CHAPTER -5

MONOPOLICTIC COMPETITION

Although the conditions necessary for the existence of perfect competitive and monopoly are unlikely to
be found in the real world, an analysis of these market structures is important because it provides insights
into more commonly encountered industry types. These insights provide guidance in the formulation of
public policy to promote the general economic welfare. Unlike monopolies, perfectly competitive firms
produce at minimum per unit cost. Thus, perfectly competitive market structures result in an efficient
allocation of society’s scarce productive resources and tend to maximize consumer and producer surplus.
Although models of perfect competition and monopoly are useful, it is important analytically to bridge the
gap between these two extreme cases. The first significant contributions in this direction were provided by
Edward Chamberlin and Joan Robinson. These economists observed that in many intensely competitive
markets individual firms were able to set the market price of their product.

Since these firms exhibit characteristics of both perfect competition and monopoly, this market structure is
referred to as monopolistic competition. The market power of monopolistically-competitive firms is
derived from product differentiation and market segmentation. Through subtle and no so- subtle
distinctions, each firm in a monopolistically competitive industry is a sort of mini-monopolist. But, unlike
a monopolist, the ability of these firms to set the market price for their product is severely constrained by
the existence of many close substitutes. Thus, the demand for the output of monopolistically competitive
firms is much more price elastic (flatter) than the demand curve confronting the monopolist.

A firm in a perfectly competitive industry faces a perfectly-elastic (horizontal) demand curve because its
output is a perfect substitute for the output of other firms in the industry. Unlike monopolies and
monopolistically-competitive firms, which may be described as price makers, perfectly-competitive firms
are price takers. Monopolistic competition is an example of an imperfectly competitive industry. Firms in
these industries exercise a degree of market power, albeit less than that exercised by a monopoly. As in
the cases of perfect competition and monopoly, profit-maximizing monopolistically competitive and
oligopolistic firms will produce at an output level where MR = MC. The characteristics of a
monopolistically competitive industry are a large number of sellers acting independently, differentiated
products, partial (and limited) control over product price, and relatively easy entry into and exit from the
industry.

1
Product differentiation refers to real or perceived differences in goods or services produced by different
firms in the same industry. Product differentiation permits market segmentation, which enables individual
firms to set their own prices within limits. As in the case of a monopoly, each firm in a monopolistically
competitive industry faces a downward-sloping demand curve, which implies that P > MR. The short-run
profit-maximizing condition for a monopolistically competitive firm is P > MR = MC. As in the case of
perfect competition, thefirm earns economic profit when P >ATC, which will attract new firms into the
industry. As new firms enter the industry, existing firms lose market share. This is illustrated graphically
by a shift to the left of each firm’s demand curve. If P <ATC, the firm earns an economic loss, which will
cause firms to exit the industry resulting in an increase in market share and a shiftto the right of the
demand curve. In the long run, the firm earns no economic profit since P = ATC. The demand curve for
the firm’s product is just tangent to the firm’s average total cost curve. The long-run competitive
equilibrium in monopolisticallycompetitive industries is P = ATC > MR = MC. As in the case of
monopoly, since MC < ATC, per unit cost is not minimized, i.e., monopolistically competitive firms
produce inefficiently in the long run. Advertising is an important element of monopolistic competition
because it reinforces customer loyalty by highlighting real or perceived product differences between and
among products of firms in the industry. The optimal level of advertising expenditures maximizes the
firm’s profits,which occurs when the firm produces at an output level where marginal cost (which include
incremental advertising expenditures) equals marginal revenue. When compared with the model of perfect
competition, in many respects monopolistic competition is considered an inferior market structure. As in
the case of monopoly, the demand curve confronting a monopolistically competitive firm is downward
sloping. Thus, monopolistic competition results in lower output levels and higher prices when compared
with perfect competition. Moreover, monopolistically competitive firms do not produceat minimum per
unit cost. On the other hand, although production is less efficient when compared with perfect
competition, the consumer is rewarded with the greater product variety. As in the case of perfect
competition, relatively easy entry into and exit from the industry encourages product innovation and
development. In the long run, monopolistically competitive firms earn only a normal rate of return.

In the first part of this course we examined two "pure" market structures: Perfect Competition and Pure
Monopoly. We defined perfect Competition as the form of market organization in which there are many
sellers of a homogeneous product. Moreover, we defined Pure Monopoly as a single seller of a
commodity for which there are no close substitutes. Between these two extreme forms of market
organization lies monopolistic competition and oligopoly. In this chapter we consider monopolistic
competition. It refers to the case in which there are many sellers of a heterogeneous or differentiated

2
product & entry into or exit from the industry is rather is easy in the long run. In summary, an industry is
characterized as a monopolistically competitive if:
 there are many buyers and sellers.
 each firm in the industry (or Product group) produces differentiated product that are close
substitute.
 there are free entry into and exit from the industry.
Thus, monopolistic competition is a market structure in which a relatively large number of small
producers are offering similar, but not identical products. Note: as the name implies, monopolistic
competition is a blend of (perfect) competition & monopoly. The Competitive element arises because:
there are many sellers in the market, each of which is too small to affect the other sells, and
Firms can enter & leave a monopolistically competitive industry easily in the long-run.
The monopolistic element arises from product differentiation. That is, since the product of each seller is
similar but not identical, each seller has a monopoly power over the specific product it sells.
Chamberlin made other heroic assumptions of monopolistic competition. That is,
 all firms face the same demand conditions which implies that consumers' preferences are evenly
distributed among different sellers; and
 all firms also face the same cost conditions, i.e. the differences in products does not give rise to
differences in costs.

These assumptions help to show the equilibrium of the firm and market on the same diagram.

Product Differentiation & the Demand Curve:

Differentiated Products are products that are similar but not identical. The similarity of differentiated
products arises from the fact that they satisfy the same basic consumption needs. Examples include
numerous brands of breakfast cereals, toothpaste, cigarettes and cold medicines on the market today. The
product differentiation may be real (when the inherent characteristics of the products are different) or
imaginary (when the products are basically the same but consumers are persuaded that the products are
different)

The real differentiation exists when there are differences in the: Specification of products, or Factor inputs
used. The case of the various breakfast cereals with various nutritional and sugar contents is best example
of real differentiation.

3
Imaginary (Fancied) differentiation is established by: Advertising, or Differences in packaging, or
Differences in product design, or Brand name; example: The case of different brands of aspirin, all of
which contain the same ingredients.

Demand Schedule: since the product of each seller is similar but not identical, each seller has a monopoly
power over the specific product it sells. This monopoly power, however, is severely limited by the
existence of close substitutes. Thus, these product differentiations create brand loyalty of consumers and
give rise to downward sloping elastic demand curve. That is, consumers are willing to pay a higher price
to enjoy the advantage of product differentiation. Since this differentiation is slight the firm faces highly
elastic demand curve.

Price Highly elastic

demand curve

0 quantity

Industry and Product Group:

Because each firm produces a somewhat different product under monopolistic competition, we cannot
define the industry, which refers to the producers of identical product. To overcome this difficulty,
Chamberlin redefined industry as a "product-group"- a group of firms/producers producing similar (or
closely related) products. Here, for simplicity we will continue to use the term "industry" (to refer to all
the sellers of the slightly differentiated products in a product group).

Price & Output Determination in the Short-run and Long run:

Before we examine equilibrium of monopolistically competitive firm let us briefly explain some basic
concepts.

*Costs: Chamberlin adopts the shape of costs of the traditional theory of the firm. That is, ATC,
AVC and MC curves are all U-shaped; and AFC has a geometric hyperbola shape.
*Actual-Sales (or share- of the market) Demand Curve: It is also called proportional (or prorata) demand
curve. It shows the actual sales of a firm at each price after accounting for the adjustments of the prices of
other firms in the group. Alternatively, it is the amount of demand going to a typical firm when all firms
are charging the same price. Here all firms adjust price simultaneously but independently.

4
*Perceived (or Planned) demand curves: the amount of demand going to a typical firm when there is no a
simultaneous price adjustment by other firms.
Actual-sales curve (DD) is derived from perceived
demand curve (dd). If price is P0, the firm's sale
Derivation of Prorata Demand Curve: could be Q0. When this typical firm reduces price to
P1 perceiving that his sales will rise to Q’0, all other
firms will adjust their price to P1 simultaneously but
Price independently and hence the typical firm will actually
sale only Q1 amount on a downward shifted
d D perceived demand curve (d1d1). If any one firm
d’ further reduces price to P2 the perceived demand
P0 curve slide-down to d2d2 because of simultaneous
price adjustment by all other firms. Thus, at P 2 a firm
d’’
plans to sell Q’1 but actually sell only Q2 on d2d2
(planned demand curve). Thus, the proportional
P1 demand curve (DD) is the locus of shifting perceived
demand curves (dd) as competitors, acting
P2 simultaneously, change their price. A movement
d along the DD curve shows changes in the actual sales
d’ of existing firms as all of them adjust their price
D d’’ simultaneously, with their share remaining constant.
0 Q0 Q1 Q2 Q’0 Q’1 Quantity However, the shift in DD curve is caused by entry of
new firms into and exit of existing firms from the
industry, and shows a decline (entry) or increase
(exit) in the share of the firm.

Short-Run and Long-run Equilibrium:

Short-Run Equilibrium: Just like any other firm, the monopolistically competitive firm will produce as
long as the marginal (extra) revenue from selling output exceeds the extra (marginal) cost of producing
that output. In the short-run, maximum profit (or equilibrium) occurs when these extra revenue (MR) and
extra costs (MC) are equal. See point E of figure (a) below. In general, short-run equilibrium conditions
are:
MR = MC and MC is rising, &
Proportional demand curve intersects perceive demand curve.

5
Graphical illustration of equilibrium:
P/C
P/C
MC
D D
MC AC
AC

Pe C A
Pe

A
B
d
E e
d
D D

Q 0 Q
0 Qe Qe
MR MR

Figure (a)—short run equilibrium Panel (b)—long run equilibrium

Long-Run Equilibrium: In monopolistic competition entry & exist are free in the long run. Firms can
enter an industry when there are excess (or abnormal) profits to be made, and firms suffering losses can
fold-up and go out of business. Consider the case where short-run excess profits are there, i.e. panel (a) of
figure above shows excess/abnormal/supernormal profit of area ABCPe. The abnormal profits will attract
new competitors into the market.

