• Game: a STRATEGIC situation in which a number of individuals make decisions, and each

cares both about his own choice and about others’ choices
• Example: game theory provides the foundation for understanding competition in industries with only
a few producers
• Example: every negotiation is a game, every battle is a game, every international crisis is a game
• Example: Penalties in football or hockey

• Game theory is the tool economists use to analyze strategic situations.

• Strategic decisions play a role when the effects of your actions depend on the actions and
reactions of other people. Strategic situations can be
1. over after one set of decisions taken SIMULTANEOUSLY (or equivalently, taken without any
information on the chosen decisions by others) by multiple players (one-stage games like within
regulation time penalty).
2. or may involve SEQUENCE of decisions (multiple-stage games – where a stage may describe a
decision taken by a single player like in a game of chess or a stage may describe a simultaneous
move game involving MULITPLE players like in a penalty shootout at the end of regulation
time).

• Sometimes all participants (players) in a game have access to the same information, but
sometimes different players have different information (asymmetric information)

For this class we will see not delve into the asymmetric information content as it is too advanced
 Two Types of Games EVEN IF
• One-stage game: each participant makes ALL of their choices before NONE HAVE
observing/knowing any choice by any other participant (effectively they make COMMITTED
SIMULTANEOUS MOVE) THEFT –
BOTH WILL
CONVICT
• Multiple-stage game: at least ONE participant observes a choice by another EACH OTHER
participant before making some decision (AT LEAST ONE of the PLAYERS – a gross
MOVES SEQUENTIALLY) miscarriage of
justice

• Describing an One-stage games require us to:

1. Identify the players and list the strategies available to each.
2. Identify each player’s payoff for every possible combination of strategies.
Two thieves are found with stolen goods. This charge alone is
enough to convict them to 2 years’ jail. Police suspect that both
have stolen it together, which if proven will give each of them 5
years’ jail. If one of them turns Government witness against the
other, he can help police establish that ONLY the other guy
engaged in the act of stealing. This means she gets 2 years of jail
with a 1 year exemption for helping the police convict a thief. The
other guy goes to jail for 5 years PLUS one year for obstructing
police.
*It is not illegal to use the
prisoners’ dilemma to get
conviction (irrespective of
guilt).

*This method is heavily used

by all investigating agencies
including ED and CBI.

*But thankfully SC has spoken

up questioning this method of
getting conviction. Though it is
surreal that top court looks for
fairness outside the existing set
of laws.

*This led to bail to K. Kavitha,

as the prosecution wanted to
convict her on the basis of
testimony provided by an
approver Raghav Magunta
Reddy, who was initially one of
the co-accused.
Now that the top court in the land has
spoken –hopefully things will change.
 Best Response
• Best response strategy with respect to a belief: a strategy that provides player with the highest possible
payoff, assuming A SPECIFIC (possibly wrong) BELIEF about what the other players will do.
• Having a belief is necessary since the game is simultaneous move, my welfare depends on BOTH of our choices.
So I cannot know what the other person will do for sure – I will have to imagine/construct a BELIEF, and then
best respond to it.
• Dominant strategy: sometimes a player finds that a single strategy or action is BEST RESPONSE to all
possible beliefs that she can have about what the other players will choose to play. Such an
action/strategy is knows as dominant strategy.
• she does not have to care how other people will behave and NEVER plays a strategy that is strictly dominated by
another strategy; and,
• other players can predict without any doubt in their minds that she will NEVER play a strategy that is strictly
dominated by another strategy.
• Strictly dominant strategy: A dominant action/strategy that gives the UNIQUE greatest payoff to ANY
possible belief about other player’s action/strategy. Without uniqueness it is called Weakly dominant
strategy.
Best Responses in the Prisoners’ Dilemma
 One-stage games

• Identify the players and list the strategies available to each

• Identify each player’s payoff for every possible combination of strategies

• And then look for,

• Best response: a strategy that provides player with the highest possible payoff, assuming other players
behave in a specified way.

• Dominant strategy: a specific strategy available to a player such that it is her best response to ANY
strategy that other players might play. You could have a STRICTLY DOMINANT strategy or a WEAKLY
DOMINANT strategy.

Three kinds of • Dominated strategy: roughly speaking any strategy OTHER than the dominant strategy, available to a
predictions/ equilibrium player. You could have STRICTLY DOMINATED strategy or a WEAKLY DOMINATED strategy. To be
notions based on best exact, it is a strategy corresponding to which you can find another unique strategy which dominates it.
responses:
• DSE • REMEMBER: A RATIONAL player will NEVER play a (weakly or strictly) dominated strategy. So we
use to this basic idea to develop our predictions of what will happen in a game. We call such a prediction
• IESDS as an “equilibrium”.
• Nash Equilibrium
• A profound application of an equilibrium prediction in a game using the logic of dominant strategy is the
“Second price sealed bid auction” – where the highest bidder wins the object but pays the second highest
bid as price. ------ MAGICALLY , HERE EVERYONE BIDS THEIR SECRET MONETARY
VALUATION FOR THE OBJECT.
 Dominant Strategy Equilibrium

S is a weakly dominated strategy for

any player P if there is ANOTHER
strategy S’ available to P that yields a
GREATER payoff than S in some
circumstances, and that never yields a
lower payoff regardless of others’
choices. So we can be sure that S will
not be played by both players.

