Dis cus si on Paper No. 08-071Barriers to Entry and Profitability
Diana Heger and Kornelius Kraft
Dis cus si on Paper No. 08-071Barriers to Entry and Profitability
Diana Heger and Kornelius Kraft
Download this ZEW Discussion Paper from our ftp server:
Die Dis cus si on Pape rs die nen einer mög lichst schnel len Ver brei tung von
neue ren For schungs arbei ten des ZEW. Die Bei trä ge lie gen in allei ni ger Ver ant wor tung
der Auto ren und stel len nicht not wen di ger wei se die Mei nung des ZEW dar.
Dis cus si on Papers are inten ded to make results of ZEW research prompt ly avai la ble to other
eco no mists in order to encou ra ge dis cus si on and sug gesti ons for revi si ons. The aut hors are sole ly
respon si ble for the con tents which do not neces sa ri ly repre sent the opi ni on of the ZEW.
This paper focuses on barriers to entry, firm profitability and the number of firms in a market.
Usually, it is expected that high profits attract entry, which is particularly important if the
highly profitable firm has a dominant position and holds a large market share. In light of the
mechanisms by which markets adjust, the absence of barriers to entry is fundamental for eco-
nomic welfare. It is necessary that a situation of considerable market power with high profits
attracts entry by challengers and that by this intensification of the competitive pressure the
profits adjust to a “normal” level.
Baumol et al. (1982) find that market performance depends in an essential way on the impor-
tance of potential entry. The problem this raises for econometric work is that potential entry is
an unobservable variable. Usually, the effects of entry are identified by the realized entry of
challengers. Using survey-based data, we are able to investigate the effect of the threat of en-
try. Information from top managers, on their perception of how strong their own competitive
position is threatened by a likely entry of competitors into their main markets, is applied to
estimate what effect this threat of entry has on profitability. We confirm that threat of entry
has a disciplinary effect on the price-setting behaviour of firms; we find a significant negative
effect of threat of entry on firm profitability.
In a second step, we derive a theoretical model linking the optimal number of firms in a mar-
ket to fixed costs. Fixed costs should, at least partly, reflect sunk costs which are assumed to
build a barrier to entry. The model predicts that the number of firms in a market depends
negatively on fixed and marginal costs. We test this model empirically and confirm the con-
jectured effects. Furthermore, we test whether firms cut prices to such an extent that entry is
not profitable. Since threat of entry turns out to have no significant effect on the number of
competitors, we can confirm that the cut of prices is sufficient to prevent entry.
Diese Studie betrachtet Markteintrittsbarrieren, Unternehmensprofitabilität und die Anzahl
von Unternehmen in einem Markt. Üblicherweise wird angenommen, dass hohe Gewinne
Markteintritt hervorrufen, was insbesondere wichtig ist, wenn hochprofitable Unternehmen
eine dominante Position und einen großen Marktanteil innehaben. In Bezug auf die Marktme-
chanismen ist das Fehlen von Markteintrittsbarrieren wichtig für die Wohlfahrt. Es ist unab-
dingbar, dass in einer Situation beträchtlicher Marktmacht einhergehend mit hohen Gewinnen
Markteintritt möglich ist und dass sich durch die Intensivierung des Wettbewerbsdrucks die
Gewinne wieder auf ein „normales“ Niveau einpendeln.
