8
COMPETITIVE
GOODS MARKETS
September 2014 beta
HOW A PERFECTLY COMPETITIVE MARKET OPERATES AND WHAT IT
MEANS FOR BUYERS AND SELLERS.
You will learn:
• What it means for a market to be perfectly competitive.
• The effect of competition on the ability of individual buyers or sellers to influence prices.
• What a differentiated product is.
• The efficiency and fairness implications of perfectly competitive markets.
• Why real world markets are not typically perfectly competitive.
• The similarities and differences between firms in competitive markets and firms with
monopoly power.
Funded by the Institute for New Economic Thinking with additional funding from Azim Premji University and Sciences Po
2
coreecon | Curriculum Open-access Resources in Economics every day, 220 million people buy an item of clothing. Different customers have
different tastes in clothes. How can each of these people buy the t-shirt or pair of
jeans they want without any of the people involved in the garment’s production, from
growing the cotton to working in the clothing store, knowing or caring anything
about the purchasers?
At every stage of the process is a market, with many buyers and many sellers. On one
side of the retail market are consumers; on the other side shops are competing with
each other to attract them. The shops participate in other markets, as buyers rather
than sellers: labour markets to employ staff and wholesale markets where they buy
large clothing in large quantities from manufacturers and importers. Manufacturers
in turn are sellers in wholesale markets, but they buy textiles (and buttons), and
employ designers and production workers. At the far end of the supply chain,
workers are employed to plant cotton.
The market participants do not know each other, but they are able to make decisions
about whether (and how much) to buy and sell by looking at the prevailing price in
the market. And when something changes in one market—a poor cotton harvest, for
example—its effects will be transmitted through prices to others.
In Unit 7 we considered the case of a product produced and sold by just one firm.
There was one seller in the market for that particular product. In this unit we look at
markets where many buyers and sellers interact, and show how the market price is
determined by both the preferences of consumers and the costs of suppliers. When
there are many firms producing the same product, each firm’s decisions are affected
by the behaviour of competing firms, as well as consumers.
8.1 BUYING AND SELLING: DEMAND AND SUPPLY
as an example of a market with many buyers and sellers, think about the
potential for trade in second-hand copies of a recommended textbook for a university
economics course. Demand for the book comes from students who are about to begin
the course, and they will differ in their willingness to pay (WTP). No one will pay more
than the price of a new copy in the campus bookshop. Below that, students’ WTP may
depend on how hard they work, how important they think the book is, and on the
available resources for buying books.
Figure 1 shows the demand curve. As in Unit 7, we line up all the consumers in order
of willingness to pay, highest first. The first student is willing to pay $20, the 20th
$10, and so on. For any price, P, the graph tells you how many students would be
willing to buy: it is the number whose WTP is at or above P.
3
UNIT 8 | COMPETITIVE GOODS MARKETS
Price, P (buyers' willingness to pay, $)
25
20
15
10
5
Demand curve
0
0
5
10
15
20
25
30
35
40
45
Quantity, Q, of books (number of buyers)
Figure 1. Market demand curve for books.
The demand curve represents the WTP of buyers; similarly supply depends on the
sellers’ willingness to accept (WTA) money in return for books. The supply of secondhand books comes from students who have previously completed the course, who
will differ in the amount they are willing to accept—that is, their reservation price.
Recall from Unit 5 that Angela was willing to enter into a contract with Bart only if
it gave her as least as much utility as her reservation option (no work and survival
rations); here the reservation price of a potential seller represents the value to her
of keeping the book, and she will be willing to sell only for a price at least that high.
Poorer students (who are keen to sell so that they can afford other books) and those
no longer studying economics may have lower reservation prices.
We can draw a supply curve by lining up the sellers in order of their reservation
prices (their WTAs): see Figure 2. We put the sellers who are most willing to sell—
those who have the lowest reservation prices—first. The first seller has a reservation
price of $2: and will sell at any price above that. The 20th seller will accept $7, and the
40th seller $12.
coreecon | Curriculum Open-access Resources in Economics 14
Price, P (sellers' reservation prices, $)
4
Supply curve
12
10
8
6
4
2
0
0
5
10
15
20
25
30
35
40
45
Quantity, Q, of books (number of sellers)
Figure 2. Supply curve for books.
Now, if you choose a particular price, say $10, the graph shows how many books
would be supplied (Q) at that price: in this case, it is 32. The supply curve slopes
upward: the higher the price, the more students will be willing to sell.
Notice that we have drawn the supply and demand curves as straight lines for
simplicity. In practice they are more likely to be curves, with shapes depending on
how valuations of the book vary amongst the students.
DISCUSS 1: SELLING STRATEGIES AND RESERVATION PRICES
Imagine that you are planning to move to a city with good public transport and
restricted parking, so you wish to sell your car. You are considering three possible
methods:
1. Advertise it in the local newspaper.
2. Take it to a car auction.
3. Offer it to a second-hand car dealer.
Would your reservation price be the same in each case? Why? If you used the first
method, would you advertise it at your reservation price? Which method do you
think would result in the highest sale price? Which method would you choose?
UNIT 8 | COMPETITIVE GOODS MARKETS
8.2 THE MARKET AND THE EQUILIBRIUM PRICE
what would you expect to happen in the market for this textbook? That will
depend on the market institutions that bring buyers and sellers together. If students
have to rely on word-of-mouth, then when a buyer happens to find a seller they can
try to negotiate a deal that suits both of them. But each buyer would like to be able
to find a seller with a low reservation price, and each seller would like to find a buyer
with high willingness to pay. Before concluding a deal with one trading partner they
would like to know about other trading opportunities.
Traditional market institutions often brought many buyers and sellers together
in one place. Many of the world’s great cities grew up around marketplaces and
bazaars along ancient trading routes such as the Silk Road between China and the
Mediterranean. In the Grand Bazaar of Istanbul, one of the largest and oldest covered
markets in the world today, shops selling carpets, gold, leather and textiles cluster
together in different areas of the bazaar. In medieval towns and cities it was common
for makers and sellers of particular goods to set up shops close to each other, so
customers knew where to find them. The City of London is now a financial centre,
but evidence of trades once carried out there can be found in surviving street names:
Pudding Lane, Bread Street, Milk Street, Threadneedle Street, Ropemaker Street, Poultry,
and Silk Street.