The result of new entry is a downward shift of the perceived demand curve (dd) and the proportional demand curve
(DD), since a large number of sellers share the market. Assuming that the new cost curves will not shift as entry
occurs; each shift to the left of the planned demand curve will be followed by a price adjustment as the firm reaches
a new equilibrium position, equating the the shifted marginal revenue curve to its marginal cost. This process will
continue until the perceived demand is tangent to the average cost curve, & hence the excess profits are wiped out
(represented by point A). In short, long-run equilibrium is defined by three conditions:

1) MR = MC & MC must be rising.

2) The perceived demand curve (dd) must be tangent to the LAC curve, representing zero economic profits.
3) The proportional demand curve (DD) must intersect both the planned demand curve (dd) & LAC curve at a point of
their tangency.

6
 CHAPTER 6
OLIGOPOLY
Definition: Oligopoly is the form of market organization in which there are few sellers of homogeneous or
differentiated product. If there are only two sellers in the market, we have a duopoly. That is, duopoly is a special
case of oligopoly in which there are only two sellers in the market. If the product is homogeneous, we have a pure
oligopoly. If the product is differentiated, we have a differentiated oligopoly.

Because there are only few firms selling a homogeneous or differentiated product in oligopolistic markets, the
action of each firm affects the other firms in the industry, and vice versa. Thus, it is clear that the distinguishing
characteristic of oligopoly is the interdependence or rivalry among firms in the industry. This interdependence is
the natural result of fewness.

The sources of oligopoly are generally the same as for monopoly, that is:
1. Economies of scale, which enable few firms supplying the entire market.
2. Huge capital investment & specialized inputs are required to enter oligopolistic industry (eg. Automobiles,
aluminum, steel,8 ...)
3. Patent right (exclusive right to produce a commodity or to use a particular production process).
4. Established firms may have loyal following customers
5. Control over entire supply of raw materials.
6. The government may award a franchise to only few firms to operate.

Types of Oligopoly: In general, oligopoly market is divided into two broad categories:

I. Non-collusive Oligopoly: in which firms may be better rivals of each other through advertising, product
differentiation, and so on.
II. Collusive Oligopoly: in which firms form coalition and cooperate.

I. NON-COLLUSIVE OLIGOPOLY

Since an oligopolist knows that its own actions will have a significant impact on the other Oligopolists in the
industry, each oligopolist must consider the possible reaction of competitors in deciding its pricing policies, the
degree of product differentiation to introduce, the level of advertising to undertake, & so on.

Because competitors can react in many different ways, we do not have a single oligopoly model. Here we present
some of the most important Oligopoly models:

A) Cournot's Model (lies between competition & monopoly)

B) The Stackelberg model (extension of Cournot model).
C) The "Kinked-Demand" model
D) Bertrand's Model (similar with perfect competition)
E) Chamberlin's model (monopoly solution)

A. COURNOT'S DUOPOLY MODEL

The French Economist Augustin Cournot introduced the first formal oligopoly model (in 1838). He made the
following assumptions to illustrate his model.
1) For simplicity, Cournot assumed that there are only two firms (duopoly)- A & B- selling identical spring
water, and operating with zero costs.
2) They sell their output in a market with a straight-line downward sloping demand curve.

7
 3) Each firm, while trying to maximize profits, assumes that the other duopolist holds its output constant at
existing level. In other words, firms are assumed never to learn from past experience, which makes their
behavior at least naive.

The third assumption is the basic behavioral assumption made in Cournot model. The result is a cycle of moves
and countermoves by the duopolists until each sells one-third of the total industry output and n/(n+1) of market
output (where n = number of firms). This is shown in figure below.

P P
12 Panel (a) 12 Panel (b)

9 Market
Demand (D)
8
6 A dE
D 6
A’
dB 4.5
4 E
B d’A
3 3 B
dB

0 3 6 0 3 4 4.5 6 8 9 12 Q
mrB mrA

In panel (a), D is the market demand curve for spring water. The marginal cost of production is assumed to be zero.
When only firm A is in the market, D=d A and the firm maximizes profits by selling Q=6 at P = $6 (at point A, given
by MRA = MC =0). When firm B enters the market it will face d B (the unsupplied portion of the market). Firm B
maximizes profits by selling Q = 3 at P = $3 (at point B, the midpoint of d B at which MRB = MC = 0).

Firm A now faces d'A given by market demand (D) minus amount sold by B (Q B=3) and maximizes profits by
selling Q = 4.5 at P =$4.5 (see panel (b) of point A'). While firm A’s equilibrium output goes on decreasing that of
B rises. This process continues until each produces one-third of the market (1/3 x 12 = 4 units) and two-third
together (8 units = 2/3 x 12) - (see point E on dE).

In general, for n-firms each will produces 1/ (n+1) and the industry output will be n/ (n+1). So, the larger the number
of firms, the closer is its output and price to the competitive level. For instance, from our illustration:
Under perfect competition, Q = 12 units at P =0 (P = MC = 0)
Under Monopoly Market, Q = ½(12) = 6 units at P = $6 (MR =MC =0)
Under Oligopoly (duopoly) Market, Q = 2/3 (12) = 4 units at P = $4 (lies between competition &
Monopoly)

NOTE: Each firm produces 1/(n+1) of the total output that would be supplied to the market at P=MC=0; industry
supply will be n[1/(n+1)] of output that would be supplied to the market at P = MC = 0.

NB: The assumption of costless production is unrealistic. But it can be relaxed without impairing the validity of
the model. This is illustrated with the help of the reaction-curves approach.

Reaction Curves Approach: is based on the concept of isoprofit curves of the competitors. An isoprofit curve
for firm A is the locus of points defined by different levels of output of A and his rival, firm B, which yield to A the
same level of profit.

8
 QB
QB
A’s Reaction
function

ΠBi = B’s
ΠAi = A’s Isoprofits
ΠB1
Isoprofits
ΠB2 B’s Reaction
ΠA3 function
ΠA2

ΠA1 ΠB3

0 0 QA
QA

Similarly, an isoprofit curve for firm B is the locus of points of different levels of output of the two competitors (A
& B) which yield to B the same level of profits.

Properties of Isoprofit curves:

1) Isoprofit curves are concave to the axis along which we measure the output of the rival firms. For example,
isoprofit of firm A is concave to QB axis & that of B is to QA axis.
2) The further the isoprofit curves of a firm lie from its axis, the lower is the profit it represents and vice versa.
i.e. ∏A1>∏A2>∏A3, and ∏B1>∏B2>∏B3.
3) The highest points of successive isoprofit curves lie to the left of each other for firm A and to the right of
each other for firm B.

NOTE:
1) The profit maximizing output of A (for any given quantity of B) is established at the highest point on the
lowest attainable isoprofit curve of A.
2) If we join the highest points of the isoprofit curves of A and B, we will obtain firm A's & firm B's reaction
curve, respectively.

Definition: Reaction curve of firm A is the locus of points of highest profits that firm A can attain, given the level
of output of rival firm, say B. Similarly, reaction curve of firm B shows how much output firm B must produce in
order to maximize its own profit, given the level of output of its rival, firm A.

EQUILIBRIUM OF THE DUOPOLIST

QB Cournot's equilibrium is determined by the intersection

of the two reaction curves. It is a stable equilibrium,
provided that A's reaction curve is steeper than B's
A’s Reaction reaction curve.
curve NB: At point e each firm maximizes its own profit.
The action & reaction of the duopolists leads toward
point e.
Be e
B2
B1 B’s Reaction
curve
0 A A2 A1 QA

9
Mathematical Derivation of Reaction Curves:

Suppose the market demand facing the Duopolist is: Q = a + bP, where Q = market output, P = price & b = slope,
which is less than zero. Assume also that the two duopolists have different costs:
C1 = f1(Q1), and C2 = f2(Q2).
Thus, 1 = R1 - C1 ------ the first duopolist profit function &
2 = R2 - C2 ----- the 2nd duopolist profit function
Note: R1 = PQ1 and R2 = PQ2 , and Q1 + Q2 = Q

i) The first duopolist maximizes his profit by assuming second firm's output (Q 2) constant, irrespective of his own
decisions; while the second duopolist maximizes his profit by assuming that Q 1 will remain constant.

The first -order (necessary) condition for profit maximization of each duopolist is:

1 =  (R1-C1)= R1 - C1 =0, and

Q1 Q1 Q1 Q1

2 =  (R2-C2)= R2 - C2 =0, and

Q2 Q2 Q2 Q2

NB: R1 = MR1, R2 =MR2, C1 = MC1 & C2 = MC2
Q1 Q2 Q1 Q2

Thus, 1 = MR1 - MC1 = 0  MR1 = MC1 .......... (1)

Q1

And 2 = MR2 - MC2 = 0  MR2 = MC2 .............. (2)

Q2
Solving from expression (1) for Q1, we obtain Q1 as a function of Q2 & hence we obtain 1st firm's
reaction function.
Solving from the 2nd equation for Q2 we obtain Q2 as a function of Q1, which represents 2nd firm's
reaction function.
Finally, solving the reaction functions of the two duopolists simultaneously we obtain the Cournot stable
equilibrium; that is, the values of Q1 & Q2 which satisfies both equations.

ii) The second-order (sufficient) conditions for equilibrium are:

 2 1  2 R1  2C1  2 R1  2C1  2 2  2 R2  2C2
 0    0   ; and  0  
Q12 Q12 Q12 Q12 Q12 Q22 Q22 Q22
That is, MC must cut MR curve from below, or (Slope of MR) < (Slope of MC) for both duopolists.

Example 1: Assume that the market demand equation is given as: Q = 12-P, where Q = total quantity of spring
water sold in the market & P is the market price. Suppose that spring water is supplied with zero costs. Thus, MC
is zero for the two firms, A & B.
a) Find the reaction function of both firms.
b) Compute QA & QB that leads to Cournot stable equilibrium.

Solution: Q = 12-P or P = 12- Q, where QA + QB = Q

Q = 12 - QA - QB
a) Revenue of A: RA = PQA = (12 - QA - QB)QA
RA = 12QA -Q2A - QAQB; and hence A = RA - CA = 12QA - Q2A - QAQB
Revenue of B: RB = PQB = (12 - QA - QB)QB
RB = 12QB - QAQB – Q2B ; and hence B = RB - CB = 12QB - QAQB - Q2B
10
 i) Necessary condition for maximum profits is:
A = 0, and B = 0
QA QB

A = (12QA - Q2A - QAQB) = 0  12 -2QA - QB = 0 .......(1)

QA QA

B =  (12QB - QAQB - Q2B) = 0 12 -QA -2QB = 0 ...... …(2)

QB QB

Solving for QA from expression (1) , we get A's reaction function, i.e.
QA = 12 - QB ------------- (3) A's reaction function
2
Solving for QB from equation (2), we obtain B's reaction function
i.e. QB = 12 - QA ------- (4) B's reaction function
2
b) The Cournot equilibrium can be obtained by substituting duopolist B's reaction function into duopolist A's
reaction function (equation 3 ). Doing this, we get:

QA = 12 - (12 - QA)/2 = 12 - (6 - 0.5QA) = 12 - 6 + 0.5QA

2 2 2
QA = 6 + 0.5QA = 3 + 0.25QA  3 + QA = QA

Multiplying both sides by 4 we get:

4QA = 12 + QA  4QA -QA = 12  3QA = 12; thus, QA = 12/3 = 4

With QA = 4, QB = (12- QA)/2 = (12 – 4)/2 = 8/2 = 4

So that, QA = QB = 4 (Cournot equilibrium).