The prediction obtained by expecting that no player would ever play a dominated strategy in equilibrium
is called DOMINANT STRATEGY EQUILIBRIUM (DSE).
In this case, it is (Down, Left), that is, Susan would play Down and Myrna would play Left.

12-7
 Iterative Deletion of Strictly Dominated Strategies (IESDS)
• Strictly Dominated strategy: if there is some other strategy that yields a strictly higher payoff regardless of
others’ choices

• Iterative deletion of STRICTLY dominated strategies: the process of removing the dominated strategies from
a game, resulting in a simplified game
1. Remove the strictly dominated strategies from a game
2. Inspect the simplified game to determine whether it contains any (new) dominated strategies. If it does, remove them
3. Repeat this process until there are no more dominated strategies left to remove.
 Prisoners’ Dilemma - Nash Equilibrium prediction
• Nash equilibrium: the strategy played by each individual is a best response to the strategies played by everyone else.
• In this case of Prisoners’ dilemma, NE prediction is (SQUEAL, SQUEAL).

Logically, any IESDS prediction is always a DSE prediction, while the latter is always an NE
prediction.
 NO NASH EQUILIBRIUM IN “PURE” STRATEGY IN “MATCHING COINS” GAME
Player B
Heads Tails

Heads 1, –1 –1, 1
Player A
Tails –1, 1 1, –1

• In this game, each player CHOOSES heads or tails and the two players reveal their coins at the same time.
If the coins match Player A wins and receives a dollar from Player B. If the coins do not match, Player B
wins and receives a dollar from Player A.
• Note that there is no Nash equilibrium in PURE (non-randomized) strategies for this game. No
combination of heads or tails leaves both players satisfied—one player or the other will always want to
change strategies unilaterally.
• There appears a Nash equilibrium IF WE ALLOW randomization in choosing strategies. This means we
allow each player, say i, to toss a biased coin that give heads with probability ‘p i’ – INSTEAD OF
CHOOSING HEAD OR TAIL SPECIFICALLY. Note that a pure strategy, thus becomes a special gamble
with degenerate probabilities. (see https://www.youtube.com/shorts/FHsL4DeKvQs)
• Alternatively, player A could use machine/computer program that plays the game instead of A playing in
person, and is programmed to play Head with some probability p A . Similarly for B.
• Then strategies become the programmed probabilities (p A and pB) – and are called MIXED
STRATEGIES. In almost all games a Nash equilibrium exists in mixed strategies – even when it may
not exist in pure strategies. Here: the Nash Equilibrium is p* A = p*B = ½.

(In game theory, unless specifically mentioned, we always assume players to be risk neutral- so even in
games with monetary payoffs, Bernoulli utility function W(x)=x.)
 How to compute Mixed Strategy Nash equilibrium

Player B
Heads Tails

Heads 1, –1 –1, 1
Player A
Tails –1, 1 1, –1

• A mixed strategy equilibrium with two players who have two available pure
strategies is given by (p*, q*) where p* and q* are positive fractions such that:
if the Genie were to come down and tell player A that player B will be playing
Heads with probability q* then, A will find herself indifferent between playing
Heads or Tails.

• That is: q*.1 + (1-q*). (-1) = q*.(-1) + (1-q*).(1) which implies that q*=1/2.

• Similarly, assuming p* is the Nash equilibrium probability of A playing Heads,

p*.(-1) + (1-p*).1 = p*.1 + (1-p*).(-1) which implies that p*=1/2.
 EXAMPLE OF A ONE STAGE GAME – HOTELLING’S LINEAR MARKET COMPETITION

• You (Y) and a competitor (C) plan to sell IDENTICAL soft drinks on a beach at IDENTICAL price. You
must choose a place along the beach to set up your stall. Your pure strategy set is [0,200] - unlike in
Prisoner’s Dilemma game where pure strategy set ={D, S}.
• If vacationers are spread evenly (meaning x/200 x 100% = x/2% of total consumers can be found to be
located in the region [0,x] as well as [200-x, 200]) across the beach and will walk to the closest vendor, it
can be predicted that the two of you will locate next to each other at the center of the beach. This is the
only Nash equilibrium, where both locate at the center of the linear market. Because
 If your competitor located at point A, you would want to move until you were just to the left, where
you could capture three-fourths of all sales.
 But your competitor would then want to move back to the center, and you would do the same.

(Note that mixed strategies in such games are mathematically very delicate. In any case, a pure strategy prediction is
usually considered to be more reliable than a mixed strategy one.)
• Big Billion Day Sale and Great Indian
Festival always start in on the same
day.

• Why? Hotelling’s market of

competition gives us some idea.

