Chapter 8

Imperfect Competition
 Reading
• Essential reading
– Hindriks, J and G.D. Myles Intermediate Public Economics.
(Cambridge: MIT Press, 2005) Chapter 8.
• Further reading
– Armstrong, M., S. Cowans and J. Vickers Regulatory Reform:
Economic Analysis and British Experience. (Cambridge: MIT
Press, 1994) [ISBN 0262510790 pbk].
– Baker, P. and V. Brechling (1992) ‘The impact of excise duty
changes on retail prices in the UK’, Fiscal Studies, 13, 48—65.
– Cowling, K.G. and D.C. Mueller (1978) ‘The social costs of
monopoly power’, Economic Journal, 88, 727—748.
– Gisser, M. (1986) ‘Price leadership and welfare losses in U.S.
manufacturing’, American Economic Review, 76, 756—767.
 Reading
– Harberger, A.C. (1954) ‘Monopoly and resource allocation’,
American Economic Review, 45, 77—87.
– Masson, R.T. and J. Shaanan (1984) ‘Social costs of oligopoly
and the value of competition’, Economic Journal, 94, 520—535.
– McCorriston, S. (1993) ‘The welfare implications of oligopoly in
agricultural input markets’, European Review of Agricultural
Economics, 20, 1—17.
– Peterson. E.B. and J.M. Connor (1995) ‘A comparison of
oligopoly welfare loss estimates for U.S. food manufacturing’,
American Journal of Agricultural Economics, 77, 300—308.
– Vickers, J. (1995), ‘Concepts of competition’, Oxford Economic
Papers, 47, 1—23.
– Vickers, J. (1996), ‘Market power and inefficiency: a contract
perspective’, Oxford Review of Economic Policy, 12, 11—26
– Waterson, M. Economic Theory of the Industry. (Cambridge:
Cambridge University Press, 1983) [ISBN 0521286867 pbk].
 Reading
• Challenging reading
– Delipalla, S. and M. Keen (1992) ‘The comparison between ad
valorem and specific taxation under imperfect competition’,
Journal of Public Economics, 49, 351—367.
– Delipalla, S. and O. O’Connell (2001) ‘Estimating tax incidence,
market power and market conduct: the European cigarette
industry’, International Journal of Industrial Organization, 19, 885
—908.
– Myles, G.D. (1996) ‘Imperfect competition and the optimal
combination of ad valorem and specific taxation’, International
Tax and Public Finance, 3, 29—44.
– Seade, J. (1985) ‘Profitable cost increases’, Warwick Economic
Research Paper, No. 260.
 Introduction
• Competitive price-taking supports economic
efficiency
• Imperfect competition arises when a large
economic agent can affect prices
– An advantage will be gained by exploiting this ability
– This must be detrimental to other economic agents
• Imperfect competition violates the assumptions
of the efficiency theorems
 Concepts of Competition
• Monopoly power occurs when the seller of a
product can influence prices
– A single seller is a monopolist
– There is oligopoly if there are several sellers
• Monopsony power occurs when the buyer of a
product can influence price
– A single buyer is a monopsonist
• Either price or quantity can be chosen
– Cournot oligopoly has quantity as strategic variable
– Bertrand oligopoly has price as strategic variable
 Concepts of Competition
• Products can be homogeneous or differentiated
• Product differentiation can be vertical (products
differ in quality) or horizontal (products differ in
specification)
• Non-price competition can accompany product
differentiation
– Advertising
– Investment
– Specification of product
 Market Structure
• Market structure refers to:
– The number of firms
– The size of firms
– Intensity of competition
• Firms are in the same market if their products are
close substitutes
– This arises if the cross-price elasticity of demand is
positive
• Empirical analysis invariably uses standard
industry classification
– This does not always guarantee close substitutability
 Market Structure
• Three dimensions are widely used to measure
intensity of competition
• Contestability represents the freedom of rivals to
enter an industry
– There can be legal monopoly rights
– Or entry barriers (economies of scale, entry-
deterrence)
• If a market is contestable the incumbent is
constrained
– The threat of entry prevents market power being
exploited
 Market Structure
• Concentration captures the number and
distribution of rival firms
• The n-firm concentration ratio is the total market
share of the n largest firms
• The Herfindahl index is the sum of squared
market share of all firms in the market