Baumol et al. (1982) zeigen, dass die Marktperformance wesentlich von der Bedeutung von
potenziellem Markteintritt abhängt. In ökonometrischen Studien wirft dies das Problem auf,
dass potenzieller Markteintritt nicht beobachtbar ist. Üblicherweise wird dieser Effekt über
sich tatsächlich manifestierten Markteintritt approximiert. Wir sind hingegen in der Lage,
durch Nutzung von Unternehmensbefragungen den Effekt von Markteintrittsdrohungen zu
analysieren. Informationen von Unternehmensmanagern bzgl. ihrer Wahrnehmung, wie stark
die eigene Wettbewerbsposition durch einen wahrscheinlichen Eintritt von Wettbewerbern in
ihrem Hauptabsatzmarkt bedroht ist, werden verwendet, um den Effekt zu schätzen, inwieweit
die Profitabilität durch Markteintrittsdrohung beeinflusst wird. Wir können den disziplinie-
rende Effekt von Markteintrittsdrohung auf das Preissetzungsverhalten der Unternehmen bes-
tätigen, da wir einen signifikant negativen Effekt der Markteintrittsdrohung auf die Profitabi-
In einem zweiten Schritt leiten wir ein theoretisches Modell her, das die optimale Anzahl von
Unternehmen in einem Markt mit Fixkosten verbindet. Fixkosten sollten zumindest teilweise
Sunk Costs abbilden, die als Markteintrittsbarriere angesehen werden. Das Modell sagt vor-
her, dass die Anzahl der Unternehmen in einem Markt negativ von fixen und marginalen Kos-
ten abhängt. Wir testen dieses Modell empirisch und können die theoretisch abgeleiteten Ef-
fekte bestätigen. Darüber hinaus, testen wir, ob Unternehmen ihre Preise so stark reduzieren,
dass Markteintritt nicht profitabel ist. Da die Markteintrittsdrohung keinen signifikanten Ef-
fekt auf die Anzahl der Unternehmen hat, können wir bestätigen, dass die Preissenkungen
ausreichend sind, um Markteintritt zu verhindern.
Barriers to Entry and Profitability
Diana Heger* and Kornelius Kraft**
September 2008 Abstract
Barriers to entry are regarded as major impediments to the working of
markets. Entry must not necessarily actually take place - the perceived
threat of entry may encourage incumbent firms to behave as if they are
in a competitive market, even if they are not. We present empirical
evidence on effects of perceived threat of entry on profitability. Using
information from managers about how they assess the existence of en-
try barriers a strong impact of these assessments on profitability is
confirmed. The number and the relative size of competitors also exert
considerable effects. We find no statistically significant relation be-
tween the perceived threat of entry and the actual number of firms if
the size of the relevant market is taken into account. Keywords:
Barriers to Entry, Profitability, Discrete Regression Models JEL-Classification:
L13, L25, C25
* Centre for European Economic Research (ZEW)
** University of Dortmund
Industrial Economics and International Management
Phone: +49 621 1235-382
Phone: +49 231 755-3152
Fax: +49 621 1235-170
Fax: +49 231 755-3155
*, ** Thanks to Georg Licht and the participants of the Brown Bag Seminar for highly valuable comments.
The observation of short-run high profits is not incompatible with the existence of a long-run
competitive equilibrium. Prospective high profits are needed to provide incentives for innova-
tion or any other activity to improve the efficiency of a firm. However, it is expected that in
the long-run excessive profits are competed away by reactions of the present competitors, and
furthermore, that high profits attract entry, which is particularly important if the highly profit-
able firm has a dominant position and holds a large market share.
In light of the mechanisms by which markets adjust, the absence of barriers to entry is funda-
mental for economic welfare. It is necessary that a situation of considerable market power
with high profits attracts entry by challengers and that by this intensification of the competi-
tive pressure the profits adjust to a “normal” level.
Economic theory has discussed extensively the conditions for barriers to entry and the effects
thereof. Although there is no consensus about the exact definition of what a barrier to entry
actually is, it is undeniable that barriers to entry play an important role in a wide variety of
competition issues. Entry barriers can retard or even entirely prevent the working of a market
and welfare may be seriously affected by them.
There are quite a few empirical studies on the determinants of barriers to entry and also some
on the effects of entry on profitability or other variables of interest, e.g. innovation. Martin
(2002, 221) notes: “Another such strand is the argument, going back to Bain (1956) and re-
cently re-emphasized by Baumol et al. (1982), that market performance depends in an essen-
tial way on the importance of potential entry. The problem this raises for econometric work is
that potential entry is an unobservable variable.” Usually, the effects of entry are identified by
the realized entry of challengers. However, Martin (2002, 221) argues this is not a convincing
way to model potential entry, as most actual entry is short and unsuccessful.