• Here is an instance in 2021, where

these were initially announced to start
at different dates – but later Flipkart
blinked/suffered from FOMO (fear of
missing out)/strategically recalibrated
their date of start to coincide with the
exact of start of Amazon’s Great
Indian Festival.
 Justifications for Nash Equilibrium
(pure or mixed strategy)

Why would we expect a group of people to settle on a stable combination of strategies?

When people play games repeatedly, they gain experience and learn how others tend to play. If
all players eventually learn to make accurate guesses, they will all play best responses to their
opponents’ actual decisions, effectively playing a Nash equilibrium.

Self-enforcing agreement/ Expectations are coordinated: every party to the agreement has an
incentive to abide by it, assuming others do the same – which is the essence of the “rational
expectations” that you may have studied in Macroeconomics.

Advantage of NE over IDSDS or DSE: ALWAYS EXISTS FOR ALMOST ALL PRACTICAL
GAMES (allowing for mixed strategies).
 Game with Perfect Information

• It is a special kind of multi-stage game

where every player makes her choice
KNOWING everything that happened
before.

• Equivalently, a perfect information

game is a multi-stage game that consists
of sequentially occurring peculiar stage
games that have a SINGLE player
choosing an action.

• A penalty shootout is NOT a perfect

information game. Chess is a perfect
information game.
Thinking Strategically in a Game with Perfect Information

Backward induction: the process of solving a strategic problem by reasoning in

reverse, starting at the end of the tree diagram that represents the game, and working
back to the beginning
 Backward induction and Nash Equilibrium
[NO QUESTIONS FROM THIS SLIDE FOR 2024]

• Tony’s NE strategy: choose the action-adventure film

• Maria’s NE strategy:
• if Tony chooses the action-adventure film, then choose the action-adventure film
• if Tony chooses the romantic comedy, then choose the romantic comedy

• Note how a perfect information game MUST involve sequential moves, and so, leads to a distinction in
concept of strategy and action (example: Maria).

• The general result is that in a game of finite perfect information, backward induction algorithm
leads us to the Nash equilibrium outcome. But we can think of other Nash equilibrium in this setup
which do not agree with Backward Induction.
• Note how this market is
almost dominated by two
players Tata Airlines
(companies) and IndiGo as
both companies together hold
around 82% market share.

• Also note how in deciding to

compete each other, the
companies think in terms of
buying planes instead of
setting prices.

• As will be argued later there

are some markets like the
airline market where there is
NO capacity available to
serve the immediate rush of
extra passengers obtained
upon price cutting.

• This is merely a quirk of the

underlying production
technology – and forces firms
to compete by choosing
outputs or production
CAPACITIES.
• Swiggy and Zomato have almost equal shares of the market.
• Both sell the same product that is food delivery.
• Question is should they compete by choosing prices or choosing quantity of output (in this caser persons working).
• As we will see later, since India is a labour surplus county, there are no capacity constraints for both producers, and so, they compete in prices.
• Finally, a new firm enters this market – Amazon.
New firm entry in an Oligopoly through commitment to avoid price war *An example where an oligopoly
structure deals with entry of a new
firm.

* Note how Grasim is committing

to enter this industry by spending a
lot of money, that is making the
first move (like in a sequential
game).

* This means that all other firms

have to expect Grasim as their
future competitor.

* What should they do? Should

they reduce prices to make
Grasim’s operations difficult/less
profitable the way Amazon and
Flipkart are fighting?

*Or should they make peace and

adjust their pricing and production
decisions accordingly.
GRASIM’S ENTRY IN THE COURNOT OLIGOPOLY PAINT MARKET
• Note how in this market, all the
firms are competing by choosing
their outputs/ ‘production
capacity’ instead of prices. Why?
Answer: Capacity Constraints.

• However, here the choices are not

simultaneous – in the sense that
the ASIAN PAINTS is reacting
over a year later to a commitment
of Grasim to enter the paint
market. So Asian Paints is a
second-mover.

• Such a type of market

competition is called Stackelberg
Oligopoly. Typically, in such
models the first mover is better
off in terms of equilibrium
payoffs.

• We will see later that if Asian

Paints could predict/anticipated
Grasim’s DESIRE entry (before
Grasim could commit itself to the
decision to enter), and become a
first mover, it could have stopped
Grasim from entering altogether.
 OLIGOPOLY
• A market is said to be an Oligopoly if:
1. There are several different firms competing in the market for homogeneous (oil) or slightly heterogeneous product (cars), but
no firm is price taker, that is, no firm treats price as fixed parameter in its profit maximizing calculations.
2. There are barriers to entry to this market – not every firm can easily enter the market.

• Every five years, the U.S. Bureau of the Census publishes four-firm concentration ratios that measure the fraction of
each industry’s NATIONAL sales accounted for by its four largest firms. Many economists believe that a
concentration ratio greater than 40% indicates an Oligopoly with all other producers being benign price takers – but
this measure has its problems.

• US Census Bureau reported in 2007 that this concentration ratio in Discount Department Stores (Walmart and
Target) was 97%; while that in Cigarettes (Phillip Morris and R.J. Reynolds) and Beer (Anheuser-Busch and
MillerCoors) were 88%.