But P = 12 - Q, where Q = QA + QB = 8 (Industry out put)

Thus, P = 12 - 8 = $4 Each duopolist will sell spring water at P = $4

Example 2 : Assume that the market demand & the costs of the duopolists are as given below:
Market Demand: P = 100 - 0.5Q = 100 - 0.5(Q1 + Q2); and Costs: C1 = 5Q1 , and C2 = 0.5Q22 .

a) Find the reaction function of each firm.

b) Find Q1 and Q2, the satisfy Cournot's equilibrium.

Answer: a) For A: Q1 = 95 - 0.5 Q2, & b) Q1 = 80 , Q2 = 30 and P = 45

For B: Q2 = 50 - 0.25Q1

Example 3: The marginal revenues of the duopolists need not be the same. Actually, if the duopolists are
unequal size the one with the larger output will have the smaller marginal revenue. Proof!

Ri
(That is, proof that  MRi  P  2bQi , where P  a  bQ , b  0 ). Mistaken it is MRi = P+BQi
Qi
NB: For the last two examples, you must show all the necessary steps.

11
B. THE STACKELBERG MODEL:

Heinrich Stackelberg (German economist) made an important extension to the Cournot model. Stackelberg
assumed that one of the duopolists, say duopolist A, knows that duopolist B behaves in the naive Cournot fashion
(i.e., firm A knows B's reaction function) and uses that knowledge in choosing its own output. Duopolist A is then
called the Stackelberg leader, duopolist B is referred to as the Stackelberg follower. All the other assumptions of
the Cournot model hold.

This model shows that duopolist A will have higher profits than under Cournot solution at the expense of duopolist
B (the Stackelberg follower).

The Stackelberg leader substitutes the follower's reaction function in market demand equation and solve for his
output (QA). The resulting expression is the demand function facing duopolist A (the leader). Then, as usual cases
the leader maximizes profit by equating his MR to his MC.

Example: Assume market demand is given by:

Q = QA + QB = 12 - P, & both firms have equal average (and marginal) costs which is zero. Thus, A's
reaction function is: QA = (12 - QB)/2 ................. (1)
B's reaction function is: QB = (12 – QA)/2 ............. (2)

Because duopolist A knows duopolist B's reaction function, we can substitute expression (2) into market demand,
i.e
QA + QB = 12-P  QA + (12 - QA)/2 = 12- P  QA = 12 - 2P ......(3)

Equation (3) is the demand function facing duopolist A when he knows duopolist B's reaction function & behavior.

From expression (3), we get MRA, i.e.

RA = PQA, where P = 6- 0.5QA
= (6-0.5QA)QA = 6QA - 0.5Q2A
MRA = RA = 6 - QA
QA
At equilibrium MRA = MCA  6 - QA = 0  QA = 6. But P = 6 - 0.5QA  P = 6 - 0.5(6)  P = 3

Thus, since MC = 0, duopolist A maximizes its total revenue and total profits by selling six units at
P = $3. Duopolist B would then sell three units; because.
QA + QB = 12 - P
6 + QB = 12 - 3  6 + QB = 9  QB = 3
With Q = QA + QB = 9, P = $3, duopolist A earns $18 (PQ A = 3 x 6 = 18) and duopolist B earns $3 (PQ B = 3 x 3 =
$9). Note that under the Stackelberg Solution, whether duopolist A or B is the Stackelberg leader, the combined
total revenue and profits of both firms would be the same ($27 = $18 + $9, in our example).

Alternative Method: The Stackelberg leader substitute the Stackelberg follower's reaction function into his own
profit function to determine his own output & market price.

C. THE KINKED-DEMAND MODEL:

Here interdependence is recognized unlike the Cournot model. This model, introduced by Paul Sweezy in 1939,
attempts to explain the price rigidity that is often observed in some oligopolistic markets. The two basic
assumptions of this model that leads to a kinked-demand curve are:

1) If an oligopolistic firm raised price, it would lose most of its customers because the other firms in the
industry would not match the price increase. Thus, the oligopolists face a demand curve that is highly
elastic (i.e. dA).

12
 2) Oligopolists could not increase its share of the market by lowering its price, since its competitors would
immediately match the price reduction as a result the firm faces a demand curve that is less elastic for
price reductions (like segment AD).
P/C Because of the above two assumptions, the demand
curve faced by oligopolists has a kink at the established
price (PA or Point A). As a result the marginal revenue
is discontinuous at the level of output corresponding to
d the kink (at QA the discontinuity segment is EF).
MC’ MC2 Oligopolist's marginal revenue curve is MR or dEFG.
PA A
Segment dE of MR curve corresponds to segment AD
MC’’
of the demand curve (see figure).
E
The discontinuity (between E & F) of the MR curve
MC1
implies that there is a range with in which costs may
F change without affecting the equilibrium price (P A) &
G quantity (QA).
Q
0 QA D
NOTE: Oligopolists' demand curve is given by
MR segment dAD and MR is dEFG.
In general, the behavioral pattern implied by the "kink" seems quite realistic in the highly competitive business
world which is dominated by strongly competing oligopolists. It can explain the stickiness of prices in a situation of
changing costs & high rivalry. However, there are some major limitations in this model:
 The model implies that price rigidity coincides with quantity rigidity (but not the case in reality)
 It doesn't explain the level of price at which the kink will occur.
 The assumption that firms don't match price increase is also questioned.

D. BERTRAND DUOPOLY MODEL:

Bertrand’s duopoly model (which was developed in 1883) differs from Cournot’s in that it assumes that each firm
expects that the rival will keep its price, irrespective of its own decision about pricing. Each firm is faced by the
same market demand, and aims at the maximization of its own profit on the assumption that price of the competitor
will remain constant. In this case the firm chooses price in stead of quantities (unlike case of Cournot). Like the
Cournot model, the Bertrand model duopoly model applies to firms that produce homogenous product.

Because products are homogenous, consumers purchase only from the lowest price seller. Thus, if the two firms
charge different prices, the higher priced firm will sell nothing. If both firms charge the same price, consumers will
be indifferent as to which firm they buy from. And it is assumed that each will supply half the market. In general,
this model leads to price competition over time. Each firm tries to cut price to capture all the market share. Thus, all
will seek to undercut their rivals and a price cutting war will result. Firms will stop cutting their prices only when
competitive price level is reached. In other words, Bertrand’s equilibrium is at a perfectly competitive price such
that no further price cutting occurs.

Consider the following example where the market demand for a good is: P = 80 – Q, where Q = Q 1+Q2; and both
firms have a constant MC of 10. Thus, if price is below this MC, there will be an incentive to reduce price with the
intension of capturing more market share. This price cut will continue until it equals MC, which is, of course, a
competitive price. Thus, for Bertrand equilibrium occurs when P = MC = 10, and over all equilibrium output equals
70 (Q=80-P), where each firm supplies half of the market; i.e. Q 1=Q2= Q/2=70/2 = 35 units.

Criticism:
Just like Cournot model, Bertrand’s duopoly model is criticized on the following points:
1. Assumption of naïve behavior; i.e. rivals don’t learn from past experience.
2. It is closed model in the sense that entry is blocked.
3. Other variables of competition are not included in the model; it considers only price competition.

13
Note that price competition is more natural (than quantity choice) when products have some degree of
differentiation. To illustrate this, consider duopolists with costs and demand schedule for each given as follows:
Costs: C1=10+3Q1 and C2=10+3Q2; Demand: Q1=12-2P1+P2 and Q2=12-2P2+P1. What will be the equilibrium of the
two firms setting their prices at the same time?
Solution:
 1
First firm’s profit: π1 = R1-C1=18P1-2P21+P1P2-3P2-46; so  0  18  4P1  P2 at maximum profit.
P1
Solving P1 as a function of P2, we obtain firm-1’s reaction function as: P1 = 4.5+0.25P2.
 1
Firm-2’s profit: π2 = R2-C2=18P2-2P22+P1P2-3P1-46; so  0  18  4P1  P2 at maximum profit. Solving
P1
P2 as a function of P1, we obtain firm-2’s reaction function as: P2 = 4.5+0.25P1.
Finally, solving the two reaction functions simultaneously, we arrive at Bertrand’s stable equilibrium
solution: P1=P2=$6, Q1=Q2=6 units and π1= π2=$8 (in thousands).

II. CULLUSIVE OLIGOPOLY

One way of avoiding uncertainty (eg. price war) arising from oligopolistic interdependence is to enter into collusive
agreements. There are two types of collusion: Cartels and Price Leadership Models. Collusion can be explicit or
tacit (implicit).

2.1 Cartel: is a combination of firms whose objective is to limit the scope of competitive forces within a
market. There are two forms of Cartels:
A) Centralized Cartel (cartels aiming at joint profit maximization), and
B) Market-sharing cartel.

A) Centralized Cartel: operates as a monopolist. It is a formal organization of producers of a commodity

whose purpose is to coordinate the policies of the member firms so as to increase joint or industry profits. The
centralized cartel appoints a central agency to which they delegate the authority to decide:
 the monopoly price for the commodity & monopoly output.
 the allocation of monopoly output among the member firms
 the share (distribution) of monopoly profits among member firms.

Note that the agency:

 will have access to the cost figures of individual firms,
 knows the market demand curve from which he estimates marginal revenue.

Thus, total marginal cost is obtained by the horizontal summation of each firm's in the cartel. We consider a
homogeneous commodity or pure oligopoly. Industry (monopoly) price & output is determined by equating market
MR and industry MC (= SMC). Consider two firms form a cartel.