With modern communications, sellers can advertise their goods and buyers can
more easily find out what is available, and where. But in some cases it is still
convenient for many buyers and sellers to meet together; large cities have markets
for meat, fish, vegetables or flowers, where buyers can inspect and compare the
quality of the produce. In the past, markets for second-hand goods often involved
specialist dealers, but nowadays sellers can contact buyers directly through online
marketplaces such as eBay. Returning to our example of second-hand textbooks, local
online sites now help students sell books to others in their university.
To analyse the market for the textbook, we assume that all the books are the same
(although in practice some may be in better condition than others) and that a
potential seller can advertise a book for sale by announcing its price on a local
website.
At the end of the 19th century, the economist Alfred Marshall described a market as
perfect if conditions were such that the law of one price would hold. In such a market,
everyone who traded would do so at the same price: no buyer would pay a high price
if the good were available at a lower price elsewhere.
5
6
coreecon | Curriculum Open-access Resources in Economics past ECONOMISTs
ALFRED MARSHALL
Alfred Marshall (1842-1924) was Professor
of Political Economy at the University of
Cambridge between 1885-1908. His work
was motivated by a desire to improve the
material conditions of working people.
Countering the prevailing pessimism, he
realised that wages would rise if labour
productivity could be improved. His
Principles of Economics, published in 1890,
laid out the supply and demand framework,
and emphasised the importance of marginal
quantities. Other economists were working
on similar ideas, but it was Marshall who
developed them fully over a period of 20
years before publishing them. John Maynard
Keynes paid tribute to Marshall’s achievement by comparing it with the lessdeveloped work of Stanley Jevons, considered one of the greatest economists of the
1870s: “Jevons saw the kettle boil and cried out with the delighted voice of a child;
Marshall too had seen the kettle boil and sat down silently to build an engine.”
We might expect that the law of one price would hold in the online marketplace for
a second-hand textbook, or at least that most trades would occur at similar prices if
the books were in the same condition. Buyers and sellers can easily observe all the
advertised prices, so if some books were advertised at $10 and others at $5, buyers
would be queuing up to pay $5, and these sellers would quickly realise that they could
charge more, while no-one would want to pay $10 so these sellers would have to lower
their price.
If all books are sold at the same price, what will the price be? To answer this
question we introduce the idea of equilibrium. In everyday language something is in
equilibrium if the forces acting on it are in balance, so that it remains still. We say
that a market is in equilibrium if the actions of buyers and sellers have no tendency
to change the price or the quantities bought and sold (as long as there is no change
in market conditions such as the numbers of potential buyers and sellers, and how
much they value the good).
7
UNIT 8 | COMPETITIVE GOODS MARKETS
By drawing the supply and demand curves on one diagram, as in Figure 3, we can
find the equilibrium price in this market. At a price P* = $8, the supply of books is
equal to demand: 24 buyers are willing to pay $8, and 24 sellers are willing to sell. The
equilibrium quantity is Q* = 24 as shown by point A in Figure 3.
25
Price, P
20
15
Supply curve
10
A
P*
5
Demand curve
0
0
10
20
Q*
30
40
Quantity, Q, of books
Figure 3. Equilibrium in the market for second-hand books.
At the equilibrium price, all those who wish to buy or sell are able to do so. If the
price were higher than $8, more students would wish to sell, but not all of them
would find buyers, so the sellers would want to lower the price. If it were lower, there
would be more buyers than sellers, and the sellers would realise that they could raise
the price. There no tendency for change only at exactly $8.
8.3 PERFECT COMPETITION
in the market equilibrium that we have described for a second-hand
textbook, none of the buyers or sellers is able to choose the price at which they
trade: if anyone tried to get a more advantageous price for themselves, no-one would
want to trade with them, because they could find an alternative buyer or seller. The
participants in this market are price-takers, because there is sufficient competition
from other buyers and sellers that the best they can do is to trade at the market price.
No individual has any power to affect the market price.
A market equilibrium in which all buyers and sellers are price-takers is called
perfectly competitive.
8
coreecon | Curriculum Open-access Resources in Economics In a perfectly competitive market equilibrium, the law of one price holds. All
trade takes place at the price that equalises supply and demand. We say that the
equilibrium price clears the market.
We have seen other examples of markets where participants do not behave as pricetakers: the producer of a differentiated product can set its own price because it has
no close competitors. In Unit 6 we saw that, because labour contracts are incomplete,
an employer may not seek to pay the lowest possible price for a worker, but instead
choose to set a higher wage.
What kinds of markets would we expect to be perfectly competitive? That is to say, in
what conditions would all buyers and sellers be price-takers?
One way to answer this question is to construct a hypothetical model of a situation in
which we would predict price-taking. Imagine a market with many buyers and sellers,
where all the goods for sale are identical. Buyers and sellers are aware of all prices
and trading opportunities in the market. Every buyer wants to buy at the lowest
available price, and every seller wants to sell at the highest possible price. In these
conditions, we can predict that that the law of one price would hold, and everyone
would have to accept the market price—to take it as given.
This hypothetical picture seems unrealistic, but it suggests where we might look
for examples of perfect competition. Markets for agricultural products such as
wheat, rice, coffee, or tomatoes look rather like this, although goods are not truly
identical, and it is unlikely that everyone is aware of all trading opportunities. But
it is nevertheless clear that they have very little, if any, power to affect the price
at which they trade: they are price takers. In other cases—for example, markets
where there are some differences in the quality of goods—there may nevertheless
be enough competition that we can assume price-taking, in order to obtain a simple
model of how the market works. A simplified model can provide useful predictions
when the assumptions underlying it are only approximately true. Judging when it is
appropriate to draw conclusions about the real world from a simplified model is an
important skill of economic analysis.
We saw in Units 4 and 5 how the institutions in which economic interactions take
place—the rules of the game—determine the bargaining power of participants. A
competitive market is an institution in which no one has any bargaining power. An
individual’s gain from participating in the market depends on the price at which that
person trades, but that individual has no power to affect that price.
In the scenario in Unit 5, competition from other workers wanting to work for Bart
would eliminate Angela’s bargaining power. Then Bart could make a take-it-or-leaveit offer, obtaining the whole of the surplus, while Angela, with no power to refuse,
would obtain only her reservation utility. In a competitive market, competition on
both sides of the market eliminates the bargaining power of both buyers and sellers.
Sellers can’t raise the price because of competition from other sellers; competition
from other buyers prevents buyers from lowering it.