• An alternative measure is the Herfindahl–Hirschman Index, or HHI which is squared sum of market shares of all
firms. According to US Justice Department guidelines, an HHI below 1,000 indicates a strongly competitive market,
between 1,000 and 1,800 indicates a somewhat competitive market, and over 1,800 indicates an oligopoly. So in an
industry with an HHI over 1,000, a merger that results in a significant increase in the HHI will receive special
scrutiny and is likely to be disallowed. In 2006-7:

PC Operating systems HHI=9182 Microsoft, Linux

Wide Body Aircraft HHI= 5098 Boeing, Airbus

Automobiles HHI= 1432 GM, Ford, Chrysler, Toyota, Honda, Nissan, VW

Retail Grocers (not same as HHI = 321 Wal-Mart, Kroger, Sears, Target, Costco, Walgreens, Ahold, Albertsons
department stores above)
*In both these cases here,
firms are competing in terms
of prices because both firms
have a production technology
that can easily expand
production to meet any extra
demand that can be earned by
price cutting.

*This is unlike airline industry,

or cement industry or paint
industry.

*Thus, the industry production

technology decides the nature
of competition that would
occur in a market- typically.

* It is not possible to rule out a

Cournot-Bertrand competition
– but that is for a more
advanced.
 BERTRAND MODEL
[https://youtube.com/shorts/aSIIeY47xM4?si=noqeqmEFvnroGWVs]

• Suppose Duopoly, that is, two sellers in the market selling identical (homogeneous) good in the market.
Such a model can be trivially generalized to multiple sellers, in which case, we shall call it an Oligopoly.
Example: Amul and Mother Dairy selling “dahi”/yogurt.

• Suppose both have zero fixed cost but constant COMMON marginal costs of production. (What if there is
a positive fixed cost instead zero? Only a mathematical game theory paper can answer that.)

• These identical firms set their prices simultaneously (recall that in the parlance of game theory, simultaneous
mover means that no firm knows any of her competitor’s strategic choice – in this case, the price charged – before
making her own strategic choice of price.)

• In case both firms set the same price, customers toss an unbiased coin to decide which firm to buy from.
 BERTRAND MODEL NASH EQUILIBRIUM – ZERO ECONOMIC PROFITS

Both (all) sellers charge an identical price equal to their COMMON marginal
cost when there are zero fixed costs as shown in the graph.

• In New Jersey (and Wisconsin) it was illegal before October 2021 to sell baked goods unless the food is prepared in a commercial kitchen that has been inspected
by the state government’s Department of Health and Senior Services. In New Jersey, meeting this requirement would typically cost $15,000 or more.

• In both states, legislators say\id that these are necessary to ensure sanitary cooking practices. Critics say that large quantities of cakes and cookies are given away
FREE each year in New Jersey and Wisconsin at church, school, and charity events, WITHOUT making many people becoming ill from eating this food. In other
48 states of US, where home bakeries are legal, there is no increase in food-borne diseases.

• You can see with BERTRAND model logic why commercial producers must be lobbying to KEEP home-baking illegal. Note that cookies are simply cookies, that
is, homogeneous products for consumers. In that case; both commercial firms and home bakeries would use the same/similar technology – commercial bakers
would be unable to charge any price above the marginal costs of the home bakeries leading to huge reduction in profits. They would also be unable to price out
home-bakeries from the market due to presence of anti-competitive regulations.

WAY FORWARD: The optimal, and hence, typical response of a commercial company in such markets is to invest heavily in advertisements to convince
customers that her product is “DIFFERENT”, and thereby build a brand loyalty that breaks the homogeneous goods assumption in Bertrand model, and
allows her to charge prices can be charged above marginal cost in Nash equilibrium. That is why the functional management disciplines of Marketing, an
particular advertising is so important. So what happens if you are successful in convincing the customer of your specialty? We study this a few slides later.

SOMETIME THIS STRATEGY LEADS TO HORRIBLE OUTCOMES, SO MORAL GUIDANCE IS A MUST IN PURSUIT OF PROFIT.
(Search for the 10-most-sexist-advertisements-ever in Google if you want an example of honest by immoral advertising.)
• Product Differentiation initiated in the
bottled mineral water market.

• All the main players, Kinley, Aquafina,

and Bisleri dominate the market with
Bertrand pricing.

• Due to huge entry barriers, in this case,

huge economies of scale that incumbent
competitors already have: new potential
entrant cannot compete with them and
make enough profits selling the plain
bottled water to enter the industry.

• Hence, we find Parle Agro and Blue Pine

enter with differentiated product for a
customer segment willing to pay more.

• This argument tells you how important is

marketing a product to differentiate it
from competitors. Also, do not stop there
but engage in aggressive surveys, and
then heavy marketing to induce higher
income customers to pay more by buying
the higher quality product.
 SOME HORRIBLE BUT POSSIBLY SUCCESSFUL ADVERTISING–
Liberal values are not as ubiquitous as the West would likes to pretend. Some of these came out when we had a female PM.