Given the market demand (DD) & each firm's MC in figure above the monopoly solution, which maximizes joint
profits, is determined by the intersection of total MC & MR (point e). The total output is Q m and it will be sold at
P . The central agency allocates the production (Q m) among firm A & firm B by equating the common MR to the
individual marginal costs (i.e. MR = MC1 = MC2; see point e1 & e2). Thus, firm A will produce Q1 and firm B will
produce Q2. Similarly, the distribution of profits is decided by a central agency of the cartel

14
 P/C P/C MC=MC1+MC2
p/c
Firm-A Firm-B D Market
AC1 MC2 AC2
MC1

P c P
d
f
g P
a b
MC1=MR MC2=MR e2 MC=MR e
e1

0 Q1 QA 0 Q2 QB 0 Qm=Q1+Q2 D Q
MR

mainly based on their cost structure. The shaded areas (or area abc P for firm A & area of dgf P for firm B) of
figures shown above represents profits of each firm. A total industry profit is the sum of the profits of the two firms.

Mathematical Presentation of this Cartel:

Suppose the market demand is given as:

P = f(Q) = f(Q1 + Q2). Thus, Q1 = PQ1 & R2 = PQ2
 And the cost curves of the two firms are:
C1 = f1(Q1) and C2 = f2(Q2).
 The profit function of each firm will be:
1 = R1 - C1, and 2 = R2 – C2
 The joint or industry profit is:  = 1 + 2, i.e.
 = (R1 + R2) - (C1 + C2) = R - C1 - C2
 The first- order condition for maximum profit is:
 
 0, and  0 . That is,
Q1 Q2
  R C1
 ( R  C1  C2 )  0    0  ( MR  MR1 )  MC1      (1)
Q1 Q1 Q1 Q1
  R C2
And  ( R  C1  C2 )  0    0;(MR  MR2 )  MC2    (2)
Q2 Q2 Q2 Q2
Note that MR = MR1 = MR2 , since all units are sold at the same price.

 From expression (1) & (2), the necessary condition for equilibrium is:

(MR1 = MR2 = MR) = MC1 = MC2 or MR = MC1 = MC2

 Second-order condition is:

2R  2C2 or (slope of MR)  (Slope of MC)
Q22 Q22

Example: assume that the four identical firms in a purely oligopolistic industry form a centralized cartel. Suppose
that the total market demand function facing the cartel is: Q = 20 -2P, and P is given in birr. Each firm's short-run
marginal cost function is given by birr 0.25q , & input prices are constant.
a) Find the best level of output and price for this centralized cartel.
b) How much should each firm produce if the cartel wants to minimize production costs?
c) How much profit will the cartel make if the average total cost of each firm at the best level of
output is birr 4?

15
{Answer: (a) Q = 160/17, & P = 90/17 (b) Q1 = Q2 = Q3 = Q4 = 40/17 and (c) Unit  = P - AC = 90/17 -4 =
38/17, & 1 = ....= 4 = 38/17 x 40/17; since total π = (unit profit)*Output  1 = 2 = ..... = 4 =
1520/289. NB: (Q1 = Q2 = Q3 = Q4) = q & Q1 + Q2 + Q3 + Q4 = Q = 4q}

B) Market - Sharing Cartel

In a market - sharing cartel the member firms agree only on how to share the market, but keep a considerable degree
of freedom concerning the style of output, their selling activities & other decisions.

There are two methods of sharing the market:

1) Non-Price competition, and
2) Determination of Quotas

1) Non-Price Competition: In this case the member firms agree on common price, at which each firm can sell
any quantity demanded. The price is set by bargaining, with the low - cost firms pressing for a lower price & the
high -cost firms for a high price. But the firms agree not to sell at a price below the cartel price.

This form of Cartel is "loose" in the sense that it is more unstable than the complete cartel aiming at joint profit
maximization. This is because with cost differences, the low - cost firms will have a strong incentive to break away
from the Cartel openly & charge a lower price, or to cheat the other firms (members) by secret price concessions to
the buyers. Such cheating will soon be discovered by the other members, who will gradually lose their customers.
Therefore, others may split from the cartel and price war and instability may develop.
P/C
MCB MC=MCA+MCB
MCA PBe
Pm
Pm Pm
PAe PA
eB D
e
dA dB
eA
MR
MRB

0 qAe MRA qA 0 qBe qB 0 Qm=qAe+qBe Q

A: Low-cost Firm B: High-cost Firm Industry Equilibrium

NB: Pm is agreed price; P Ae & PBe are equilibrium price of firm A & B if they separately want to maximize . Low-
cost Firm wants to cheat by lowers of price.

2. Market-Sharing by Agreement on Quotas: This method represents the agreement on the quantity that
each member may sell at the agreed price (prices). If all firms have identical costs, the monopoly solution
will emerge, with the market being shared equally among member firms; otherwise shares of the market
will differ. (See figure below for a monopoly solution).

16
 SMC
NB: Another popular method of
B

Pm
sharing the market is the
=d
dA definition of geographical
D region in which each firm is
allowed to sell

0 qA=qB Qm
mrA=mrB MR

2.2 Price-Leadership:
In this form of coordinated behaviour of oligopolists one firm-usually the dominant or the largest firm in the
industry -initiates price changes, and allows all the other (small) firms to sell all they want at that price.

There are two most common types of price-leadership:

a) Low -cost price -leadership &
b) Price -leadership by large (dominant) firm.

a) The low-cost price -leadership

In this model the firms may come to an agreement to share the markets equally or may agree to have unequal
shares. The important condition here is that the firms have unequal costs. s
are
Sh
-
ket
ar
M
ual
eq
Un
2.
Fg
P/C P/C
Fig. 1: Equal Market -Shares MCB
MCA
MCB ACB
MCA PB
PA
PB ACA
PA
eB
eB dA
d D eA
A =d
B =M
eA R dB

Q
0 qeB qeA=qB Q 0 qeB qB qeA
MRA=MRB MRA
MRB

17
* In figure 1, firm A has the lowest cost and charges a lower price P A & produces qA to maximize profit. Firm
B with the highest cost would like to charge P B & produce qeB (point eB); but it prefers to follow the leader
(PA) in order to avoid a price war. If B charges P B his sales become zero because customers switch to a low
price firm, firm A. Thus, Firm B must change its price to P A and sale qB which equals qA since there is
equal market sharing.

* In fig. 2, the leader (A) maximizes his profit by setting his price at P A & selling qA amount of output. The
follower (B) must supply qB which is sufficient to maintain the price set by the leader, although firm B
maximizes his profit at point eB, selling qeB at price PB.

b) The Dominant Firm Price Leader

In the price leadership model by the dominant firm, the dominant firm sets the price for the commodity that
maximizes its profits, allows all the other smaller firms in the industry to sell all they want at that price, and then
comes in to fill the market. Thus, the small firms in the industry behave as perfect competitors or Price-takers, and
the dominant firm acts as the residual supplier of the commodity. (See figure below).
P/C
MCd NB:
A d is the demand curve of the dominant
∑MCS=SS of small firms firm (i.e. d = HKFG).
 MRd is the marginal revenue curve of the
H B dominant firm
K L D (ABCFG) is the market demand curve.
P e
C

D Since the small firms can sell all they want at

the price set by the dominant firm, they
d F behave as perfect competitors and produce
P3
where P = MCS. The horizontal distance
between market demand (D) and the total
0 G Q supply of smaller firms (MCS) at each price
Qd QS Qm Q3
gives the product (residual) of the leader. For
MRd
instance, at price P the demand for the
product of the leader will be given by segme-
nt LC, which is equal to Qd. Because at this price market demand is Q m of which QS is supplied by smaller firms. So
the difference between Qm and Qs equals leader’s output, LC=Qd.

As price falls down the supply of the smaller firms decrease while the leader’s output increases. For example, at
price P3 total market demand is Q3 which is, of course supplied by the leader because at that high price the small
firms don’t supply any quantity. Then below this price (P 3), market demand will coincide with the leader’s demand
curve.

Now having derived the leader’s demand curve (HKFG=d), and given the MC curve (MC d), the dominant firm will
be at equilibrium when his MR equals his MC. This is given by point e, where OQ d output is produced and price of
P is set. Thus, at this price, small firms supply 0Qs (point L on their supply curve) and total market demand will be
OQm, which is equal to OQd+OQS.
NB:
 The dominant firm is maximizing his profit by equating his MR to MC, and the smaller firms are just price-
takers and maximize profit by equating their MC to market price.
 In order to maximize profit the leader must make sure that the small firms:
1. will follow his price, as well as
2. will produce the right output (OQs).

18
Mathematical Presentation of the dominant firm model: The case of pure duopolists!
Suppose the market demand is given as: P = f(Q), or Q = f(P), where Q=Q d+QS.
And the cost structure of the two groups are: Dominant firm: C d=f1(Qd), and many small firms: C s=f2(Qs),
from which their supply is derived; QS=fs(P)
Then the demand function of the dominant firm is: Qd=Q-Qs; Qd=f(P) – fs(P). Thus, Rd=PQd.
Calculate MR of dominant firm from its demand schedule (MR d=dRd/dQd), and MC from its cost structure:
MCd= dCd/dQd.
Finally, leader equates MCd to MCd to determine the equilibrium market price & its optimal output. The
followers adopt the price set by the leader and then determine its optimal output from their supply schedule.
NB: The low-cost price leader set price by equating his MC with his MR just like any other rational firm. The
follower (the high-cost) firm takes the price set by the leader & substitute it in to his market-share to determine
quantity produced.
Example 1: Price-leadership by low -cost firm. Assume the duopolists share the market equally for a
homogeneous product. If short -run MC (SMC) for firm 1 and firm-2 is given respectively as = SMC 1 = 2Q1 and
SMC2 = 6 + 2Q2, what price & quantity of the commodity will each duopolist produce? (Q = 12-P is given to be
market demand). (Answer: Q1 = Q2 = 2 units, & P = 8).

Example 2: Price -leadership by Dominant firm: Assume that in a purely oligopolistic industry, there is one
dominant firm and ten small identical firms. The market demand for the commodity is: Q = 210-20P. Suppose
MCd = 1.5 + Qd/2, while MCs = 1 + Qs/4 (each small firms' MC curve).

a) Find the price the dominant firm will set.

b) How much will the small firms supply together?
c) How much will the dominant firm supply?

(Ans. a) P = 49/12 , Qm = 385/3 b) Qs = 370/3 & C) Qd = 5

19
GAME THEORY & OLIGOPOLISTIC BEHAVIOR

A. Basic Concepts of Game Theory

*Game theory is concerned with the choice of the optimum strategy in conflict situations. It can help an oligopolist
choose the course of action of best price, best level of advertising, and optimal degree of product differentiation that
maximizes its benefits or profit after considering all possible reactions of its competitors.