UNIT 8 | COMPETITIVE GOODS MARKETS
DISCUSS 2: CONSUMERS IN PERFECTLY COMPETITIVE MARKETS
A market is more likely to be competitive if consumers always choose to buy at the
lowest price available. Think about some of the goods you buy: different kinds of
food, clothes, transport tickets, electronic goods, etc. Do you try to find the lowest
price? If not, why not? For which goods would price be your main criterion?
In the online marketplace for a second-hand textbook, where everyone can observe
what is going on, the sellers compete with each other and so do the buyers. But even
if you agree that competition will prevent sellers from advertising their books for sale
at different prices, you may not believe that they will all immediately pick $8, leading
to market equilibrium. Initially they are unlikely to know enough about demand and
supply to know the equilibrium price.
Marshall introduced the idea of a perfectly competitive market, and he recognised
this problem. In the 19th century most English towns had a Corn Exchange (known
in the US as a grain exchange)—a building where farmers met with merchants to sell
their grain. Marshall described how on market day in the Corn Exchange “the price
may be tossed hither and thither like a shuttlecock”. But he argued that it would soon
settle at the equilibrium level, and then stay there unless supply or demand changed.
Economists have studied the behaviour of buyers and sellers in laboratory
experiments, to assess whether prices do adjust to equalise supply and demand. In
the first such experiment, in 1948, Edward Chamberlin gave each member of a group
of Harvard students a card, designating them “buyers” or “sellers” and stating their
willingness to pay or reservation price in dollars. They could then bargain amongst
themselves, and he recorded the trades that took place. He found that prices tended
to be lower, and the number of trades higher, than the equilibrium levels. One of
the students who took part, Vernon Smith, later conducted his own experiments. He
modified the rules of the game so that participants had more information about what
was happening: buyers and sellers called out prices that they were willing to offer
or accept. When anyone agreed to a proposed deal, a trade took place and the two
participants dropped out of the market. His second modification was to repeat the
game several times, with the participants keeping the same card in each round.
Figure 4 shows his results. There were 11 sellers, with reservation prices between
$0.75 and $3.25, and 11 buyers with WTP in the same range. The diagram shows the
corresponding supply and demand functions. You can see that, in equilibrium, six
trades will take place at a price of $2. But the participants did not know this, since
they did not know the price on anyone else’s card. The right-hand-side of the diagram
9
coreecon | Curriculum Open-access Resources in Economics shows the price for each trade that occurred. In the first period there were five trades,
all at prices below $2. But by the fifth period most prices were very close to $2, and
the number of trades was equal to the equilibrium quantity.
4.0
Supply
3.6
3.2
2.8
2.4
2.0
P
1.6
0
Period 5
Demand
Period 4
0.4
Period 3
0.8
Period 2
1.2
Period 1
Price ($)
10
0 1 2 3 4 5 6 7 8 9 10 11 12
01234512345123451234567123456
Quantity
Number of transactions in each period
Figure 4. Vernon Smith’s experimental results.
Results like these suggest that the perfect competition model provides a good
prediction of what will happen in a market where goods are identical, and there
are enough buyers and sellers, who are well-informed about trading by others, to
generate competition. The outcome was close to equilibrium even in the first period,
and converged quickly towards it subsequently as the participants learned more
about supply and demand, just as Marshall argued that it would do.
But although prices may adjust to equalise demand and supply in controlled
laboratory conditions, we shouldn’t necessarily expect this to happen in real markets
for all kinds of goods. We will look later in this unit at evidence suggesting that the
effects of competition on prices may be weak, and in Unit 9 at some examples where
prices do not clear markets.
UNIT 8 | COMPETITIVE GOODS MARKETS
DISCUSS 3: PRICE MATCHING
Retailers sometimes promise to match their competitors’ prices: if you can find the
same item on sale cheaper elsewhere, they will refund the difference. Do you think
that this strategy is a sign of a competitive market? Large supermarkets go further,
checking the prices of competitors themselves, and giving you an automatic refund if
the items in your shopping basket would have cost less (in total) elsewhere. This BBC
news report investigates their strategies: LINK.
8.4 FIRMS IN COMPETITIVE MARKETS
in the second-hand textbook example, both buyers and sellers are individual
consumers. Now we look at markets where the sellers are firms. In Unit 7 we analysed
the decisions of a firm producing a differentiated good, and saw that if other firms
made similar products the demand curve would be almost flat. The firm’s choice of
price would be restricted, because raising its price would cause consumers to switch
to other, similar, brands.
What would happen if there were many firms producing identical products? If
consumers were well-informed about the suppliers of the product, and could easily
switch from one firm to another, then we would have a perfectly competitive market.
Firms would be price-takers, with no discretion over price at all.
To see how competition affects the behaviour of firms, consider a city where many
small bakeries produce bread and sell it direct to consumers. Figure 5 shows what
the market demand curve—the total daily demand for bread of all consumers in the
city—might look like. It is downward-sloping as usual because, at higher prices,
fewer consumers will be willing to pay.
11
coreecon | Curriculum Open-access Resources in Economics 4.5
4.0
3.5
Price, P (€)
3.0
2.5
2.0
1.5
1.0
Demand curve
0.5
0.0
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,000
Quantity, Q: number of loaves
Figure 5. The market demand curve for bread.
Suppose that you are the owner of one small bakery, with the isoprofit curves and
marginal costs shown in Figure 6. Your marginal cost curve—indicating, at each
quantity, Q, the cost of making one more loaf—slopes upward. When the quantity
is small the marginal cost is low, close to €1; having installed mixers, ovens and
other equipment, and employed a baker, the additional cost to produce a loaf of
bread is relatively small. Marginal costs rise gradually as the number of loaves
per day increases because you have to employ extra staff and use equipment more
intensively.
7
6
Price, P; Cost (€)
12
5
Marginal cost curve
4
Isoprofit curve: €200
3
P*
2
Isoprofit curve: €80
Zero-economic-profit
curve (AC curve)
Feasible set
1
0
0
20
40
60
80
100
120
140
160
180
200
Quantity Q: number of loaves
Figure 6. Isoprofit curves and marginal cost curve for the bakery.