A criminal ad by Valentino shoes

 Bertrand Duopoly with Asymmetric Costs
• What happens when one of the duopolists has a lower
(say, constant) marginal cost of production than the
other.

• Here is the curious case of Imagine Store which WAS

an Apple premium reseller closing its very popular
shop in the South Delhi’s Select City-Walk Mall.

• Why? To understand this you must first understand

that resellers are third party private businesses that buy
Apple products from Apple at discount and then sells
to final customer. (
https://www.techtarget.com/searchitchannel/definition/
reseller
and
https://www.gearrice.com/update/the-price-the-main-d
ifference-between-the-apple-store-and-the-apple-resell
er/
)

• Imagine closed its store at Select City-Walk mall

because Apple itself opened its store at the same mall.
Apple did not ask it to be closed, like it asked around
20 other brands’ shops. Why?

• VERY SIMPLY HERE IMAGINE WAS SELLING A

HOMOGENOUS PRODUCT AT HIGHER MARGINAL COST AT
THE SAME PLACE AND SO HAD TO QUIT - and do what? -
Example of Cournot Oligopoly (not Monopolistic competition) market.
 COURNOT MODEL OF DUOPOLY
(a model of quantity competition)
 Firms choose how much to produce simultaneously (more
importantly, without any knowledge of what the other is
choosing), and the price clears the market given the total
quantity produced.

 Compared to Bertrand’s model, everything is same except

that the choice variable for both firms is quantity of
output. One may also interpret output as capacity because
firms would be unable to compete in prices unless they
have the capacity to produce the extra demand that they
may get by price reduction.

 This may be more realistic in markets where output

choices are heavily determined by machinery/ capital
input choices which cannot be changed in the short
run. That is, firms just cannot undercut and hope to meet
the extra demand quickly

 For example, Boeing vs. Airbus; American Airlines vs.

British Airways on the Chicago –London route during
new year. The number of airplanes dedicated to a sector at
a time of year cannot be changed SWIFTLY enough to
meet a surge in demand.
No content after this slide
will come in the
examination.
 COURNOT MODEL
(notion of best response to a belief ABOUT THE QUANTITY OF VOLUME THAT OPPONENT WILL BE
ABLE TO SELL using the idea of RESIDUAL DEMAND curve)
 COURNOT COMPETITION – NASH EQUILIBRIUM

• REACTION CURVE: shows a firm’s best choice in response to each possible action by its

rival.
Intersection of these reaction curves gives us the Nash equilibrium/ optimal output choice in
this market.
 MATHEMATICL EXPOSITION OF COURNOT DUOPOLY
• 2 firms
• Identical products
• Output choice competition
• Simultaneous choice of production amount: that is, one does not know the choice of the other
while making own choice
• Two firms face the same market demand curve P = 30 - Q where Q =Q 1+ Q2 = the total output produced
by the two firms, say, firm 1 and firm 2. Thus, Q i denotes the output choice made by the firm i.

• Both firms 1 and 2 are assumed to have the same constant marginal costs of production = C.
This is a simplifying assumption for class exposition purposes. So, MC 1= MC2=C.

• Total revenue of firm 1, TR1= PQ1= 30 x Q1 - Q12 – Q2 Q1, and marginal revenue of firm 1, MR1= 30 - 2Q1
– Q2. Total revenue of firm 2, TR2= PQ2= 30 x Q2 – Q22 – Q1 Q2, and marginal revenue of firm 1, MR2= 30
- 2Q2 – Q1.

• To find the Nash equilibrium outputs (Q1*, Q2*), we can think each firm 1 to be monopolist with the
residual demand curve while believing that the other firm 2 to be producing Q2* and vice-versa.

• That means, firm 1 thinks that she is facing a (residual) demand curve P = (30 - Q2*) – Q1 , and must best
respond to this belief by choosing Q1* that maximizes her profit given this belief. This requires her to
equate her marginal revenue from producing Q1* given her belief to her marginal cost. Thus, we get that
MR1= 30 - 2Q1* – Q2* = C =MC. We call this equation the reaction function of firm 1 in the Q1-Q2
plane.

• Arguing similarly for firm 2, we can arrive at the reaction function of firm 2 that MR2= 30 - 2Q2* – Q1*
 Cournot Duopoly - A Linear Demand Curve
• Solving the two reaction functions (one for each firm) we get that Nash equilibrium outputs produced by
each firm is same, that is, Q1*= Q2*= (30-C)/3, and the Nash equilibrium price in the market is (30+2C)/3.

• Note the demand (the link below) by Hyundai chairman (2019) to reduce GST to help tide over the then
current auto industry crisis demand crisis (it was a time when demand for televisions, vests, and cars were
falling alarmingly even before advent of Covid). So why is he asking the Govt. to introduce a supply side
relief to tide over a demand side problem?

