Theoretical Economics Letters, 2011, 1, 88-90
doi:10.4236/tel.2011.13018 Published Online November 2011 (http://www.SciRP.org/journal/tel)
Copyright © 2011 SciRes. TEL
Endogenous Sunk Cost,
Quality Competition and Welfare
George S. Ford1, Michael Stern2
1The Phoenix Center for Advanced Legal & Economic Public Policy Studies,
Washington, DC, USA
2Department of Economics, Auburn University, Auburn, USA
E-mail: ford@phoenix-center.org, sternml@auburn.edu
Received July 2, 2011; revised August 4, 2011; accepted August 12, 2011
Abstract
Competition in quality with escalating levels of endogenous sunk costs may produce levels of concentration
even higher than expected in their absence. We show that consumers may very well benefit from such
expenditures despite the effects on concentration and likely attenuation of price competition.
Keywords: Sunk Costs, Endogenous Sunk Costs, Quality Competition, Welfare, Market Structure
1. Introduction
The number of firms that can supply a given market is
not infinite, but bound by the underlying supply-side and
demand-side conditions of the market. An excellent pres-
entation of these ideas is provided in John Sutton’s
seminal treatise, Sunk Costs and Market Structure [1].
Under some simplifying assumptions such as Cournot
Competition and unit demand elasticity, the predicted
equilibrium number of firms in a market, N*, is deter-
mined by the formula:
S
N* (1)
where S is market size and
is the sunk cost of entry.
Thus, as market size grows, other things remaining con-
stant, the equilibrium number of firms rises. However, as
demonstrated in Sutton [1], this is true only if the level of
sunk cost is determined exogenously. By “exogenous” we
mean that entry costs are determined solely by the tech-
nology of production (e.g., plant size, start-up working
capital, and so forth), so that the firm has little discretion
in choosing the level of
.
In some cases quality competition may raise the fixed
and sunk costs of a firm’s participation in an industry
and, in turn, increase equilibrium industry concentration.
With only exogenous sunk costs, Expression (1) indi-
cates that the number of firms in equilibrium rises
monotonically (always) with market size and market con-
centration falls. With endogenous sunk costs, once mar-
ket size reaches some critical value, say S*, the incumbent
firms begin to intensify competition in quality by esca-
lating investment expenditures in R & D, advertising,
and activities that will raise the consumers’ willingness
to pay and attract customers away from firms offering
lower quality products. If the increased endogenous sunk
cost investments are large enough, then concentration
begins to rise as market size grows.
High concentration, particularly in large markets, in-
variably attracts the attention of policymakers and regu-
lators. A theoretical expectation that increases in mar-
ket size may result in higher, rather than lower, concen-
tration invites the question: Is the higher concentration
resulting from quality competition necessarily good or
bad for consumers and economic welfare? From a wel-
fare perspective, the reduced well being of consumers
from higher prices due to higher concentration (at least
under some assumptions of competitive interaction) is
offset, to some extent, by consumers’ higher willing-
ness-to-pay for better products and services. Sutton [1],
however, eschews welfare analysis, thereby providing no
answer. In this Letter, we fill in that gap.
2. Analysis
Since our interest in this topic is drawn from Sutton’s
book, we look there to begin our analysis. In Section
3.2, Sutton utilizes a Cobb-Douglas specification for
modeling consumer utility. This choice implies that the
representative consumer chooses a good that maximizes
the ratio of quality (u) to price (p). The quality-to-price
ratio (u/p) is a simple index of consumer welfare. This
89
G. S. FORD ET AL.
metric is superior to alternatives such as aggregate utility
since it does not automatically scale when market size is
increased due to either an increase in the number of con-
sumers or consumer income.
Let market concentration be defined as x = 1/N, where
N is the number of firms. Inserting this measure of con-
centration into Equation (3) from Sutton [1] and rear-
ranging yields:
32
222
a
xxx
S