UNIT 8 | COMPETITIVE GOODS MARKETS
Now look at the isoprofit curves. As in Unit 7, costs include the opportunity cost
of capital, so our calculation of profit gives economic profit , which is what you,
as the owner, will obtain in addition to the normal profits you would expect as a
return on your capital. The light blue curve is the zero-economic-profit curve: the
combinations of prices and quantities at which you break even. Remember from Unit
7 that this is the firm’s average cost curve. The darker blue isoprofit curves show
higher levels of profit. As we explained in Unit 7, isoprofit curves slope down where
price is above marginal cost, and up where price is below marginal cost. Where price
is equal to marginal cost, the isoprofit curve is horizontal—the marginal cost curve
passes through the lowest point on each isoprofit curve.
You have to decide what price to charge and how many loaves to produce each
morning. Suppose that neighbouring bakeries are selling loaves identical to yours at
€2.35. Since the market is competitive, you are a price-taker: you cannot sell loaves
at a higher price than other bakeries. So the feasible set is the grey-shaded area in
Figure 6: all the points where price is less than or equal to the market price P* =
€2.35.
The problem looks similar to the one for the carmaker in Unit 7 except that, for a
price-taker, the demand curve is completely flat. For your bakery, it is not the market
demand curve in Figure 5 that affects your own demand; it is the price charged by
your competitors. If you charge more than P* your demand will be zero; at P* or less
you can sell as many loaves as you like.
To maximise profits you should choose the point in the feasible set that reaches the
higher possible isoprofit curve. This is at Q*, or 120 loaves, where the horizontal line
at P* is tangent to the middle isoprofit curve. You should produce 120 loaves and sell
them at the market price.
Figure 6 illustrates a very important characteristic of firms in competitive markets.
They choose to produce a quantity at which the marginal cost is equal to the market
price (MC = P*). This is always true. Perfect competition means that the firm’s own
demand curve is a horizontal line at the market price; so maximum profit is achieved
at a point on the demand curve where the isoprofit curve is horizontal. At this point
the price is equal to the marginal cost.
Another way to understand why a firm in a competitive market produces at the
level of output where MC = P* is to think about what would happen to its profits if it
deviated from this point. If the firm were to increase output to a level where MC > P*,
the last unit would cost more than P* to make, so the firm would lose on this unit and
would make higher profits by reducing output. If it were to produce where MC < P*, it
could produce at least one more unit and sell it at a profit. Therefore it should raise
output as far as the point where MC = P*. This is where profits are maximised.
Put yourself in the position of the bakery owner again. What would you do if the
market price changed? The three isoprofit curves we have drawn give you part of the
answer. From Figure 7 you can see that, if P* were to rise to €3.20, you would be able
13
coreecon | Curriculum Open-access Resources in Economics to reach the $200 isoprofit curve by producing 163 loaves per day. If the price fell
to €1.52 you could reach only the zero-economic profit curve, and your best choice
would be 66 loaves. But we don’t really need to draw any more isoprofit curves: if
you know the market price, the marginal cost curve tells you the profit-maximising
choice of quantity. In other words, the firm’s marginal cost curve is its supply curve,
showing you how much it will produce at each possible level of the market price.
7
7
Marginal cost curve
Isoprofit curve: €200
6
6
Isoprofit curve: €80
Zero economic profits
MC
4
3
5
Price, P; Cost (€)
5
Price, P; Cost (€)
14
Supply curve
4
3
2
2
1
1
0
0
40
80
120
160
200
0
0
40
80
120
160
200
Quantity Q: number of loaves
Figure 7. The firm’s supply curve.
Notice, however, that if the price fell below €1.52 you would be making a loss. The
supply curve shows how many loaves you should produce to maximise profit, but
when the price is this low, the economic profit is nevertheless negative. On the
supply curve, you would be minimising your loss. If this happened you would have
to decide whether it was worth continuing to produce bread. If you expected market
conditions to remain this bad, it might be best to sell up and leave the market—
elsewhere, you could obtain a better return on your capital. But if you expected the
price to rise soon, you might be willing to incur some short term losses—and it might
be worth continuing to produce bread if the revenue helped you to cover the costs of
maintaining your premises and retaining staff.
15
UNIT 8 | COMPETITIVE GOODS MARKETS
8.5 MARKET SUPPLY AND EQUILIBRIUM
the market for bread in the city is competitive: there are many consumers and
many bakeries. Let’s suppose there are 50 bakeries in the city. Each one has a supply
curve corresponding to its own marginal cost curve, so we know how much it will
supply at any given market price. To find the market supply curve, we just add up the
total amount that all the bakeries, together, will supply at each price.
Figure 8 shows how this works if all the bakeries have the same cost functions. In
this case we know, for example, that if the price is €2.35, each bakery will produce
120 loaves, and we can say that the market supply at €2.35 is 50 x 120 = 6,000 loaves.
At a price of €1.52 they each supply 66 loaves, and market supply is 3,300.The market
supply curve looks like the firm’s supply curve, except that the scale on the horizontal
axis is different. If the bakeries have different cost functions, then at a price of
€2.35 some bakeries will produce more loaves than others, but we can still add them
together to find market supply.
5
5
Market supply (marginal cost)
4
4
3
3
Price, P (€)
Price, P (€)
Firm supply (marginal cost)
2
1
0
2
1
0
40
80
120
160
200
0
0
2,000
4,000
6,000
8,000 10,000
Quantity Q: number of loaves
Figure 8. The firm and market supply curves.
Notice that the market supply curve still represents the marginal cost of producing
a loaf, just as the firm’s supply curve does. Suppose the price is €2.75 and all firms
have the same cost functions. Then each firm will supply 140 loaves and we know
that its marginal cost is equal to the price: the cost of producing one more loaf would
be €2.75. Total supply in the market is 7,000 loaves. The cost of producing one more
loaf (that is, the 7,001st loaf) must be €2.75 too. It doesn’t matter which firm produces
the extra loaf, because if the market price is €2.75 they all have marginal cost equal
coreecon | Curriculum Open-access Resources in Economics to €2.75. This would be still be true if the bakeries had different marginal cost curves:
they would have different levels of output, but would all choose their output so that
their marginal cost was €2.75.
Now we know both the demand curve (Figure 5), and the supply curve (Figure 8) for
the bread market as a whole. Figure 9 shows that the equilibrium price is exactly
€2.00. At this price, the market clears: consumers demand 5,000 loaves per day, and
firms supply 5,000 loaves per day.
4.5
Supply
(marginal cost)
4.0
3.5
3.0
Price, P (€)
16
2.5
A
2.0
1.5
1.0
Demand
0.5
0
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000 10,000
Quantity Q: number of loaves
Figure 9. Equilibrium in the market for bread.