• In a competitive market, this logic should lead to further collapse in the market price, which may NOT lead
to any increase in margins of cars sold in spite of lower costs of production – and may even lead to fall in
profits (if the demand at 2019 was too inelastic)!! In any case, this is a risky maneuver to demand of the
Government without CLEAR computable benefits. So once again, how can we rationalize this demand?

• The answer is that the hatchback market, and indeed the car market, is a Cournot oligopoly market. He is
making this demand because auto sales were down (for very specific political and economic reasons
beyond the scope of this course) so much that the value “30” in original demand curve had now become,
say, “20” leading to reduction in output produced and hence, labour employed.

• If taxes are reduced sufficiently, then “C” will reduce to say, “C’:=C-10”, and so, the initial output and
employment levels would be regained – albeit at a significant loss of Government revenue.

(https://indianexpress.com/article/business/companies/govt-support-may-help-revive-demand-hyundai-motor-india-5921942/)
 ANOTHER EXAMPLE - COURNOT MARKET

• As a part of deregulation of electricity markets 1994 onward, the state of California decided to setup
wholesale markets where buyers and sellers could trade electricity one day ahead of its actual use.

• The buyers were predominantly three electric utilities, Pacific Gas & Electric, Southern California Edison,
and San Diego Gas & Electrics.

• The rules of the wholesale market allowed each producer to choose the amount of its capacity that would
be available to supply electricity to the buyers.

• It was empirically recorded in research papers and otherwise, that during periods of high demand (and low
supply from other very small capacity sources), the five large producers purposefully limited the amount
of capacity available for sale to enhance the market price – but there was NO EVIDENCE OF CARTEL
TYPE COLLUSION.

• Thus, this must have been Cournot competition playing out where competitors simultaneously decided
how much output they wanted to sell in the market.
 STACKELBERG DUOPOLY

• The same setup as Cournot –with only one difference that one of the two firms, say G (say, Grasim) chooses production capacity Q G in an
IRREVERSIBLE manner (say public disclosure under SEBI norms) before firm A (say Asian Paints) makes a production capacity choice.
• Thus, apparently A has an advantage because she can see what her competitor has chosen to do, and then best respond rationally.

• However, this is now a two-period perfect-information game like the Tony-Maria game that we saw earlier. So just like there, it is the first
mover that has advantage and will get greater paypoff in the backward induction equilibrium (also known as SUBGAME-PERFECT-NASH-
EQUILIBRIUM.
• Thus, as in Tony-Maria game, G will predict the BEST RESPONSE of A (to ANY possible choice that G can make) – and then make the choice
that maximizes A’s profit.

• If the same functional forms as the Cournot example used earlier hold, then the BEST RESPONSE. REACTION function of A will now be
QA(QG), since A can now observe what G has done irreversibly, and as argued in the Cournot case, to maximize profit, it must choose MR= 30 -
2QA(QG) – QG = C=MC, which means that QA(QG) = .

• Since A is rational, and is aware of all the parameter values (a plausible assumption given these are huge industrial powerhouses with
excellent technical, statistical and espionage facilities while working with an old and mature production technology) – she can compute that
QA(QG) = , and so, now believes that the demand curve she faces is P= 30 - QG - QA(QG) = 30 - QG -
• Now behaving as a monopolist with respect to this demand curve, G equates MR = = C =MC, and produces , while A produces .

In reality, Asian paints is likely to have greater economies of scale than Grasim, meaning that its marginal cost is to be lower than that of Grasim – which explains the symmetry of
equilibrium observed in real life (where both decide to invest 10000 crores)
 WHAT IF ASIAN PAINTS (AP) COULD HAVE GUESSED GRASIM’S PLAN TO ENTER BEFORE GRASIM ACTUALLY STARTED INVESTING
MONEY/MAKING FACTORIES. THEN AP COULD HAVE BECOME THE FIRST MOVER IN THE ENSUING STACKELBERG GAME.
(Could AP have done something to stop Grasim from entering? Let us see using game theory’s Stackelberg model. For the purposes of propriety – let us say AP is
Rebecca and GRASIM is Joe.)

• Let us suppose that REBECCA’s firm is incumbent, while the JOE’s is planning to enter the industry to set up shop by spending $15000. So IF Joe ENTERS, he
gets to choose an output production AFTER observing what Rebecca’s output choice is or has been over the years. Consider all the following numbers to be net
present values.
• This means Rebecca can calculate the exact reaction curve of Joe.

Since the entry (eventually sunk) cost of Joe is 15000$ to enter the industry. Then Rebecca will choose to produce (or commit to produce) 4000 units, which
will deter Joe from entering (as post entry profits would be 10000$), and thus, yield Rebecca a profit of 80000$.
• Product Differentiation initiated in the
bottled mineral water market.

• All the main players, Kinley,

Aquafina, and Bisleri dominate the
market with Bertrand pricing.

• Due to huge entry barriers, in this

case, huge economies of scale that
competitors already have: new
potential entrant cannot compete with
them and make enough profits selling
the plain bottled water to enter the
industry.

• Hence, we find Parle Agro and Blue

Pine enter with differentiated product
for a customer segment willing to pay
more.