Game Theory
 Game theory is concerned with the general analysis of strategic interaction. Though it has a wide variety of
application, here we will provide a brief introduction of game theory with a primary focus on its use in
explaining pricing & entry behavior in oligopolistic markets.
 Any situation in which individuals must make strategic choices & in which the outcome will depend on
what each person chooses to do can be viewed as a game. All games have three basic elements: Players,
Strategies, and Payoffs. Game may be cooperative, in which players make binding agreements or non
cooperative, where such agreements are not possible. Here we will concentrate on the latter.
 A Player implies each decision-maker in a game. Each player is assumed to select the course of action that
yields the most favorable outcome.
 Strategy represents each course of action open to a player in a game. Each strategy is assumed to be a well-
defined, specific course of action. Each player is uncertain about what the other will do. Strategy can be
pure or "mixed".
 Payoffs incorporate all aspects associated with outcomes of a game; that is both explicit monetary payoffs
& implicit outcomes. The outcome of each combination of strategies adopted by the two players or firms is
called the payoff. The payoff of all strategies is called Payoff Matrix

B. Nash Equilibrium
Are there strategic choices that, once made, provide no incentives for the players to alter their behavior further?
According to John Nash, a pair of strategies, say (a, b) for two respective firms (A & B), is defined to be an
equilibrium if "a" represents player A's best strategy when B plays "b" & "b" represents B's best strategy when A
Plays "a". Even if one of the players reveals the optimal strategy she/he will use, the other players cannot benefit
from knowing this. Thus, strategies "a" & "b" are called dominant strategies.

C. Two-Person zero-sum game:

Certainty model: A game where the gains of one firm equal the losses of the other (so that the total gains plus the
total losses sums to zero) is called zero-sum game. E.g. A duopoly model in which each duopolist attempts to
maximize his/her market share. Given this goal, whatever a firm gains the other firm losses.

Basic Assumptions
1. Firms have well defined goal
2. Each firm knows the strategies open to it and its rival
3. Each firm knows with a certainty the payoffs of all combinations of the strategies being considered.
4. The actions chosen by the duopolists do not affect the total size of the market.
5. Firms are 'rational', i.e. each firm chooses its strategy expecting the worst from its rival.

20
Now in order to find the equilibrium solution we need information on the payoff matrix of the two firms. Assume
that firm A has two strategies open to it & firm B has three strategies.

Combined payoff matrix

Firm A's Strategies

Firm B's Strategies
B1 B2 B3
0.15 0.25 0.50
A1
A2 0.55 0.35 0.40

NB: The sum of payoffs of firm A & B is equal to 1. Thus, if player B responds with B 1 to strategy A1 (by firm
A), A will gain 0.15 (first entry in the first row), and firm B will gain 0.85 (1-.15) of the market. However, if B
responds with strategy B2 to strategy A1, A will gain 0.25 of the market and B will gain the remaining (0.75). The
same interpretations are made for others.

Before we deal with equilibrium solution, we have to understand the basic concepts of maximin and minimax
strategies.
Maxmin: In game theory, firm A knows that firm B will always respond to A's gains because that is the strategy
that minimizes B's loss. For example if firm A adopts strategy A 1, the worst (minimum) outcome that it may expect
is the share of 0.15 (which corresponds to best strategy of firm B, B 1). If firm A adopts A2, the worst outcome will
be a share of 0.35 (if firm B adopts the best action for him, B 2). Thus, among all these minima (or the worst
outcomes) firm A chooses the maximum, i.e. the 'best of the worst'. This is called maximin strategy-a strategy that
maximizes its minimum gain. In our examples, the maximin of firm A is A 2, which yields a share of 0.35.

Minimax:
Firm B considers the columns of its payoff table, since it shows the payoffs of each of the strategies open to firm B.
For each strategy (or column) firm B looks for the worst outcome (on the assumption that the rival will choose the
best) and should choose the best among these worst outcomes.

Firm B behaves exactly the same way as firm A. If firm B adopts strategy B1, the worst outcome will be A's best
outcome, i.e. 0.55. If firm B adopts B 2, the worst outcome will be 0.35, for strategy B 3 the worst outcome is 0.50.
Among these maxima of each column (or strategy) firm B will choose the strategy with minimum value. Therefore,
the strategy of firm B is minimax strategy-B chooses a strategy that minimizes its loss or minimizes A's gains. In
our example, the minimax strategy of firm B is B 2, which yields a share of 65 % (=1-0.35). Now it is direct forward
to talk about equilibrium solution. The equilibrium solution is strategy A 2 for firm A and B2 for firm B. This
solution yields shares of 0.35 for firm A and 0.65 for firm B. The preferred strategies A 2 and B2 are called
dominant strategy - a strategy that provides a greater outcome to any one firm no matter what the rival does. These
strategies obey the Nash Criterion for Equilibrium.

In Summary:
Firm B's Strategies Row Minima
Firm A's Strategy

B1 B2 B3
A1 0.15 0.25 0.50 0.15
A2 0.55 0.35 0.40 0.35
Column 0.35
maxima 0.55 0.35 0.50 0.35

Definition:
* Maximin Strategy is the strategy of maximizing the minimum payoff in game theory.
* Minimax Strategy is the strategy of minimizing the maximum payoff of the rival player in game theory.

21
NB: In a game where the maximum of the row minima (maximin) coincides with the minimum of the column
maxima (or minimax), each firm is said to choose a pure strategy. The game is then said to be a strictly determined
game and the solution is called the saddle point. If the game is not strictly determined, a firm should choose a
mixed strategy.

*Pure strategy is the single best strategy in game theory. Mixed Strategy is the best strategy for each firm in a
game, which is, not strictly determined (a game in which maximin equals minimax). The solution or outcome of a
strictly determined game is called Saddle Point.

D. Non-Zero - Sum Game: is game in which the sum of the payoffs for each combination of strategies not equal to
zero. This model can be illustrated with a duopolistic market in which the firms aim at the maximization of their
profit.

For simplicity, suppose that there are two strategies open to each competitor-low price & high price strategies. The
payoff table of the two firms is shown as follows:

Combined Payoff Matrix

Firm B's Strategies
Firm A's Strategies

low Low Price High Price

price (1000, 1000) (3000, -1000)

high
price (-1000, 3000) (2000, 2000)

Note that in the above pay-off matrix, the first entry of each combination of strategies refers to firm A’s pay-off and
the 2nd refers to firm B. Here each firm expects the worst from the rival. Thus, both firm A's and B's choice is a
maximin strategy. If firm A charges a low price the minimum gain is birr 1000; if it sets a high price its minimum
return is a loss of birr 1000. Thus, firm A chooses the maximum among these two minima-i.e. profit of 1000 with
preferred strategy of low price.

Similarly, if firm B charges a low price the worst it can expect is a profit of 1000 birr; if it charges a high price the
minimum possible return is a loss of 1000 birr. Thus, firm B should go for strategy of low price since it yields the
"best of the worst". This is a strictly determined non-zero-sum game with a saddle point occurring when each firm
charges a low-price (which is a dominant strategy) and earns a profit of birr 1000. NB. Collusion, though illegal,
enable both firms to charge a high price & increase their profits to birr 2000.

E. Prisoner's Dilemma (Negative payoffs)

Suppose two suspects are arrested after committing a big bank robbery. Each suspect has two strategies open to
him: to confess or not to confess. Suspects are interrogated separately & no communication is allowed between
them. Each suspect is faced with a dilemma to confess & go free if the other doesn't confess, or get a ten year
prison sentence; or not to confess & go free it the other does not confess or get a heavy twenty-year sentence if the
other suspect confess. The payoff of the two prisoners is given as follows:
Prisoner A's Strategies

Suspect B
Confess Not confess
Confess 10, 10 0, 20
not confess 20, 0 0,0

22
Clearly, suspect A gets 10 years or goes free by confessing, as opposed to 20 years or zero year by not confessing.
Thus, suspect A confesses. Similarly, suspect B either gets a 10-year sentence or goes free by confessing as
opposed to 20 years or zero years by not confessing. Thus, the dominant strategy is to confess, so that both get a 10
year sentence.

CHAPTER-III

III. PRICING & EMPLOYMENT OF INPUTS

Income disparities exist among individuals in any economy. The main reason for this income disparity is:
-differences in the quantity and quality of resources owned,
-and differences in the price received for the services each type of resources.

In a free-enterprise economy, therefore, the distribution of income among individuals in the economy is determined
to a considerable extent by the configuration of input prices. Thus, in this chapter we concentrate on the
determinants of the price and employment of inputs. This, of course, requires the interaction of demand for and
supply of inputs. Thus, we first examine the demand for a competitive firm then pass to imperfect market
situations. We follow the same step for input supply.

3.1 Input Demand Curve of a Competitive Firm

a) The Firm's Demand curve-One variable input case:

Simplifying assumptions:
i. Consider only one variable input, labor, non labor inputs are fixed.
ii. There is perfect competition in both commodities and factor/input markets. That means, price of a
commodity, say X, as well as the price of labor services (wage rate) is given. In other words, the demand for a
commodity X and the supply of labor (workers) is perfectly elastic. That is, firms can sell any amount of
commodity demanded at a given market price, P ; and firms can hire any amount of labor at the going market wage
rate, W .
P
Perfectly elastic W Perfectly Elastic
output Demand Labor Supply

P d W SL

0 Q W L

iii) Firms have perfect information concerning the productivity of all inputs including workers.
iv) The goal of the firm is profit-maximization.
v) Technology is given, i.e. production changes only if labor input changes.

23
Profit Maximization and Derivation of Labor Demand
It is clear that a firm's hiring of inputs is directly related to its desire to maximize profits. Hence any profit-
maximizing firm should hire (employ) additional units of each factor of production up to the point at which extra
revenue yielded by hiring one more unit is equal to the extra cost of hiring that unit.

Some basic concepts:

 The marginal revenue product (MRP) from hiring extra unit of any input is the extra revenue yielded by
selling what that extra input produces. It can be obtained by multiplying the input's marginal productivity
with the marginal revenue obtainable from the firm's output in the market for goods. i.e.
MRPL = dR/dL = MPL*MR. Proof! Mistaken It has to be multiplied by Q
d ( PQ) Q P Q Q  P  Q P
MRPL  P Q  P  ; where  MPL , P   MR .
dL L Q L L  Q  L Q
Thus, MRPL = MPL (MR), which is the general formula for MRPL.

However, we know that MR is equal to price in a perfectly competitive market. Thus, the marginal revenue product
of labor in this competitive market becomes: MRPL = MPL(P), which also indicates the value of marginal product of
labor (VMPL). That is, for a competitive firm: MRPL = VMPL = MPL*P.