Would you expect the market to adjust quickly to equilibrium? In this example it
seems quite plausible, since the supply and demand curves are unlikely to change
much from day to day. Bakeries that were left with unsold loaves, or had to turn away
disappointed customers, would quickly adjust their prices and quantities to bring
supply in line with demand.
In the market equilibrium, each bakery is producing on its marginal cost curve, at
the point where its marginal cost is €2.00. If you look back to the isoprofit curves in
Figure 7, you will see that the firm is above its average cost curve—the isoprofit curve
where economic profits are zero. So the owners of the bakeries are receiving economic
rents—profit in excess of normal profit. We might expect this to attract other bakeries
into the market. We shall see in Unit 9 how the entry of more firms would increase
the supply of bread in the longer term, and could eventually reduce economic profits
to zero, eliminating rents.
17
UNIT 8 | COMPETITIVE GOODS MARKETS
8.6 GAINS FROM TRADE, ALLOCATION AND DISTRIBUTION
we can calculate the gains from trade in a competitive market, as we did for
the markets in Unit 7. At the equilibrium price of €2 in the bread market, shown in
Figure 10, a consumer who is willing to pay €3.50 obtains a surplus of €1.50. The red
shaded area shows total consumer surplus—the sum of all the buyers’ gains from
trade. Remember from Unit 7 that the producer’s surplus on a unit of output is the
difference between the price at which it is sold, and the marginal cost of producing
it. The marginal cost of the 2,000th loaf, for example, is €1.25; since it is sold for €2,
the surplus is €0.75. The purple-shaded area is the sum of the bakeries’ surpluses on
every loaf that they produce. The whole shaded area shows the sum of all gains from
trade in this market, known as the total surplus.
4.5
Supply
(marginal cost)
4.0
3.5
Price, P (€)
3.0
Consumer
surplus
2.5
A
2.0
Producer
surplus
1.5
1.0
Demand
0.5
0
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000 10,000
Quantity Q: number of loaves
Figure 10. Gains from trade.
The equilibrium allocation of bread is the one that maximises the total surplus. If
fewer than 5,000 loaves were produced, there would be consumers without bread
who would be willing to pay more than the cost of producing another loaf, so there
would be unexploited gains from trade. And there are no gains to be made from more
than 5,000 loaves, because none of the other consumers is willing to pay more than
they would cost to make. At the equilibrium, all of the potential gains from trade are
exploited.
This result is true in general: the equilibrium allocation in a perfectly competitive
market maximises the sum of the gains achieved by trading in the market, relative to
the original allocation. We demonstrate this result in EINSTEIN 1.
18
coreecon | Curriculum Open-access Resources in Economics EINSTEIN 1
However the market works, and whatever prices are paid, we can calculate the
consumer surplus by adding together the difference between WTP and price paid of
all the people who buy, and the producer surplus by adding together the difference
between price received and marginal cost of every unit of output:
Consumer surplus = sum of WTPs – sum of prices paid
Producer surplus = sum of prices received – sum of MCs of each unit
Then when we calculate the total surplus, the prices paid and received cancel out:
Total surplus = sum of WTPs of consumers – sum of MCs of producers
The equilibrium allocation in the competitive market makes this as high as possible
by including the consumers with the highest WTPs, and the units of output with the
lowest marginal costs. Every trade involves a buyer with higher WTP than the seller’s
reservation value, so the surplus would go down if we omitted any of them. And if we
tried to include any more units of output in this calculation, the surplus would also
go down because the WTPs would be lower than the MCs.
In other words, it is not possible to make any of the consumers or firms better off
(that is, to increase the surplus of any individual) without making at least one of
them worse off. Provided that what happens in this market does not affect anyone
other than the participating buyers and sellers, we can say the equilibrium allocation
is Pareto efficient (We will consider below a case where this is not true, when noise
from the bakeries disturbs their neighbours).
UNIT 8 | COMPETITIVE GOODS MARKETS
DISCUSS 4: MAXIMISING THE SURPLUS
Consider a market for the tickets to a football match. Six supporters of the Blue team
would like to buy tickets; their valuations of a ticket (their WTP) are 8, 7, 6, 5, 4 and
3. Six supporters of the Red team already have tickets, for which their reservation
prices (WTA) are 2, 3, 4, 5, 6 and 7.
1. Draw the supply and demand “curves” (They are step functions like the ones in
Vernon Smith’s experimental market in Figure 4).
2. Show that, in equilibrium, four trades take place. What is the equilibrium price?
Calculate the consumer (buyer) surplus by adding up the surpluses of the four
buyers who trade.
3. Similarly calculate the producer (or seller) surplus, and hence find the total
surplus in equilibrium.
4. Suppose that the market operates through bargaining between individual buyers
and sellers, rather than competitively. Find a way of matching the buyers and
sellers so that more than four trades occur. (Hint: suppose the highest WTP buyer
buys from the highest WTA seller.) Work out the surplus from each trade. How
does the total surplus in this case compare with the equilibrium surplus?
5. Starting from the allocation of tickets you obtained through bargaining in
question 4, in which at least five tickets are owned by Blue supporters, is there
a way through further trade to make one of the supporters better off without
making anyone worse off? (You can assume that none of them are football
hooligans.)
When an allocation of economic resources is not Pareto efficient, we sometimes say
that there is a deadweight loss. We can illustrate the deadweight loss by looking at
Figure 10a. Suppose that the market price was €2.00, but the bakeries were producing
only 4,000 loaves. Then the gains from trade in the market would be lower. There
would be a deadweight loss, shown by the triangle-shaped area, consisting of the
consumer and producer surplus that would be lost. Producers would be missing out
on potential profits, and some consumers would be unable to obtain the bread they
were willing to pay for.
19
coreecon | Curriculum Open-access Resources in Economics 4.5
Supply
(marginal cost)
4.0
3.5
3.0
Price, P (€)
20
Consumer
surplus
2.5
DWL
2.0
A
Producer
surplus
1.5
1.0
Demand
0.5
0
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000 10,000
Quantity Q: number of loaves
Figure 10a. Deadweight Loss.
But we know this is not an equilibrium. The price is higher than the marginal cost
of the 4,000th loaf, and in a competitive market the bakeries have an incentive
to expand production. The market will clear, with 5,000 loaves sold, and the
deadweight loss will be eliminated.
