economics and Statistics, 58(February 1976), 33-39.

 

Sylos-Labini, Paolo. Oligppoly and Technical Progress. Cambridge,
Mass.: Harvard University Press, 1962.

 

Theil, H. Economic Forecasts and Polipy. Amsterdam: North-Holland
Publishing Co., 1961.

 

Tullock, Gordon. "The Welfare Costs of Tariffs, Mbnopolies, and Theft."
Western Economic Journal, 5(June 1967), 224-232.

 

Turner, Phillip. "Import Competition and the Profitability of United
Kingdom Manufacturing Industries." Journal of Industrial Economics,
29(December 1980), 155-166.

Walter, Ingo. "Non-Tariff Protection Among Industrial Countries: Some
Preliminary Evidence." Economia Internazionale, 25(May 1972), 333-
354.

 

1

Weiss, Leonard. "The Concentration-Profit Relationship and Antitrust."
In Industrial Concentration: The New Learning. Ed. H. Goldschmid,
et. a1. Boston: Little, Brown, 1974, pp. 201-220.

 

140

Wenders, John. "Excess Capacity as a Barrier to Entry." Journal of
Industrial Economics, 20(November 1971), 14-19.

 

 

Williamson, Oliver. Corporate Control and Business Behavior. Englewood
Cliffs, N.J.: Prentice Hall, Inc., 1970.

 

---------- . "Managerial Discretion and Business Behavior."

Economic Review, 53(December 1963), 1040-1047.

American

 

Winter, Sidney G. "Satisficing, Selection, and the Innovating Remnant."
Quarterly Journal of Economics, 85(May 1971), 237-261.

 

United States Government Public Documents
(Washington, D.C.: Government Printing Office)

 

Bureau of the Census. Annual Survey_of Manufactures. 1963.

 

---------- . Census of Manufactures. 1958, 1963, 1967.

 

Bureau of Mines. Aluminum, Mineral Industry Survey. 1959-1968.

Department of Commerce. U.S. Commodity_Exports and Imports as Related

Federal Trade Commission. Economic Report on the Baking Industry. 1967.

 

---------- . Economic Report on the Brewing_InduStry, 1978.

 

---------- . Economic Report on the Development and Structure of the U.S.
Electric Lamp Industry. 1980.

 

 

---------- . Economic Report on the Investigation of the Petroleum

Industry. 1973.

 

---------- . Economic Report on the Manufacture and Distribution of
Automotive Tires. 1966.

 

 

---------- . Economic Report on Mergers and Vertical Integration in the
Cement Industry. 1966.

 

 

---------- . Economic Report on the U.S. Steel Industry and Its Inter-
national Rivals. 1977.

 

 

---------- . Economic Report on the U.S. Sugar Industry. 1975.

 

---------- . Rates of Return in Selected Manufacturing Industries. 1959-

 

---------- . Staff Report of the Bureau of Economics on Cigarette
Advertising and Output. 1964.

 

 

141

Internal Revenue Service. Sourcebook of Income. 1959-1968.

 

Other Material

 

Annual Reports of Corporations.
Barrons. 1959-1968.

Business Week. 1959-1968.

 

Fortune Plant and Product Directory of the Largest 1000 U.S. Industrial
Corporations. 1961, 1964, 1967.

 

Moody's Industrial Manual. 1959-1968.

 

New York Times. 1959-1968.

 

Pit and Quarry. 1959, 1961, 1963, 1965, 1967, 1969.

 

Standard and Poor's Compustat annual industrial tape. 1959-1968.

---------- . Industrnyurveys. 1959-1968.

 

---------- . Register of Corporations, Directors, and Executives. 1959-

 

Wall Street Journal. 1959-1968.

 

Unpublished Sources

 

Brock, James. "The Photography Industry." Diss. Michigan State
University 1981.

Martin, Stephen. "Gaskins Models of Entry Impedance." Econometrics
Workshop Paper #7902. Michigan State University, April 1979.