H  i si2
• Tab.8.1 reports the concentration ratio and
Herfindahl index for several US industries
 Market Structure
Industry Number of 4-firm concentration ratio Herfindahl
firms index
Cereal breakfast 33 0.87 0.221
foods
Pet food 130 0.61 0.151
Book publishing 2182 0.24 0.026
Soap and 683 0.65 0.170
detergents
Petroleum refining 200 0.32 0.044
Electronic 914 0.43 0.069
computers
Refrigerators / 40 0.85 0.226
freezers
Laundry machines 11 0.93 0.286
Greeting cards 147 0.85 0.283
Source: Concentration rations in Manufacturing, 1992, US Bureau of the Census
Table 8.1: Market concentration in US manufacturing, 1987
 Market Structure
• Collusiveness relates to the degree of
independence of firms’ strategies
– Sellers may agree to raise prices or reduce quantities
in unison
• A cartel agreement between firms is an explicit
form of collusion
• Collusion can also be tacit through the
reluctance to engage in competition
• Explicit collusion is illegal but more easily
detected than tacit collusion
 Welfare
• Imperfect competition is a form of market failure
• Firms with monopoly power restrict output to
raise price above marginal cost
• Consider a monopoly choosing output y given
inverse demand p(y) and marginal cost c
• Profit is  = p(y)y – c so the necessary condition
for maximization is
dp
p y c  0
dy
• Since dp/dy < 0 it follows that p > c so price
exceeds marginal cost
 Welfare
• Using the elasticity of demand  = pdy/ydp
pc 1

p 
• (p - c)/p is the Lerner index with value between 0
(competition) and 1 (maximum market power)
• The index provides a measure of market power
• When there is Cournot oligopoly with a
homogeneous product and m firms
pc 1 1

p m 

• The Lerner index is deflated by market share

 Welfare
• Inefficiency can also arise from incomplete
information
• Trade may sometimes not take place even when
two parties could gain
• If a buyer’s willingness to pay is unknown firm
may set price too high
• Let the firm assume the probability that value is v
or less be F(v)
• Firm chooses p to maximize profit
max p1  F  p  p  c 
 Welfare
• Fig. 8.1 shows standard
monopoly and incomplete Price
information Price

• If willingness to pay is
uniformly distributed then p* “Inverse” p*
“Expected” Demand
1  F  p
1 – F(p) = 1 – p and MR = Demand

1 – 2p c c
Marginal Marginal
• The profit-maximizing Revenue Revenue

price is * 1  c Quantity Prob.(Trade)

p  c Monopoly with Monopoly with

2 Uniform Pricing Incomplete Information

• This is the same outcome

as monopoly with linear Figure 8.1: Monopoly pricing
demand
 Welfare
• The welfare loss due to Price
monopoly is shown in Fig. A
8.2
pm B
• The competitive price is p
= c with quantity yc and E D
c
consumer surplus ADC
• The monopoly price is pm
and quantity ym
• Consumer surplus falls to MR AR
ABpm, profit is pmBEc ym yc Quantity
• BDE is the deadweight Figure 8.2: Deadweight loss with
loss of monopoly monopoly
 Welfare
• If the demand function is linear then deadweight
loss (BDE) is one half monopoly profit (pmBEc)
• Using the monopoly pricing condition relating
price and elasticity gives
pm ym
Deadweight loss  
2
• Several estimates are reported in Tab. 8.2
• These range from small (Harberger) to
significant (Masson and Shaanan)
• The larger estimates include the cost of
defending monopoly position
 Welfare
Author Sector Welfare loss (%)
Harberger US Manufacturing 0.08
Gisser US Manufacturing 0.11 – 1.82
Peterson and Connor US Food Manufacturing 0.16 – 5.15
Masson and Shaanan 37 US Industries 3
16
McCorriston UK Agricultural Inputs 1.6 – 2.5
20 – 40
Cowling and Mueller US 4 – 13
UK 3.9 – 7.2