We can apply the concept of deadweight loss to the car manufacturer in unit
7, which sets its price above marginal cost (the cost of the last unit produced),
leaving potential gains from trade unexploited. The total surplus in that case is
less than it would be if the firm behaved like a firm in a competitive market and
produced at the point where price equals marginal cost. The loss of surplus relative
to the hypothetical competitive case is the deadweight loss and, because the car
manufacturer has no incentive to expand production, the deadweight loss persists.
Pareto efficiency comes about in our model of the bread market because of some
particular conditions that we have assumed. First, perfect competition means that
participants have no market power. They are price takers. Hence the suppliers will
choose their output so that the marginal cost (the cost of the last unit produced) is
equal to the market price. In contrast, a firm producing a differentiated product (such
as a car) faces less competition and has the power to set its own price as a result. It
chooses a higher price, above marginal cost.
Second, the exchange of a loaf of bread for money is a very simple economic
interaction, in which each party knows exactly what he or she is getting. Compare
it with the employment contract in Unit 6: the firm can buy the worker’s time, but
cannot be sure how much effort the worker will put in. The firm chooses to pay
more than the worker’s reservation wage as an incentive for effort. In the labour
market equilibrium there will be some unemployed workers who are willing to work
UNIT 8 | COMPETITIVE GOODS MARKETS
at the equilibrium wage, so the allocation of jobs will not be Pareto efficient. The
key difference between the bread market and the labour market is that the labour
contract is incomplete: it can specify hours of work but not effort.
Third, when we say that that the allocation of bread is Pareto efficient, we are
assuming that what happens in this market affects no one except the buyers and
sellers. But if, for example, the early morning activities of bakeries disrupt the sleep
of local residents, then there are additional costs of bread production; we ought
to take the costs to the bakeries’ neighbours into account when we assess Pareto
efficiency. Then, we may conclude that the equilibrium allocation is not Pareto
efficient after all. We will investigate this type of problem in Unit 10.
Even if we think that the market allocation is Pareto efficient, we should not
conclude that it is necessarily a desirable one. Remember from Unit 4 that there are
two criteria for assessing an allocation: efficiency and fairness. What can we say
about fairness in the case of the bread market? We could examine the distribution
of the gains from trade between producers and consumers: Figure 10 shows that
both consumers and firms obtain a surplus, and in this example consumer surplus
is slightly higher than producer surplus, and in this example consumer surplus is
slightly higher than producer surplus. You can see that this happens because the
demand curve is relatively steep compared with the supply curve. Recall from Unit 8
that a steep demand curve corresponds to a low elasticity of demand. Similarly the
slope of the supply curve corresponds to the elasticity of supply: in Figure 10, demand
is less elastic than supply. In general the distribution of the total surplus between
consumers and producers depends on the elasticities of demand and supply.
We might also want to take into account the standard of living of participants in the
market. For example, if a poor student buys a book from a rich student, we might
think that an outcome in which the buyer paid less than the market price (closer
to the seller’s reservation price) would be better, because it would be fairer. Or, if
the consumers in the bread market were exceptionally poor, we might decide that
it would be better to pass a law setting a maximum bread price lower that €2.00 to
achieve a fairer, although Pareto inefficient, outcome. In Unit 9 we will look at the
effect of regulating markets in this way.
So we need to be careful with the result that the equilibrium allocation in a perfectly
competitive market is Pareto efficient. It is often interpreted as a powerful argument
in favour of competition, and of markets as a mechanism for allocating resources.
But first, it may not true if we have not taken everything into account; second, even if
it is true, there are other important considerations such as fairness; and third, as we
shall in the next section, most markets are not perfectly competitive.
21
22
coreecon | Curriculum Open-access Resources in Economics DISCUSS 5: SURPLUS AND DEADWEIGHT LOSS
1. Sketch a diagram to illustrate the competitive market for bread, showing the
equilibrium where 5,000 loaves are sold at a price of €2.00. Suppose that the
bakeries get together to form a cartel. They agree to raise the price to €2.70, and
jointly cut production to supply the number of loaves consumers demand at
that price. Shade the areas on your diagram to show the consumer surplus, the
producer surplus, and the deadweight loss caused by the cartel. (Remember that
the deadweight loss is the difference between the total surplus in the competitive
market and the total surplus under the cartel.)
2. For what kinds of goods would you expect the supply curve to be highly elastic?
Draw diagrams to illustrate how the share of the gains from trade obtained by
producers depends on the elasticity of the supply curve.
8.7 TESTING FOR PERFECT COMPETITION
if we look at a market in which conditions seem to favour perfect competition—
many buyers and sellers of identical goods—how can we tell whether it is, in fact, a
perfectly competitive market? Economists have used two tests:
1. Do all trades take place at the same price?
2. Are firms selling goods at a price equal to marginal cost?
The difficulty with the second test is that it is often difficult to measure marginal
cost. But Lawrence Ausubel was able to do this for the US bank credit card market
in the 1980s. 4,000 banks were selling an identical product: credit card loans. The
cards were mostly Visa or Mastercard, but the individual banks decided the price of
their loans—that is, the interest rate. The banks’ cost of funds—the opportunity cost
of the money loaned to credit card holders—could be deduced from other interest
rates in financial markets. Although there were other components of marginal cost,
the cost of funds was the only one that varied substantially over time. If the credit
card market were competitive, we would expect to see the interest rate on credit card
loans rise and fall with the cost of funds. Ausubel found that this didn’t happen. As
23
UNIT 8 | COMPETITIVE GOODS MARKETS
Figure 11 shows, when the cost of funds fell from 15% to below 7%, there seemed to
be almost no effect on the price of credit card loans. Why do the banks not cut their
interest rates when their costs fall?
Credit card
interest rate
Interest rate & cost of funds (%)
20
15
10
Cost of funds
5
0
1982
1983
1984
1985
1986
1987
1988
1989
Figure 11. Ausubel’s credit card data.
He suggested two possibilities. One is that consumers do not respond much to
differences in interest rates because they find it difficult to change credit card
provider. In that case the banks are not forced to compete with each other; so they
to keep prices high when costs fall. A second explanation might be that there is a
problem in this market: banks cannot tell which of their customers are bad risks, and
it is the bad risks who are most sensitive to prices. The banks do not want to lower
their prices for fear of attracting the wrong kind of customer.
Perfect competition requires that consumers are sufficiently sensitive to prices to
force firms to compete, and this may not be the case in any market where consumers
have to search for products. If it takes time and effort to check prices and inspect
products, they may decide to buy as soon as they find something suitable, rather
than continue the search for the cheapest. When the internet made online shopping
feasible, many economists hypothesised that this would make retail markets more
competitive: consumers would easily be able to check the prices of many suppliers
before deciding to buy.