Definition: The value of marginal product of labor (VMP L) measures the extra revenue a competitive firm receives
by selling the additional output generated when employment of an input increased by one unit.

 Marginal Cost of input, say labor (MC L): is the extra cost or expense of a firm incurred from hiring
additional unit of that input. It is a change in cost of input per unit change of the input used. Symbolically:
MCL = dCL/dL.

In a competitive factor markets, marginal expense (extra cost) is simply equal to the inputs price. That is,
MCL = dCL/dL = W . Proof!
dCL d (WL) dL
MCL  , but CL  WL; thus MCL  W  W , W can be factored out since it is fixed ( a
dL dL dL
constant doesn’t affect the derivative). Thus, in a competitive factor markets, marginal expense (or extra cost) is
simply equal to the input’s price; MCL = W .

The first order condition for profit maximization therefore, becomes: MRP L = VMPL = W ., since P = MR in
perfect competition.

Profit Maximization:
Any profit maximizing firm should hire additional units of each factor of production up to the p oint at which extra
revenue yielded by hiring one more unit of an input is equal to the extra cost of employing that unit. That is,
equilibrium occurs only when:
 MRPL = MCL, in general, and
 VMPL = W for competitive firm in particular.
Why other points are not equilibrium? Because:
1. When MRPL > MCL, one additional employment of worker adds more to revenue than costs, and hence
firms are motivated to hire more labor & produce more.
2. If MRPL > MCL, an extra unit of hiring labor adds more to costs than revenue and hence the firm should
cut employment.
3. It follows from the above two conditions that a stable (or equilibrium) condition occurs only when
MRPL equals MCL in general.
Since MR = P in a perfectly competitive market, the equilibrium condition in this market will be:

24
 (MRPL = MPL*MR= MCL) = (MPL*P = VMPL= W ).
The proof: Any firm's profit () can be expressed as the difference between revenue (R) and costs (C), each of
which can be regarded as functions of inputs used, labor. That is,
C = f(L) = W L + FC, and R = PQ, where Q = f(L); thus R = f(L). We know that π = R – C.
The first order condition for maximum  are:
d d dR dC d ( PQ ) d (WL  FC ) dQ dL
 (R  C)  0    0  0 P W 0
dL dL dL dL dL dL dL dL , d(FC)=0.
P ( MPL )  W  0  VMPL  W

Graphical illustration: Note that the relevant portion of production function for a rational firm is the portion of
following but positive marginal productivity of inputs - the law of variable proportion. Multiplying this MP of an
input by a given commodity price we obtain VMP, which measures the extra revenue a competitive firm receives by
selling the additional output generated when employment of an input increased by one unit. By equating this VMP L
to inputs price (W), we determine profit maximizing level of employment.

MPL/VM W/VMPL
Equilibrium
llPPL
VMPL
e
MPL W SL

VMPL
0 L* L
0 L

At any point to the left of L*, VMP L  W. Thus, profit is maximized by hiring more labor. But to the right of L*,
VMPL < W, implying that profit can be raised only by cutting the level of employment. Therefore, it follows that,
maximum profit is maintained only at point e (L* units of labor) where VMP L = W .

Derivation of Individual Labor Demand Curve

The firm's demand for any input depends on:
 how productive a particular input is in making goods, & thereby yielding revenue to the firm, and
 how hiring of inputs affects costs.

We expect the demand for labor to be downward sloping (slope = dW/dL)  0; i.e. a decrease in wage (W) will
cause more labor to be hired to bring about the equality of W & VMP L (=MPL* P ); this is because a fall in W must
be met by a fall in MPL, which is possible only by increasing L (NB: product price is fixed).

25
 Equilibrium at different wages Wage Derived Labor Demand Curve

W/VMPL

W1 e1 SL1 W1 e1’

e2 e2’
W2 SL2 W2

W3 e3
SL3 W3 e3’

dL = VMPL
VMPL

L1 L3 L
0 L1 L2 L3 L L2
If the market wage rate is W 1 the firm is in equilibrium by hiring L 1 (point e1 where MVPL = W1). If market wage
rate falls to W2, the firm will maximize its profit by increasing its employment to L 2 at point e2 (where MVPL = W2).
Thus, if the firm demands the optimal amount of input, labor, at each wage rate, its demand curve for the input
(labor) must be the VMPL curve. Therefore, the marginal-value product curve is the firm's demand curve for a
given input (labor) when all other inputs are fixed.

Example: Suppose that the number of quintals of coffee harvested in a particular region during one season is given
by: Q = 100√L where L is the number of workers hired to plant and harvest coffee. Assuming coffee sell for $ 60
per quintal, and workers' seasonal wages are $500, what will be the level of employment determined by the owner
to maximize profit?

1
dTR d (6000 L ) 
Solution: TR = PQ = 60 (100√L)=6000√L, and MRPL =   3000 L 2 .
dQ dL
Optimal labor is determined by equating MRPL to wage rate, W: MRPL = W; 3000L-1/2 = 500.
L-1/2=1/6, or L1/2 = 6; or L* = 36. With 36 workers, the MRPL is $500, which is precisely what the owners must pay
in wages. The 36 workers produce a total of 600 quintal of coffee during the season.

b) Firm's Demand Curve: The case of Several Variable Inputs

Suppose now that the firm uses a number of inputs that can be varied in quantity and labor is only one of them.
Under these circumstances, the firm's demand curve for labor is no longer the VMP curve, since the change in price
of labor will result in a change in the quantities of the other variable inputs used. For simplicity consider only two
variable inputs, labor (L) and capital (K). Now if W falls, there will not only be a change in labor but also a change
in capital as a new cost minimizing combination of inputs is chosen. When capital changes the entire MP L function
changes, because labor now has a different amount of capital to work with.

Note that when wage rate falls, we can decompose the total effect on quantity of labor hired into two components:
The substitution effect and output effect. These effects can be analyzed by the concept of isoquants.

26
 Capital
b) Product
Effect of fall in Wage on
A1 equilibrium! P/C market

MC
A MC’

C
K3 e3
K1 e1
e1 e3 d
K2
e1’
e2

Q3
P

Q1 Q2 e’1

B D B’ B1
0 K1 L’1 L2 L3 Labor 0 Q1 Q3 Q

Initially the firm produces the profit-maximizing output, Q 1, with combination of K1 & L1 at point e1. When the
price of labor falls, the isocost line becomes flatter (since slope W/r falls) i.e. the new isocost line is AB'. Thus, the
firm using the same expenditure can now produce a higher output, Q 2, using K2 & L2 at e2.Movement from e1 to
point e2 shows total effect of fall in wage, which can be split is to two:
 One of these, the substitution effect, would cause more of labor to be purchased if output were held constant
at Q1. This is shown as a movement from point e 1 to e'1 on the same isoquant curve. To produce an
unchanged output, the firm uses more labor and less capital when the relative cost of labor falls (i.e. it
substitutes L for K).
 The other, the output effect, is shown by the movement from e’1 to e2. The hiring of labor will also increase
(to L2) due to output effect of reduction in wage rate (while the total expenditure remains unchanged).

NB: Q2 or e2 is the equilibrium of the firm only if the firm were to spend the same amount of money as initially.
But firms produce as much as the available demand allows, by increasing its expenditure further.

A change in wage, because it changes relative factor costs, will shift all the firm's cost curves. In figure (b) the
marginal cost curve for the firm has shifted downward to MC' due to fall in wage rate. Consequently, the firm's
profit maximizing output rises from Q2 to Q3. As the firm produces more output, it moves to a higher isocost line
A1B1. That is, the firm's expenditure rises by an area e 1Q1Q2e3 (see fig. b) which is equal to AA 1 (of fig. a). This
new equilibrium (movement from e2 to e3) is sometimes called the profit- maximizing effect.

In summary:
The substitution effect of a fall in W tends to raise the demand for labor but reduces that of capital. Thus, MP L
will decrease (since more L is used with smaller K) i.e. MP L shifts inward. On the other hand, the output effect
(and profit maximizing effect) of a fall in W results in use of more L and K & hence MP L will rise. This shifts
MPL to the right which outweighs the inward shift due to substitution effect. Thus, the net effect is an outward
shift in MPL and hence the VMPL (= P*MPL).

Derivation of Labor Demand: Several Variable Input Case!

The firm employs more workers at a lower wage rate because it uses more L per unit of output (the substitution
effect) and because it is profitable to increase output when production costs falls (the output effect).

27
 W/VMPL In this figure, let the initial price of L be W1. The firm is in
equilibrium, using OL1 quantity of L at point A. If the price of L
VMP2 falls to W2, it changes the quantity of the other factors (K), which
shift the VMPL to the right. It becomes VMP 2 instead of VMP1.
W1 A The firm then equates W2 and VMP2, to arrive at the amount OL2 of
B factor L. If W falls farther to W3 VMP curve shift again to VMP3
W2
and the firm reaches equilibrium at point C, where W 3 = VMP3,
W3 C hiring more labor, L3, at lower wage rate, W 3. Joining these points
dL
of equilibrium on the shifting VMPL (like points A, B, C), we
VMP1 VMP3 obtain demand curve for labor, d L, when the quantities of others are
0 also variable.
L1 L2 L3 L

Factors that affect Firm's Demand for a variable input:

1. Price of the input. For instance, there is inverse relation between W and L.
2. Marginal product of the input, which is derived from the production function. Positive relation!
3. Price of a commodity produced by the input also affects it directly.
4. The amount of other factors which are combined with labor affects d L.
5. The prices of other factors. Positively if other inputs are substitute, & inversely if it is complement.
6. Technological change/progress. For example, labor intensive technology rises labor demand.

c) Market Demand Curve

The price of the product remained unchanged when we derive the competitive firm's input demand curve (in the
previous section). That is, when W fell, the firm expanded output and sold the larger output at unchanged price.
P
S This is appropriate when we are dealing with only one firm. Hence, when all
S’ firms simultaneously increase output they can sell more (total industry
P* e output) only at a lower price. For instance, at the initial wage rate firms
P’ e’ supply Q* amount of output B sell it at P*. If wage rate falls, then all firms
tend to hire more labor which, in turn, leads to more industry output. This
D more output is represented by a rightward shift in supply curve (to S'),
leading to a new lower equilibrium price, P'. This decline in price again
0 Q* Q’ Q
reduces VMPL ( = P'*MPL), implying a downward shift in individual demand
curve for labor (VMPL shifts to left from dL to d’L), see figure below.