Table 8.2: Monopoly welfare loss

 Tax Incidence
• Tax incidence is the study of the economic
effects of a tax
• Formal incidence refers to legal responsibility for
paying a tax
• Economic incidence refers to the changes in
behavior because of tax
• Economic and legal incidence are usually not
the same
• Economic incidence is a consequence of the
supply and demand curves for a good
 Tax Incidence
• Fig. 8.3 illustrates Price
economic incidence in a
competitive market with
constant marginal cost
• Supply is perfectly elastic q=c+t

at a price equal to t
p=c
marginal cost
• Before tax the equilibrium
price is p = c
• After tax the price is q = c
+t Quantity

• The tax falls entirely upon Figure 8.3: Tax incidence with
consumers perfectly elastic supply
 Tax Incidence
Price
• When the supply curve is
upward sloping the price
rises by less than the tax t
q
• The initial price is p
p
• A tax t is introduced
• Price rises to q
• The price rise q – p < t
• Some of the tax increase
falls on producers and Quantity
Figure 8.4: Tax incidence in
some on consumers
the general case
 Tax Incidence
• With imperfect Price
competition price is
above marginal cost and q
taxation can affect profit p

• Fig. 8.5 shows one b

c+t
possible outcome with t
monopoly c a

• A tax t is levied and price MR AR

rises from p to q
• The price increase is less yt yo Quantity
than the tax Figure 8.5: Tax undershifting
• This is tax undershifiting
 Tax Incidence
• It is also possible for price Price
to rise by more than the
tax increase q

• Fig. 8.6 illustrates tax

overshifting p
• The price rise is q – p c+t b
AR t
• This rise is greater than a
c
the tax increase t
• Overshifting can occur
MR
when the demand
yt yo
function is concave Quantity

• Estimates of tax shifting Figure 8.6: Tax overshifting

are given in Tab. 8.3.
 Tax Incidence
Baker and UK Beer UK Tobacco
Brechling 0.696 0.568
Delipalla and “Northern” EU “Southern” EU
O’Connell, 0.92 2.16
Tobacco
Tasarika, UK
Beer 0.665

Table 8.3: Calculations of tax shifting

 Tax Incidence
y2
• If there is oligopoly a tax
r1 y2 
can cause profit to rise
• In Fig. 8.7 join profit
m2
maximization occurs at b
• The Cournot equilibrium
c
is at c where the reaction
functions cross b

• A tax can move the r 2  y1 

Cournot equilibrium m1 y1
closer to joint profit Figure 8.7: Possibility of a
maximization profit increase
 Specific and Ad valorem
Taxation
• A specific tax is a fixed amount per unit of
output, q = p +t
• An ad valorem tax is a fixed proportion of the
producer price, q = [1 + ]p
• In a competitive market specific and ad valorem
taxation are equivalent: if t and  lead to the
same price they also raise the same revenue
• When there is imperfect competition the taxes
are not equivalent
 Specific and Ad valorem
Taxation
• Specific taxation raises cost
  qx  c  t x
• An ad valorem tax reduces revenue
1
 qx  cx
1
• An ad valorem tax raises more revenue than a
specific tax for the same consumer price
• Let c = t and observe the profit tax element
1 1
 qx  cx  qx  c  cx
1 1
 Specific and Ad valorem
Taxation
• The contrast between the
tax is shown in Fig. 8.8 Price Price