But often consumers are not very sensitive to prices, even in this environment. You
can test the law of one price in online retail competition for yourself, by choosing a
particular product—a book, DVD, or household appliance, for example—that should
be same wherever you buy it, and checking the prices at which it is available. Figure
12 shows the prices of UK online retailers for a particular DVD in March 2014. The
range of prices is high: the most expensive seller is charging 66% more than the
cheapest.
24
coreecon | Curriculum Open-access Resources in Economics The Hobbit: An Unexpected Journey
SUPPLIER
PRICE INCLUDING
POSTAGE ($)
Game
14.99
Amazon UK
15.00
Tesco
15.00
Asda
15.00
Base.com
16.99
Play.com
17.79
Zavvi
17.95
The HUT
18.25
I want one of those
18.25
Hive.com
21.11
MovieMail.com
21.49
Blackwell
24.99
Figure 12. Differing prices for the same DVD, from UK online retailers, March 2014.
DISCUSS 6: PRICE DISPERSION
What explanations can you suggest for the differing prices for The Hobbit?
The economist Katy Graddy studied the Fulton Fish Market in Manhattan, an
institution that appeared to encourage competition. There were about 35 dealers,
with stalls close to each other; customers could easily observe the quantity and
quality of fish available and ask several dealers for a price. She recorded details of
2,868 sales of whiting by one dealer, including price, quantity and quality of fish, and
characteristics of the buyers.
UNIT 8 | COMPETITIVE GOODS MARKETS
Of course, prices were not the same for every transaction: quality varied, and fish
supplies changed from day to day. But her surprising result was that on average Asian
buyers paid about 7% less per pound than white buyers. (All of the dealers were
white.) There seemed to be no differences between the transactions with white and
Asian buyers that could explain the different prices.
How could this happen? If one dealer was setting high prices for white buyers,
why did other dealers not try to attract them to their own stalls by offering a better
deal? Graddy’s results suggested that that the dealers did have some discretion in
price-setting; the price difference arose because the Asian buyers were more price
sensitive, and persisted because the two groups of buyers were not aware of the
difference.
8.8 ASSESSING THE MODEL OF PERFECT COMPETITION
the evidence in the previous section suggests that many real markets do not
conform to the model of perfect competition. So how should we assess the model? Is
it still useful? Why don’t we see more cases of perfectly competitive markets?
We have encountered some reasons already. For example, buyers or sellers may
not be well-informed about the prices at which others are trading, as in the case
of the Fulton fish market. Another reason that the market for a product may not
be perfectly competitive is if the market demand function and the production cost
function, together, are not compatible with perfect competition.
To understand how this can happen, remember that in the face of competition from
other firms, the best a firm can do is to produce where its marginal cost is equal to
the market price. If you look back at the isoprofit curves and the marginal cost curve
for the bakery in Figure 7 you will see that it must sell at least 66 loaves at a price
of at least €1.52, if it is to make normal profits. It must operate at a scale where its
marginal cost is at least as high at its average cost. In a small town there may not be
enough demand for bread to sustain a market with many firms. For example, if the
market demand for bread was 80 at a price of €1.52 there wouldn’t even be enough
demand to support two bakeries. One bakery could survive, but then the market
would no longer be competitive and the bakery could choose its own price.
This may seem to be an extreme example. But it is true, in general, that a firm in a
competitive market can only make normal profits if it can operate at a scale where its
marginal cost is greater than or equal to its average cost—that is, where its average
cost is rising. If the cost function for a product is such that the average cost is high
at low levels of output and the marginal cost rises slowly, if at all, then perfect
competition may be impossible.
25
coreecon | Curriculum Open-access Resources in Economics 26
We also know that markets are not perfectly competitive when products are
differentiated. Consumers’ preferences differ, and we saw in Unit 7 that firms have an
incentive to differentiate their product, if they can, rather than to supply a product
similar or identical to others. Nevertheless, the model of perfect competition can be a
useful approximation, to help us to understand how some markets for non-identical
products behave.
Figure 13a shows the market for an imaginary product, Choccos, for which there are
close substitutes: many similar products compete in the wider market for chocolate
bars. Due to competition from other products, the demand curve is almost flat. The
range of feasible prices for Choccos is narrow, and the firm chooses a point where its
marginal cost is close to its price. So this firm is in a similar situation to a firm in a
perfectly competitive market.
Isoprofit
curve
P*
A
Demand
for Choccos
Market
supply (MC)
Price, P ($)
Price, P ($)
Marginal cost
of Choccos
B
Demand for
chocolate bars
Q*
Quantity of Choccos
Figure 13a. The market for Choccos. Total quantity of chocolate bars
Figure 13b. The market for chocolate bars.
The narrow range of feasible prices for this firm is determined by the behaviour of
its competitors. So the main influence on the price of Choccos is not the firm, but
the market for chocolate bars as a whole.Since all the firms will be producing at
similar prices, which will be close to their marginal costs, we lose little by ignoring
the differences between them and assuming that each firm’s supply curve is its
marginal cost curve. Then we can construct the market supply curve in Figure 13b by
finding the total number of chocolate bars produced at each price. Similarly, if most
consumers do not have strong preferences for one manufacturer’s product, we can
draw a market demand curve for chocolate bars. Then, looking back at Figure 13a,
we can see that it is the equilibrium price for chocolate bars at B that determines the
decisions of the producer of Choccos.
UNIT 8 | COMPETITIVE GOODS MARKETS
DISCUSS 7: RESTAURANTS IN CHINA
This market research report describes the “full-service” restaurant industry in China
(LINK). With the help of this information, discuss whether you would you expect
restaurants to produce at a point where their marginal cost is close to the price they
receive for a meal.
8.9 CONCLUSION
looking back over Units 7 and 8 we now have two different models of how
firms behave. In the Unit 7 model the firm produces a product that is different from
the products of other firms, giving it market power—the power to set its own price.
This model applies to the case of a monopolist, who has no competitors; common
examples are water supply companies, and national airlines with exclusive rights
granted by the government to operate domestic flights.
The price-setting model also applies to firm producing differentiated products such
as breakfast cereals, cars, or chocolate bars—similar, but not identical, to those
of their competitors. In this case the firm still has the power to set its own price,
although if it has close competitors demand will be quite elastic and the range of
feasible prices narrow.