W Demand of Firm W DL
Market Demand
a A
W1 W1

b b’
W2 W2 B
dL ∑dL
d’L ∑d’L

0 l1 l2 l’2 L 0 L1 L2 L

In the 1st panel, the decline in the individual input demand attributable to the decline in the commodity price (P* to
P’) is represented by the shift leftward from d L to d’. Thus, at input price W2, b is the equilibrium point, with 0l 2
units employed. So aggregating for all employers, 0L 2 units of the productive service are used (as indicated by point
B of the 2nd panel). Here it is the quantity demanded of labor on this new demand curve (d' L) that must be added in
order to get a new market demand (DL) when wage falls.

d) Supply of a Variable Productive Service: Labor

28
All variable productive services may be broadly classified into three groups: natural resources (or land),
intermediate goods, and labor.
 Intermediate goods are those produced by one entrepreneur and sold to another who, in turn, utilizes them
in the productive process. The supply curves of intermediate goods are positively sloped just like any other
commodities, because they are also the commodity outputs of manufacturers, even if they are variable
inputs to others.
 Natural resources may be regarded as the commodity outputs of (usually) mining operations. As such they
also have a positively sloped supply curve. If you consider land, however, its supply is usually constant in
the short run and hence we face a perfectly inelastic (vertical) supply curve. It doesn’t change with change
in rent in the short-run. But the supply curve of land is also upward sloping in the long-run.
 Here our attention is restricted to the most important category, labor. Note that the change in the size of the
population, age structure (labor force participation), and the occupational & geographic distribution of labor
force will shift the labor supply & hence they are constants (independent of its slope). Thus, we have to
analyze, what induces a person to forego leisure for work?

Indifference Curve Analysis of Labor Supply:

The supply of labor offered by one individual can be determined by indifference curve analysis. (see figure below).
Hours of leisure are measured along the horizontal axis, OZ representing the maximum, or the total number of
potential work hours in a day. The total money income from work is measured along the vertical axis.
Definition: Indifference curve (IC) shows the various combinations of leisure & income that yields the worker the
same level of satisfaction. And the income (wage) line shows the locus of points of leisure and income that the
worker can have with a given wage income. The slope of IC is marginal rate of substitution between leisure &
income; and the slope of a wage line represents a wage per hour. For instance, if 0M 1 is the money income that
could be received for 0Z hours of work (maximum hours of work), the hourly wage rate is OM 1/OZ, (say W1),
which is the slope of wage-line ZM1.

When the wage rate is W1 (the slope of ZM1), the tangency condition for maximization establishes equilibrium at
point e1 on IC I. The individual works Z1Z (or L1) hours for income of Z1e1. Leisure, therefore, is OZ1. Let the wage
rate increases to W2), as given by the slope of wage line (ZM 2). The new equilibrium is e2 on indifference curve II.
Now hours pf work expands from Z 1Z to Z2Z (or from L1 to L2) as a result of increase in the wage rate; workers
would further expand to Z3Z (or L3) if wage rises to W3 (on ZM3 income-line). But if wage rate further increases to
W4 and above, workers start preferring leisure time to work, and start reducing work time. With wage-line ZM 4,
equilibrium is e4 on IV curve. Here hours of work reduces from Z 3Z to Z4Z; but leisure time rises from OZ 3 to OZ4.
The equilibrium points e1, e2, e3, e4, --- can be connected by the dashed curve S, showing the supply of labor offered
by a single individual.

29
 Income from Fig. Derivation of Indv. Labor
work(M) Supply
NB: 24-Z3 24-Z2 24-Z1 & Z=24
M4
M3 W Sl
E4
W4
e4 S
M2
e3
E3
W3
M1
e2
IV
E2
III
e1 W2
II E2
I W1
0 Z3 Z2=Z4 Z1 Z Leisure 0 L1 L2 L3 Labor
Z 24-Z3 24-Z2 24-Z1 0 Work
Note that the individual supply curve of unskilled labor is always upward sloping both in the short-run and in the
long-run. Most economists argue that the slope of supply curve of skilled workers may be positive, negative, or both
in the short-run. In the long-run, however, it would be positively sloped (on average). Market supply curve of labor
has positive slope in the long run (for detail, see below).

Market Supply of Labor: In general, the supply curve of specialized workers is upward sloping mainly due
Wage to two factors:
SL 1. The newly educated people would like to join an industry where hourly wage rate
is relatively high. Young people plan their education in line with current &
expected returns.
2. Old (& retired) people also retrain themselves and join high wage sector.
Thus, though some workers prefer leisure to work at a very high wage rate, still
there are a large part of people would like to offer more services at high wage
0
Labor rate, leading to upward sloping labor supply curve in the long-run.
Even if labor is not specialized to one industry, the labor supply curve to any one industry must be positively
sloped, because to expand employment, workers must be obtained from the other industries, thereby increasing the
demand price of labor. In general, the supply curves of non-specialized types of labor are positively sloped. And in
the short-run the supply of specialized labor may take any shape or slope; but in the long-run it, too, tends to be
positively sloped.
e) Market Equilibrium: The demand for and supply of a variable productive factor jointly
Market equilibrium determine its market equilibrium price. As indicated in figure below, D L
Wage and SL are the market demand and supply curves, respectively. Their
SL intersection at point e determines the stable market equilibrium price
OW* and quantity demanded & supplied OL*. Thus, the commodity
market equilibrium and factor market equilibrium is the same – equality
W* e between market demand & supply. The only features unique to factor
market are the methods of determining the demand for and the supply of
productive factor (labor) services. Input demand is based on the value-
DL of-the-marginal product of that input; and labor supply is determined by
workers attitude between leisure & work. But commodity supply is cost-
Labor determined.
0 L*

3.2 Factor Pricing In Imperfectly Competitive Market

30
Just like the case of competitive markets, the demand and supply determine the price of the factor and the level of
employment when there are imperfections. But the determinants of demand and supply differ in this case which will
consider three models with three kinds of imperfections:

Model A: Input demand and employment assuming imperfect competition (monopoly) in the product market.
Model B: Monopsonistic power in the factor market & monopolistic power in the output market
Model C: Bilateral Monopoly

Model A: Monopolistic Power in the Output Market while

Factor Market is Competitive

1. The firm's Demand Curve: The case of one variable input

A firm can be a sole seller of a product and still compete with large number of firms in hiring inputs. In that case,
the firm is a monopoly in the product market and a competitor in its input markets.

Just like a competitive firm, the price that must be paid for an input measures the added (marginal) cost of
employing it in the case of monopoly also. The difference in the two market settings rests on how the added (one
more) input affects the firm's revenue. For a competitive firm, employing one more input unit adds to revenue an
amount equal to VMP (=P*MP). For monopoly, one more input unit adds to revenue by expanding output which
amounts marginal revenue product - MRP (= MR*MP).

w Graphically: Competitive firm's equilibrium is:

(VMPL = MRPL) = W, since P = MR, VMP=MRP; and
Monopoly's equilibrium is: MRPL = W; note that
VMPL > MRPL in monopoly because P is always
w SL greater than MR (since demand is down-sloping)
VMPL
MRPL

0 L1 L2 L
Thus, under down-ward sloping product demand curve, we will show that the demand for labor of an individual
firm is not the VMPL but the MRPL defined by multiplying the MPL with the MR of selling the commodity
produced. That is, MRPL = MPL*MR

Proof: Let the demand function for the product be P = f(Q), and total revenue (R) of the firm is: R = PQ. Hence,
marginal revenue (MR) becomes: MR = R/Q = (PQ)/ Q = P*Q/Q + Q*P/Q. Thus,
MR = P + QP/Q ----- (1)

Suppose production function with one variable input, labor, is Q = f(L) and hence marginal product of labor
becomes: MPL = Q/L ----- (2)
Now by definition the marginal revenue product of labor (MRP L) is the change in revenue (R) attributable to a unit
change in labor, i.e.
MRPL = R/L = (PQ)/ L = P*Q/L + Q*P/Q*Q/L---- with the help of chain rule
= Q/L(P + Q*P/Q)

But from equation (1) and (2): P + Q*P/Q = MR and Q/L = MPL.
 MRPL = MPL (MR)

Note that all firms demand and hire inputs to maximize profits. The monopoly firm that use a single variable input
in his production maximizes profit when the input's price is equal to its MRP. That is, equilibrium is reached when:
MRPL = W

31
Proof: Obviously, firms want to maximize profits (), which is the difference between revenue (R) & cost (C).
 = R - C, where R = PQ & C = WL + FC, where WL is variable cost & FC is fixed cost.
  = PQ - (WL + FC)
 = PQ - WL - FC
At equilibrium (or maximum profit) the first order derivative of profit with respect to labor must be zero; that is,
/L = 0
MRPL = /L = (PQ)/L - (WL)/L = 0
Q P Q  dL 
P* Q *  w   0
L Q L  dL 
, but P+Q*P/Q=MR and Q/L=MPL
Q  P 
 P Q* w  0
L  Q 
MPL ( MR )  w  0;
----- Necessary condition for profit maximization
MRPL  w

Therefore, given the perfectly elastic supply of labor for an individual (monopoly) firm and the MRP L, equilibrium
is attained when W = MRPL.

Graphically: If wage is W2 a monopoly firm demands l2 on MRPL at e2 and if wage rate fall to W1 the monopoly
hires l1 units of labor at point e 1 on MRPL. Thus, it follows that if the firm is a monopoly and uses only one
variable factor, the marginal revenue product (MRP L) is the demand curve of the input.

2) The Firm's Demand Curve: The Case of Several Variable Inputs

As in the case of perfect competition, firm's demand curve for input, labor (L), is no longer its marginal revenue
product curve rather it is formed from points on shifting MRP curve due to three effects of change in wage rate:
 Substitution effect,
 Output effect, and
 Profit-maximizing effect. Assume that the market price of labor is w and 1
its MRP is given by MRP1. Thus, the
MRP2 monopolistic firm is in equilibrium at point e1
W
hiring l1 units of labor on MRP1. If wage rate
MRP1
falls to W2 the MRP curve shifts to the right (to
e1 MRP2) due to the net effect of the above three
W1 effects. Thus, the new equilibrium occurs at
W2 e2 point e2 on MRP2 & the firm hires l2.

The locus of points such as e 1 & e2 is called the

demand curve for labor by the monopolist when
dL several factors are used. In general, the demand
curve for inputs is negatively sloped regardless of
0 l1 l2 L the degree of competition in the product market.