• The left-hand figure

shows a specific tax q q

raising tax-inclusive cost p p

• The right-hand figure ct

c
MR
1 
c
shows an ad valorem tax MR AR
AR
1  AR

shifting average revenue Quantity Quantity

() and marginal Specific Tax Ad Valorem Tax

revenue (MR)
• This explains the different Figure 8.8: Contrasting taxes
effects of the taxes
 Regulation of Monopoly
• The natural policy is to encourage competition
• This can be done directly by enforcing division of
monopolists
– US antitrust legislation applied to Standard Oil and
Bell System
• It can be done indirectly by reducing barriers to
entry
• Legal barriers can be removed by changing the
law
– But why were they imposed initially?
 Regulation of Monopoly
• Technological barriers can be reduced
insistence on knowledge sharing
– US insists Microsoft provides information
• Patents are also a barrier to entry
– Optimum length trades reward for innovation against
stifling of competition
• Advertising and excess capacity can be part of
an entry deterrence strategy
– Advertising expenditure can be limited (e.g. tobacco)
– Proving excess capacity is held to deter entry is
difficult
 Regulation of Monopoly
Price
• Competition cannot be
increased when there is
natural monopoly
• Fig. 8.9 illustrates natural
p
monopoly
• One firm can be profitable
AC
charging price p MC
• It is not possible for two MR 2 MR1 , AR 2 AR1
firms to make a profit y2 y1 Quantity
• This is a consequence of Figure 8.9: Natural monopoly
increasing returns
 Regulation of Monopoly
• Natural monopoly occurs when there are
significant fixed costs
– Water supply, electricity, gas, railways
• Public ownership (nationalisation) was the
standard response in Europe
– But this caused a lack of incentive to innovate
• Alternative is private ownership with regulation
– The rate of return or the rate of increase in prices
– The fixed cost (e.g. the rail track) may be placed into
a company separate to the operators (e.g. train
companies)
 Regulation of Oligopoly
• Collusion among firms allowed price to rise and
profit to increase
• Tacit collusion may be difficult for a regulator to
detect
– Does a high price represent lack of substitutes or
price collusion?
• This question is answered by calculating price
elasticities and using these to construct Lerner
index with and without collusion
– Breakfast cereal: collusion implies Lerner of 65-75%
– Competition implies Lerner of 40-44%
– Actual index about 45%, implying no collusion
 Regulation of Oligopoly
• Mergers can damage economic efficiency by
increasing monopoly power
• Mergers are regulated by governments
• A merger is not permitted if it is judged to harm
the public interest
• Estimated demand elasticities can be used to
predict the outcome of a merger
• Tab. 8.4 illustrates this method for a proposed
merger between Kleenex, Cottonelle,
ScotTissue)
• The predicted price changes favor the merger
 Regulation of Oligopoly
Bath Tissue Brand Market share Own-price Price change [cost change]
(%) elasticity (%)
Kleenex 7.5 -3.38 +1.0 [-2.4]
Cottonelle 6.7 -4.52 -0.3 [-2.4]
Scot Tissue 16.7 -2.94 -2.6 [-4.0]
Charmin 30.9 -2.75
Northern 12.4 -4.21
Angel Soft 8.8 -4.08
Private Label 7.6 -2.02
Other 9.4 -1.98
Market demand -1.17

Source: Data from Tables 1 and 2 in Hausman and Leonard (1997)

Table 8.4: Estimating the Effect of Merger in the Bath Tissue Market
 Unions and Taxation
• Unions create market power for their members
in labor markets
• Unions trade-off an increased wage rate against
reduced employment
• Let employed members receive w(1 – t) and
unemployed receive b
• If there are m members and the labor demand
function is n(w) the union maximizes
nw m  nw
U u w1  t   u b 
m m
 Unions and Taxation
• The utility-maximizing wage for the union
satisfies
u w1  t    u b 
u

•  u
is the union mark-up given by  u

 1 1 u  n 
where n is the elasticity of labor demand and u
the elasticity of utility
• This pricing rule shows that any increase in the
tax rate t is met by an increase in to keep w(1 – t)
constant
• Employment must fall and union members must
be worse off on average
 Monopsony
• A monopsonist is a single buyer
– Such as the only firm using a specialized form of labor
• Monopsony results in price (or wage) being
below the competitive level
• The monopsonist takes account of the fact that a
higher price (or wage) must be paid to all units
purchased
• This provides a disincentive to raising the wage
to the competitive level
• Monopsony causes a deadweight loss
 Monopsony
Wage
• Monopsony is shown in
Marginal
Fig. 8.10 Cost
• Labor demand is given by Labor Supply
the marginal revenue of w(L)

labor (MRL)
wc
• The competitive wage is
wc wm Labor Demand
MRL
• The marginal cost for the
monopsonist is w plus Lm Lc Quantity
additional payment to of Labor
existing workers Figure 8.10: Monopsony in the
• Monopsony wage is wm labor market