Our contribution is a different one: We investigate the effect of the threat of entry. Informa-
tion from top managers, on their perception of how strong their own competitive position is
threatened by a likely entry of competitors into their main markets, is applied to estimate what
effect this threat of entry has on profitability.
Aside of the value of using information on the perceived threat of entry we are able to identify
the relevant markets as assessed by the managers themselves.
2 General Theoretical Considerations
Entry and exit conditions are important factors that determine existing firms’ possibilities to
exert market power. A dominant firm with a very high market share might not be able to
make use of its position, if any significant deviation of the price from marginal costs will lead
to entry by new competitors. Entry by new firms can also affect innovativeness and put pres-
sure on the existing firms not only to refrain from misusing their market power, but also to
operate as efficiently as possible. Therefore, cost conditions might also improve. Hence mar-
ket shares and concentration are just one part of the story; in addition, the conditions for po-
tential competition are important contributors to the functioning of markets.
A firm, deciding whether to penetrate a market or not, compares the benefits and costs of en-
try. The benefits are the expected profits and growth of demand connected with entry. The
costs are, among other things, determined by barriers to entry, which may be caused by ex-
ogenous factors like economies of scale1 or by strategic factors like excess capacity, limit
pricing or advertising.
In the literature, the first important contribution to the discussion of entry barriers is Bain
(1956). Following this a lively debate took place which, however, did not succeed in finding a
generally accepted definition. Bain defined a barrier to entry by its effects on profitability, in
particular in terms of the ability to earn above-normal profits without inducing entry. Stigler
(1968) later defines an entry barrier as a cost advantage of incumbents over entrants and von
Weizsäcker (1980) argues that a cost differential is only an entry barrier if it reduces welfare2.
The discussed reasons for barriers to entry are manifold. Economies of scale may or may not
be regarded as entry barriers. Clearly, with large scale economies there is only place for a few
producers in an industry, and thus entry might be difficult. However, in the view of Baumol,
Panzar and Willig (1982) it is the nature of the cost structure which determines entry barriers.
In the absence of sunk costs, entry is not impeded and every firm presently active in the in-
dustry is disciplined by potential entry. Therefore, prices will be close to average costs. The
problem is that in most industries a part of the costs will always be sunk.
1 Whether this is really a barrier to entry is debated.
2 See McAfee, Mialon and Williams (2004) on this issue.
Excess capacity plays an important role in the theoretical discussion. Spence (1977), Dixit
(1979, 1980), Bulow, Geanakoplos and Klemperer (1985) are among the first to point out the
asymmetry of an incumbent and a potential entrant. Typically, in such models the incumbent
selects a level of capacity in the first period and the potential entrant and the incumbent simul-
taneously determine quantities in the second period. These models assume that the incumbent
produces at or below the capacity limit in the second period and that the incumbent’s mar-
ginal costs are lower than the potential entrant’s marginal costs because the incumbent is able
to avoid the costs associated with expanding capacity in the second period. Hence, the incum-
bent enjoys a first mover advantage.
Another cause of entry barriers is product differentiation. Consumers view products as imper-
fect substitutes for a number of reasons, such as different varieties (horizontal product differ-
entiation) or product quality (vertical product differentiation). If introducing a new brand is
connected with significant fixed costs, horizontal product differentiation may well lead to
persistent entry barriers. Shaked and Sutton (1982, 1983) analyse a game where firms choose
whether to enter at the first stage of the game, choose quality at the second stage and prices at
the third stage. Surprisingly, they show in their model that only a few and at the limit only one
firm will operate in the industry despite of free entry.
Obviously, firms have an interest in product differentiation and they will attempt to increase
the perceived difference or quality advantage of their products by the use of advertising. Ad-
vertising intends to increase consumers’ loyalty to specific brands, and therefore, to deter en-
try. Clearly, advertising expenditures are sunk costs and can as such increase the impediments
to enter a market. This is the way advertising is introduced into the Sutton model. It can, how-
ever, also be argued that advertising informs consumers about the existence and the character-
istics of new products, and thus, eases entry.