• This argument tells you how important

is marketing a product to differentiate
it from competitors. Also, do not stop
there but engage in aggressive
surveys, and then heavy marketing to
induce higher income customers to
pay more by buying the higher quality
product.
 DIFFERENTIATED PRODUCT BERTRAND OLIGOPOLY – an evolution

• The modern automobile industry was created by Henry Ford, who first introduced assembly-line production. This
technique made it possible for him to offer the famous Model T at a far lower price than anyone else was charging for a
car; by 1920, Ford dominated the automobile business (as predicted by our Bertrand model).

• But to ensure maximum cost reduction, and hence, price reduction, he insisted on making his cars in ONLY black colour.

• He quickly captured all the market, and reaped what we call economies of scale (falling average costs), and gained a
huge cost advantage over his other competitors, who were obviously late in catching up to the new technology (invented
by Henry Ford). So by the Bertrand competition logic, the low cost firm would put other firms making the same product
(that is, mechanised transport device) out of business, and become a monopolist. THAT DID NOT HAPPEN…why?

• Alfred P. Sloan, who emerged as a challenger to Henry Ford by merging a number of smaller automobile companies into
“General Motors” so that the aggregate merged entity could achieve the economies of scale due to falling average costs
(note that this strategy was also taken by competitors of RELIANCE JIO when it launched its telecom product) .

• However, that was not enough – so he did what we noted in an earlier slide. He introduced PRODUCT
DIFFERENTIATION, that he differentiated his product from the Model T produced by “Ford” which was always
BLACK.

• So Sloan’s strategy was to offer a range of car types, differentiated by quality and price. Chevrolets were basic cars that
directly challenged the Model T, Buicks were bigger and more expensive, and so on up to Cadillacs – and each model
was available in several different colors.

• By the 1930s , Sloan had won. Customers obviously preferred a range of styles, and General Motors, not Ford, became
the dominant auto manufacturer for the rest of the twentieth century.

THEREIN LIES THE VALUE OF ADVERTISING/MARKETING THAT CAN INSERT PRODUCT DIFFERENTIATION
IN MINDS OF CUSTOMERS EVEN WHEN THERE IS NONE – EXAMPLE: CARS, MINERAL WATER ETC.
 DIFFERENTIATED PRODUCT BERTRAND DUOPOLY
• What if there are no infrastructural constraints binding? For example, each player in the market has
enough free capacity to serve any customers it can win away from its competitors simply because product
in question is perhaps easy to make. In that case, it is natural that such market would witness price
competition instead of quantity competition.

• Recall that we noted that in a Bertrand model with homogeneous objects, all players would like to invest
in marketing or advertising campaign to differentiate its own object vis-à-vis others. Such a market
becomes a differentiated product price competition where similar products are treated as separate goods by
consumers owing to their brand loyalty.

• Best example: Coke and Pepsi. Lets take an example, say demand for Coke depends on its own price as
well as the price differential from Pepsi’s chosen, and vice-versa.
QCoke = 45 - 50PCoke + 200(PPepsi - PCoke )
QPepsi = 45 - 50PPepsi + 200(PCoke - PPepsi )

• Both firms choose prices simultaneously to compete with each other. They have a CONSTANT marginal
cost of 0.3 rupees per can. Note that both firms are profit maximizers, and so, for any given price that they
might believe to be chosen by their competitor, P
they equate their marginal revenues to 0.3 to decide their
MR (Q ) P (Q)  ( )Q
own price choice. Recall that Q ; and so, MRCoke = 0.18 + 0.8PPepsi - 0.008QCoke.
Equating this expression with 0.3, we get QCoke = -15 + 100PPepsi which implies that
PCoke = 0.24 + 0.4PPepsi

• This is the reaction curve for Coke. Note that it is best response for her to reduce price if she believes that
Pepsi would do so. Similarly, we can obtain the reaction curve for Pepsi as below:
PPepsi= 0.24 + 0.4PCoke
•
 MONOPOLY FORMATION IN OLIGOPOLY AND ITS REGULATION

• Often a handful of sellers in an Oligopoly decide to initiate communication and coordination

to fix prices in a manner that they effectively become a monopoly.

• Governments strongly discourage such behaviors by passing laws which typically are called
Anti-trust regulations.

• In 1996 Archer Daniels Midland (ADM) and two other producers of lysine (an animal feed
additive) pled guilty to charges of price fixing in USA. In 1999 three ADM executives were
sentenced to prison terms of two to three years. (See the Steven Soderbergh movie THE
INFORMANT starring Matt Damon – AFTER THE EXAMS).

• In 1999 four largest drug and chemical companies of the world - Hoffman-La Roche of
Switzerland, BASF of Germany, Rhone Poulenc of France, and Takeda of Japan – accepted
their guilt in fixing prices of vitamins sold in U.S. and Europe. The companies paid about $1.5
billion in penalties to the U.S. Department of Justice (DOJ), $1 billion to the European
Commission, and over $4 billion to settle civil suits. Executives from each of the companies
did prison time in the U.S.