3) Market Demand for Labor: is the summation of the demand curves of the individual monopolistic firms
after taking into account their shift (as the price of the factor changes). For example, a fall in wage initiates
all monopolistic firms to expand their output, which of course, leads to a fall in market price. (This is
because the demand and MR curve for the commodity shifts to the left with fall in wage). Thus, the MRP L
(and hence dL) shifts inward. Derivation of market demand for input is just similar to the competitive case;
see figure below.

32
 w
w DL
w1 a A
w1 MRPL1

w2 b B B’
b’ w2

MRPL1 MRPL2
MRPL2

0 l1 l2 l’2 L 0 L1 L2 L

4) The Market Supply of Labor: is not affected by the fact that firms have monopolistic power in the product
market. Thus, individual and market supply of labor is just as derived earlier in competitive case. That is,
they are usually upward sloping, indicating a direct relationship between labor supply and wage rate.

5) Market Equilibrium: Given the market demand for and the supply of labor, the market price of labor is
determined by the interaction of the two curves.

(a) Firm Equilibrium (b) Market Equilibrium

w w
SL
Monopolistic
b Exploitation

wc

A
w a SL wm ∑VMPL
VMPL
MRPL ∑MRPL

0 l1 L 0 Lm Lc L

When the firm has monopolistic power in the output market the factor is paid its MRP which is smaller than the
VMP of an input. According to some economists (like Joan Robinson), this difference between VMP and MRP is
called monopolistic exploitation. The difference ‘ab’ in figure (a) and w cwm in figure (b) shows that the profit
maximizing behaviour of imperfectly competitive firms causes the factor price to be less than their VMP. Further
more, employment is higher under perfect competition than imperfect market (i.e. L c > Lm).

Some economists argue that Robinson’s exploitation cannot be accepted on its face value, because lower wage rate
in imperfect markets reflects the downward sloping demand curve which is due to brand loyalty of consumers (or
consumers’ desire for variety due to product differentiation).

Example: Assume that production function and demand function of a monopoly firm is given respectively by: Q =
100L1/2, and P = 100 – Q. If the price of labor is fixed at w, what is the demand function for labor? (State L as a
function of w). Answer: L = 50002/(w+10,000)2, where dL/dw < 0.

Model B: Monopsony in Input Markets and Monopoly in Product Markets

33
Up to this point we have assumed that there is perfect competition in input market. Now we begin by considering
the case of monopsony. Monopsony is a situation in which there is a single buyer (of a product and/or inputs). Thus,
pure monopsony in input markets occurs when a firm is the sole purchaser of an input. In this case the demand for
labor by the individual firm is the same as in model – A. That is, the demand for labor by a monopolistic firm is the
marginal revenue product of labor (MRP L). The supply curve of labor to the individual firm, however, differs,
which is our concern in the next section.

1) Input Supply Curves & Marginal Expenditure Curves:

The supply curve of the input facing the monopsonist is the market supply curve. The reason is that the
monopsonist is the entire market for the input> It is the sole buyer of the input. As a sole buyer, a monopsony
faces the market supply curve of the input, a curve that is often upward sloping.
W
Thus, in an input market the monopsony firm has some discretion over the
SL price paid for an input. If the monopsonist wants more of the factor, it must
pay a higher price not only for the additional units but for all the units of the
factor that it purchases. Hence, the marginal cost of the input exceeds the
price of that factor under monopsony; for example, MC L > W. That is, when
the firm faces an upward sloping input supply curve, marginal input cost is
always greater than average expenditure (or input price, which is also input
0 L supply)----MCL>W, where, W=AEL = fL(L) is input supply function.
Marginal expense of an input, say labor (ME L) is the change in total expenditure on the factor (say labor) arising
from one additional unit of the factor. That is, MEL=d(TEL)/dL, where TEL shows total expenditure on L, W*L.

Derivation of MEL: By definition MEL is change in TE when labor changes by one unit; i.e.
d (W * L) L W W
MEL  W L .Thus, MEL  W  L ------- (1)
dL L L L
Given the supply function fL(L), its slope is dW/dL >0. So the MEL is greater than W (or AEL) from expression (1).
Moreover, the MEL has a steeper slope as compared with the AEL or labor supply curve as shown in figure.
W
MEL
 The slope of the supply function (W=fL(L)) is dW/dL>0.
W=SL=AEL  And the slope of MEL curve is: d(MEL)/dL, or
d ( MEL ) d W dW  2W dW dL
 (W  L ) L 2  *
dL dL L dL L dL dL
dW 2 L
2 L 2
0 dL L
L Clearly the slope of MEL is greater than that of AEL (dW/dL). For
instance, if labor supply curve is linear, then d 2W/dL2 = 0 and slope of ME L = 2(slope of SL); if SL is a curvature,
MEL is even more steeper.

2) Equilibrium Price & Employment: Single Variable Input, Labor:

Note that firms demand & hire factors of production to maximize profit. The equilibrium of the monopsonistic firm
is given by the point where the marginal expenditure on the factor is equal its marginal revenue product.. That is, at
equilibrium: MEL = MRPL.

Proof: The firm maximizes its profit: π = R – C, where R = PQ and C = WL + FC. But at equilibrium the 1 st order
condition must be equal to zero; dπ/dL = 0.
  R C
 (R  C )  0    0  MRPL  MCL  0  MRPL  MCL
L L L L
Graphical illustration of equilibrium:

34
 MEL
W The monopsonist sets a price of OWe and employs OLe
SL=W is determined by equating MRPL with MEL at point e.
e But the monopsonist sets the input’s price (W) at the
level at which the quantity it demands (Le) will be
supplied; that is, wage rate (We) is determined by
substituting equilibrium level of employment into the
We
labor supply curve.
MRPL=DL

0
=
L
Le

3) Equilibrium Price & Employment: Several Variable Input, Labor:

The monopsonistic firm will stop increasing its use of each input when the input’s MRP equals the ME of the input.
Therefore, since an input’s MRP equals its marginal product times the firm’s marginal revenue, it follows that the
monopsonist will hire inputs so that:
 MRPL (=MPL*MR) = MEL, and hence MR = MEL/MPL ------- (1),
 MRPK (=MPK*MR) = MEK, and hence MR = MEK/MPK ------ (2) and so on.
 From equation (1) and (2): MEL/MPL = MR = MEK/MPK; or after rearranging this expression we get the
equilibrium situation of a monopsonist who uses several variable input:
MPL MPK MPN
    -------- (3)
MEL MEK MEN
Recall that in a perfectly competitive market ME of an input is equal to the price of the input; ME L=W, MEK=r.
Thus, for a competitive firm who uses several variable input equilibrium condition is given as:
MPL MPK MPN
 
W r PN
Monopsonistic (and monopolistic) Exploitation:
When the firm has a monopsonistic power in the input market it pays to the factor a price which is less not only than
its value-of-marginal product, but also less than its marginal revenue product. This gives rise to “ monopsonistic
exploitation”, which is something in addition to “monopolistic exploitation.” Workers are said to be exploited if
they receive a wage rate less than their VMP. These concepts are illustrated with the aid of a diagram as follows.
Wage
MEL  If we assume competition in both output & factor markets
equilibrium will occur at point A where SL(=W) and VMPL (or
dL) intersects leading to WC levels of wage & LC units of
SL=AEL employment.
Wc e S ec  If we assume monopoly in the output market while factor market
is competitive, the input market is in equilibrium at point e m
Wm em where MRPL=W. Now equilibrium wage & employment falls
respectively to Wm & Lm due to monopoly power. Thus, the
WS C difference between WC and Wm is the monopolistic exploitation.
VMPL
 If the firm has a monopsony power in the factor market,
MRPL equilibrium is reached at point eS where MRPL=MEL. To
maximize its profit the firm pays even a lower wage rate (W S) &
further reduces its employment to L s. Thus, the difference
0 LS Lm Lc Labor between WC & WS (=WCWm+WmWS) represents monopsonistic
monopsonistic elements. It would exist even if the firmexploitation,
were not a where the part
sole buyer in W CW
the M is not
labor uniquely
market. Theattributable
part W mW tos is
due to the monopsonistic power of the firm in the labor market.
Model C: Bilateral Monopoly
In this model we assume that all firms are organized in a single body (example, federation of manufacturers or
management of a firm) which acts like a monopsonist, while labor is organized in a labor (or trade) union which
acts like a monopolist (or as a single seller of labor). Thus, we have a model in which the participants are two

35
monopolies: (1) Trade union as a single seller of labor (monopoly)-supply side; (2) Sole buyer of labor
(monopsony) –demand side.

Next we will show that the solution to bilateral monopoly situation is “indeterminate”. The model gives only the
upper and the lower limits within which the wage rate will be determined by bargaining.

1. Supply side (Monopoly): From the point of view of the monopolist D b represents his
average revenue curve (ARS). Thus, given ARS marginal revenue curve of the seller (MR S) can be derived
by the usual graphic technique. If AR S is linear, then MRS is twice as steep as the AR curve; i.e. MR S <
ARS. Similarly, from the monopolist’s view point, S L represents the MC curve of the seller (labor union),
since the union considers the buyer as if they act as a perfect competitor. Thus, given the cost and revenue
curves, the monopolist (labor union) maximizes his gains at point U, where his MC intersects his MR. So,
the monopolist will want to supply LS units of labor and receive a wage equal to WS.
Wage 2. Monopsony (Single buyer of labor): In this
MEb figure the monopsonist’s demand curve is D b. It
is also the MRP of labor being demanded.
A SL=AEb=MCS Similarly, SL represents the supply of labor
WS (average expense of labor) facing the
F
monopsonist). Corresponding to this average
cost curve (AEb) is the marginal expenditure
(MEb) curve of the buyer. Thus, the
Wb b
U monopsonist (federation of manufacturers)
maximizes profit at point F, where his
MEb=MRPL. So the monopsonist will desire to
MRPL=Db=ARS hire Lb units of labor and pay the wage rate
equal to Wb.

0 LS Lb Labor
In summary, price desired by the monopsonist is
the lower limit (Wb), which can be realized only if
he could force the monopolist seller to act as a
MRS
perf-
ct competitor. Likewise, the price desired by the labor union (monopolist) is the upper limit (W S), which could be
realized if he could force the monopsonist (single buyer) to act as a perfect competitor.

Therefore, since the goal of the two parts can’t be realized, the price and quantity in the bilateral monopoly market
are indeterminate. The level at which the price will be settled depends on the bargaining skills and power of the
participants. The power of each participant is, in turn, determined by his ability to inflict losses to the opposite part
and his ability to withstand losses inflicted by the opponent. Usually, labor union use strike to get higher wages, and
buyers use dismissal of workers to pay lower wages.

36