 

 (2)
where a is the cost per advertising message, and meas-
ures the returns to advertising or other expenditures which
increase consumers’ willingness-to-pay. Given a set of
parameter values (a, , ) and market size (S), this cubic
equation yields the equilibrium level of market concen-
tration when firms are in the interior of their profit
maximization problem with respect to quality (u > 1). If
the market size is relatively small, the firms will be on
the quality boundary (u = 1) and the equilibrium concen-
tration is given by:
2
xS
(3)
The critical point (S*, x*) where the equilibrium con-
centration switches from (3) to (2) is characterized by the
solution to the system:
01
2
2
2
 x
a
x
(4a)
and 2
/Sσ
x
(4b)
The above system merely involves a quadratic equa-
tion, and is therefore easy to solve. With the critical point
(S*, x*) calculated, Equation (3) can be used to calculate
the equilibrium concentration for any market size, S S*.
Finally, Equation (2) can be used to calculate the equilib-
rium concentration for any market size S > S*.
After calculating the equilibrium concentration level
for a given market size, one can calculate the equilibrium
price and quantity in the market. The equilibrium price is
given at Sutton [1, p. 50], and can be written in terms of
concentration:
(1 )pc x, (5)
where c is marginal cost. Whenever S S*, the equilib-
rium quality is simply equal to one. However, when S >
S*, the equilibrium quality level can be calculated using
Equation (1) on Sutton [1, p. 54]. Writing the equation in
terms of concentration and substituting in for the fixed
cost function (F) yields the following characterization of
equilibrium quality:
1/
2
2(1 )uSxx
a


1
u if S S* (6b)
Finally, the ratio of the equilibrium quality (u) to market
price (p) can now be formed in order to characterize con-
sumer welfare.
Generally speaking, in a model with fixed costs, one
would expect efficiency gains from increased market size.
Given the endogenous entry and zero profit condition in
Sutton’s model, firm welfare cannot rise and hence we
would expect the welfare gains to be captured by con-
sumers. Before firms invest in quality (u), an increase in
market size simply results in more firms, a lower price
level, and hence higher consumer welfare.
After S*, firms begin investing in quality, the number
of firms slowly falls, and the market price rises (by the
Cournot assumption), but the rate of increase in quality
will generally far outweigh the higher market price. Thus,
the consumer welfare ratio (u/p) will continue to rise as
market share increases.
As an example, we plot in Figure 1 the consumer wel-
fare ratio (u/p) as a function of market size (S) for the
following parameter values: a = 300,
= 101,
= 2, and
c = 1. The simulation produces the expected size-con-
centration relationship with S* = 1000. Consumer wel-
fare rises as market size increases, even after the concen-
tration “switch point” (S*). Thus, our analysis demon-
strates that when higher concentration results from com-
petition in quality with escalating endogenous sunk costs
where those costs are incurred to raise willingness-to-pay,
consumer welfare rises despite the reduction in the equi-
librium number of firms.
3. Conclusions
Competition in quality with escalating levels of endoge-
nous sunk costs may produce levels of concentration
even higher than expected in their absence. But we show,
based on the analysis in Sutton [1], that consumers may
very well benefit from such expenditures despite the ef-
fects on concentration and likely attenuation of price
competition. This result may have significant public pol-
if S > S* (6a) Figure 1. Welfare and market size.
Copyright © 2011 SciRes. TEL
G. S. FORD ET AL.
Copyright © 2011 SciRes. TEL
90
icy relevance for competition in communications mar-
kets, where sunk costs are prevalent and quality com-
petetion is a primary instrument of rivalry.
4. Acknowledgements
The authors are grateful for helpful comments and sug-
gestions on this work provided by Randy Beard, Jerry
Duvall, and Lawrence Spiwak.
5. Reference
[1] J. Sutton, “Sunk Cost and Market Structure,” MIT Press,
Cambridge, 1991.