In the model of a perfectly competitive market, developed in this unit, firms are
price-takers. Competition from other firms producing identical products means
that they have no power to set their own prices. We have seen that this model can be
useful as an approximate description of a market in which there are firms producing
very similar products, even if they are not identical.
In practice economies are a mixture of more competitive markets, and less
competitive ones in which firms have more power to set prices—monopoly power.
But in some respects firms act the same whether they are the single seller of a good
or one of a great many competitors. Most important among their similarities, all
firms decide how much to produce, which technologies to use, how many people to
hire, and how much to pay them so as to maximise their profits.
27
28
coreecon | Curriculum Open-access Resources in Economics But there are important differences. Look back at the decisions made by price-setting
firms to maximise profits (Figure 15 in Unit 7). Firms in competitive markets lack
either the incentive or the opportunity to do many of these things.
A firm with a unique product will advertise (Buy Nike!) to shift the demand curve
for its product to the right. But why would a single competitive firm advertise
(Drink milk!)? This would shift the demand curve for all of the firms in the industry.
Advertising in a competitive market is a public good: the firm pays the cost and the
benefits go to all of the firms in the industry. If you see a message like “Drink milk!”
it is probably paid for by an association of dairies, not by a particular one.
The same is true of expenditures to influence public policy. If a monopolist is
successful, for example, in relaxing environmental regulations, then it will benefit.
But lobbying, contributing money to electoral campaigns and other expenses of this
type will be unattractive to the competitive firm because the result (a more profitfriendly policy) is a public good.
Similarly, investment in developing new technologies is likely to be undertaken
by monopolies because if they are successful in finding a profitable innovation,
the benefits will not be shared with the other firms in the industry. If the firm is a
monopoly the extra profits they will make by a successful innovation are less likely
to be competed away by those who copy the innovator. However, successful large
firms can emerge by breaking away from the competition and innovating with a new
product. The UK’s largest organic dairy, Yeo Valley, was once an ordinary farm selling
milk like thousands of others. In 1994 it established an organic brand, creating new
products for which it could charge premium prices. With the help of imaginative
marketing campaigns it has grown into a company with 1,400 employees and 65% of
the UK organic market.
Figure 14 summarises the differences between price-setting and competitive firms:
UNIT 8 | COMPETITIVE GOODS MARKETS
PRICE-SETTING FIRM OR MONOPOLY
FIRM IN A PERFECTLY COMPETITIVE
MARKET
Sets price and quantity to maximise
profits (“price-maker”).
Takes market determined price as
given and chooses quantity to
maximise profits (“price-taker”).
Chooses an output level at which
marginal cost is less than price.
Chooses an output level at which
marginal cost equals price.
Deadweight losses (Pareto inefficient).
No deadweight losses for consumers
and firms (can be Pareto efficient if
no-one else in the economy is affected).
Owners receive economic rents (profits
greater than normal profits).
If the owners receive any economic
rents, the rents are likely to disappear
as more firms enter the market.
Firms advertise their unique product.
Little advertising: it costs the firm, but
benefits all firms (it’s a public good).
Firms may spend money to influence
elections, legislation and regulation.
Little expenditure by individual firms
on this (same as advertising).
Firms invest in research and
innovation; seek to prevent copying.
Little incentive for innovation; others
would copy (unless the firm can succeed
in differentiating its product and
escaping from the competitive market).
Figure 14. Price-setting and competitive firms.
29
30
coreecon | Curriculum Open-access Resources in Economics Unit 8 Key Points
1. In a market with many buyers and sellers, individuals and firms have little influence
on prices, due to competition. They are called price-takers.
2. When a perfectly competitive market is in equilibrium, all trade takes place at the
equilibrium price, and the market clears.
3. A perfectly competitive market would exploit all possible gains from trade,
eliminating all deadweight losses to consumers and firms. Most markets for real goods
don’t conform exactly to the model of perfect competition, but it can be a useful
approximation.
4. Firms in competitive and single firm markets alike seek to maximise profits but the
former are restricted in the ways they can pursue this objective.
UNIT 8 | COMPETITIVE GOODS MARKETS
UNIT 8: READ MORE
INTRODUCTION
• The company of strangers
Use this link you to download Who’s in Charge?, Chapter 1 of Paul Seabright’s book on
how market economies manage to organise complex trades among strangers: LINK.
Seabright, P. 2010. The company of strangers: A natural history of economic life. Princeton
University Press.
8.3 PERFECT COMPETITION
• Economic experiments in the laboratory and the field
You can read more about how laboratory experiments are used in economics in this
article by Vernon Smith: LINK.
Smith, V. 1994. Economics in the Laboratory. Journal of Economic Perspectives 8(1), pp. 113131.
John List tested the model of perfect completion in a field experiment: LINK.
List, J. 2004. Testing Neoclassical Competitive Theory in Multilateral Decentralized
Markets. Journal of Political Economy, vol. 112(5), pp. 1131-1156.
8.6 GAINS FROM TRADE, ALLOCATION AND DISTRIBUTION
• The deadweight loss of Christmas
Joel Waldfogel suggests that gift-giving at Christmas may result in a deadweight loss:
LINK.
The Economist. 2001. Is Santa a deadweight loss? 20 December.
• Is more competition always good?
Antitrust lawyer Maurice Stucke questions whether it is always desirable to increase
competition: LINK.
OUP blog. 2013. Is competition always good? 25 March.
31
32
coreecon | Curriculum Open-access Resources in Economics 8.7 TESTING FOR PERFECT COMPETITION
• Lessons about markets from the internet
Glenn Ellison and Sara Fisher Ellison describe what economists have learned from
the Internet about markets and market competition: LINK.
Ellison, G. and Ellison, S. F. 2005. Lessons About Markets from the Internet, Journal of
Economic Perspectives, 19(2), pp. 139-148.
• The Fulton Fish Market
Katy Graddy describes how competition in the Fulton Fish Market works: LINK.
Graddy, K. 2006. The Fulton Fish Market, Journal of Economic Perspectives, 20(2), pp. 207220.
This work is licensed under the Creative Commons Attribution-NonCommercialNoDerivatives 4.0 International License. To view a copy of this license, visit http://
creativecommons.org/licenses/by-nc-nd/4.0/ or send a letter to Creative Commons,
444 Castro Street, Suite 900, Mountain View, California, 94041, USA.