• Eight companies, mostly in Korea and Japan, fixed DRAM (memory chip) prices from 1998
to 2002. In 2007, 18 executives from these companies were sentenced to prison terms in the
USA.
 MONOPOLISTIC COMPETITION
• A market UNLIKE Oligopoly – there are no barriers to entry, but products are differentiated. The closest we can get to
perfectly competitive (physical) markets in real life. Typically, assumed to have quantity competition.

• Example: Soap, shampoo, deodorants, shaving cream, cold remedies, and many other items found in a ‘kirana’ shop are
sold in monopolistically competitive markets. So is most retail trade, because goods are sold in many different stores that
compete with one another by differentiating their services according to location, availability and expertise of salespeople,
credit terms, etc. In each such case, entry is relatively easy, and so, if profits are high in a neighborhood because there are
only a few stores, new stores will enter.

• In the long run, these profits

attract new firms with competing
brands. The firm’s market share
Residual
falls in the new Nash equilibria
with larger number of firms, and Residual
this is gets reflected in a
downward shift of its product’s
demand curve.

• In long-run equilibrium, described

in figure (b), price equals average
cost, so the firm earns zero profit
even though it has monopoly
power. So free entry and exit in
market eliminates any monopoly
profits in long run like the same
feature in a perfectly competitive
market.
 COMPARISON OF MONOPOLISTICALLY COMPETITIVE LONG RUN EQUILIBRIUM AND
PERFECTLY COMPETITIVE LONG RUN EQUILIBRIUM
• Under monopolistic
competition, price exceeds
long run marginal cost, and
production is carried out at
an inefficient level where
costs are not minimized.

• Thus there is a deadweight

loss vis-à-vis the
competitive market, as
shown by the yellow-
shaded area.

• The demand curve is

downward-sloping, so the
zero profit point is to the In both types of markets, free entry occurs until profits are driven to
left of the point of zero.
minimum long run average
cost. This is known as
presence of excess
capacity.
 MONOPOLISTIC COMPETITION AND DIFFERENTIATED GOOD OLIGOPOLY

Monopolistic Competition = Differentiated good Oligopoly – Entry barriers

So restaurant market constitutes Monopolistic competition but Wide body airplane market is an
Oligopoly due to presence of entry barriers in the latter market. But why may there be barriers
to entry in an Oligopoly? Possible reasons are as follows:

• Economies of Scale: The falling long run average costs are such that the output at which this
average cost is MINIMIZED constitutes a large proportion, at least 30% of the LARGEST
possible market demand. In such a scenario, the market cannot support more than 3 or 4
firms.

• Patents and licenses: Governments often forbid entry of new firms without requisite
permissions either to protect the quality of product traded, or to protect the incumbent who
has created the market through aggressive spending on R&D.

• Ownership of Critical Input: Until the 1990s, Ocean Spray, in USA, faced negligible
competition in the market for fresh and frozen cranberries as it controlled almost the entire
supply of cranberries. Today, Ocean Spray controls about 65% of the cranberry crop supply
through agreements with 650 cranberry farmers. Similarly, for many years, the Aluminum
Company of America (Alcoa) controlled most of the world’s supply of high-quality bauxite,
the mineral needed to produce aluminum, so much so that the only way other companies
could enter the industry was to recycle aluminum.
 Entry Deterrence
(Output choice to ensure there is not new entrant in the market)
• Thus, when the first move is an INCUMBENT, by expanding one’s output sufficiently, a firm may reduce
the profit its rival foresees enough to deter them from entering the market – and so, discourage any entry.
This is a costly strategy since often this requires sacrifice of own profits.

• As an alternative and less costly strategy, often firms prefer to set up the excess capacity (by signing a
labour contract or setting up a plant) as a first mover but not actually increase production prior to possible
entry of competitor. This is less costly than the previous strategy.

• Effectively, by setting up the excess capacity, the existing first mover firms are threateningly
COMMITTING TO the entrant that they will render this decision to enter a financially COSTLY decision
by flooding the market with output and collapsing market price if the new firms DARE to enter. And to do
so they have already incurred some cost.

• For example, DuPont (https://www.dupont.com/) once tried to deter rivals from producing titanium dioxide
(a whitener used in paints and plastics) by committing to an extensive expansion of its facilities.
 Entry Deterrence and associated risk

• Commitment versus flexibility: Though a firm can sometimes gain by pre-committing to

future actions, there is a potential downside to the strategy if market conditions are uncertain.

• Put simply, pre-committing limits the firm’s ability to respond to changing market conditions
as the entry that is to be discouraged may take over a year to take place

• For example, DuPont once tried to deter rivals from producing titanium dioxide (a whitener
used in paints and plastics) by committing to an extensive expansion of its facilities.

• Unfortunately for DuPont, an unexpected recession struck, and market demand turned out to
be much lower than DuPont forecasted, causing the firm’s clever strategy to yield a
disappointing loss.

(HENCE THE NEED FOR PROFESSIONAL STATISITICIANS/ ECONOMISTS WHO CAN PREDICT BUSINESS CYCLES.)