Another possible way of incumbents to raise entry barriers is by innovation. Bain (1956) al-
ready identifies absolute cost advantages as a major reason for entry barriers, and obviously,
process innovation that aims at cost reductions. Secondly, newly developed products will lead
to (at least temporary) advantages compared to competitors and therefore R&D activities can
lead to reduced entry. At this point, it can also be argued in the opposite direction, as many
entering firms are new foundations of innovators, who want to benefit from the market poten-
tial of their inventions. Hence innovativeness might spur entry and not impede it.
3 A Model with Fixed Costs
We consider a representative firm i
which competes with n-1
other producers in a Cournot
oligopoly3. In addition to constant marginal costs c
every firm has to cover fixed costs F,
which are completely sunk4. The inverse demand function is defined in the following (stan-
(1) p = a − bq − b(n −1)q
where the price demanded by firm i
) depends on the reservation price, the output level of
) and the individual outputs of all other competitor firms j (qj
) . All outputs are char-
acterized by the same price elasticity b
These assumptions lead to a simple profit function:
π = (a − bq − b(n −1)q )q − cq − F .
Optimizing this function with respect to output q , and solving, assuming homogenous firms
with identical output levels, leads to:
p − c
a − bnq − c
a − c
Equation (3) shows that the optimal output level q
of every competitor depends negatively on
the marginal costs c
and on the number of firms n
active in the market.
Since a general assumption is that entry occurs until all profit is dissipated, we determine the
optimal number of firms in a market by solving the zero profit condition. As firms’ sales (s
are identical, they can be defined as the average industry sales (S/n
) determined by the opti-
mal output level q
derived in equation (3).
a − c
− cq − F = − c
− F = 0 .
3 Breshnahan and Reiss (1990, 1991) and Berry (1992) discuss the relation between market size and the number
of firms. Clearly, Sutton (1991, 1998) is also highly relevant if this question is considered. Cf. also the surveys
by Berry and Reiss (2006) and Sutton (2006).
4 The assumption that all fixed costs are sunk is used for simplicity. Any sunk costs larger than zero would pro-
duce results similar to the ones presented below. If fixed costs are not sunk an entrant would not take them into
account, as they can be recovered after leaving the industry. However in practice a part of capital will always be
Solving for n
, leads to the following expression:
Sb − ac + c − bF + ((Sb − ac + c − bF) + 4b FS)
The industry sales volume S
is simply defined as S=pnq
. Replacing q
by the optimal output
determined by (3) and also including (3) into the inverse demand function (1) leads to:
(a + nc)n(a − c)
S = pnq =
This relation is inserted into the equation (5), and solving for n
, leads to an explicit expression
for the number of firms in a market (n
a − c
n = 1
This relation implies that the maximal number of firms falls with a larger fixed costs F
as with higher marginal costs c
, and rises with a
. More interesting, however, is the average
firm size, as the costs have opposing effects: On the one hand in the absence of entry barriers
higher costs lead to a lower output level. On the other hand fixed as well as marginal marginal
costs restrict the number of firms and therefore increase the market shares of the existing
firms. Hence, both the nominator and the denominator of n/S
are affected by the cost condi-
tions and it remains unclear which effect dominates. The number of firms in an industry is
determined by the following condition:
bF + c(bF)
The number of firms in an industry is negatively affected by fixed costs F
as well as by mar-
ginal costs c
. Thus, this implies in turn that average firm size is positively affected by both F
. Given that output is reduced if marginal costs c
rise, this result is not self-evident. Sur-
prisingly, the constant term of the demand curve, a
, has no effect on the maximal number of
The model is based on the assumption of entry until profits are dissipated. In such a scenario,
by construction a connection between fixed costs and profitability cannot exist as profits are
always zero. However, this is clearly the result of ignoring the integer constraint on the num-
ber of competitors n
. If the integer condition is taken into account, raising fixed costs can lead