Let us continue with our example
of the pushcarts, but let us change assumptions. Let the
beach be very, very long, and let buyers face a cost in traveling to and from sellers. To make this second
assumption specific, assume that it costs buyers $1.00 to
travel a mile. Finally, assume that there is one seller on
the beach and that everyone values the product he sells at $10.00.
If the seller prices his product at $6.00, how many
people will buy it? He should get all those people who are
located less than two miles from him. To see this answer,
consider someone located one mile away. This person finds
that the product costs $8.00 to get because she must pay
$6.00 and travel two miles (one going and one returning).
Because $8.00 is less than the benefits of $10.00 she gets
from the product, she will buy. If she was located three
miles away, the product would now cost her $12.00 ($6.00 to
the seller and $6.00 of transport costs), so she would not
Now, suppose that another seller decides to locate on our
very long beach. Unlike the Hotelling beach, he will not
locate next to the original seller. The assumptions of
transport costs and long beach change this result. In fact,
the new seller should move down the beach until he has his
own four-mile stretch of beach.
Suppose that both sellers are doing a great business and
getting rich. Because it is very easy to roll another push
cart onto the beach, the unusually high returns in this
business should attract new sellers. Suppose that a new
seller locates exactly two miles away from our first seller.
How much of the beach will each control? If they both charge
the same price, then they will evenly split the beach
between them, or each will attract customers within one mile.
However, to make the story a bit more complicated,
suppose that the new seller raises his price to $7.00. If
the original seller still charges $6.00, then the dividing
line separating customers between them will be 1.25 miles
from the original seller and .75 miles from the
higher-priced seller. (If you check the cost of the buyer at
this point, you should see that it costs her $8.50 to go to either seller.)
In the previous example each seller has some control over
price. Each realizes that if he charges a higher price, he
loses some customers, but not all of them. In this way the
sellers are like monopolies: they are price searchers facing downward-sloping demand curves. However, because they also are in an industry that has easy entry and exit, there will be no long-term profits (as economists define profits) and in this this way resemble price takers in perfect competition, the sellers we discussed when we drew supply curves. Because it has both elements of monopoly and competition, economists classify an industry of this type as monopolistic competition.
The model of monopolistic competition puts our situation
into a more familiar form of demand and cost curves. The
illustration below shows a seller with a downward-sloping
demand curve and a conventional marginal cost curve. Because
the demand curve slopes downward, the marginal revenue curve
lies below it. The seller maximizes profit by selecting that
output at which marginal revenue equals marginal costs and
charges as much as he can, which is price P2. In the
long run, there can be no economic profit because there is
free entry into the industry. If there are any profits,
others will enter the industry, positioning themselves to
take away customers from the most profitable sellers. The
zero profit condition implies that at equilibrium, average
revenue (which is demand) must just equal average cost. When
average revenue equals average cost, average profit is zero,
and so total profit must also be zero.
The model of monopolistic competition was considered
important when it was introduced for two reasons. First, the
situation it described seemed the most common form of
industry. Both the single sellers of monopoly and the many
sellers of price-taking competition are uncommon in comparison. Furthermore, monopolistic competition describes more than traveling costs due to geographical distance. The distance between sellers can be in the minds of buyers.
Product differentiation, which results in many products that
are similar but not identical, also creates a distance
between products. It was this distance that seemed
especially important to the developers of monopolistic competition.1
Second, notice that because price exceeds marginal cost,
the graph contains a gray area of welfare loss, an
unexploited value that neither firms nor customers obtain.
Because monopolistic competition was seen as both common and
inefficient, it was argued that market systems were
However, the welfare loss in the case of monopolistic
competition may be illusionary. Firms could obtain this
value if they could price
discriminate, selling beyond q* up to qo.
If they do not do so, then the resources needed to obtain
the information required to price in this fashion must be
more valuable than the triangle of unexploited value. Or
consider possible government solutions. The problem is that
sellers are too small and charge too high a price. The
government could react by limiting the number of sellers and
forcing them to charge lower prices. But this policy would
increase traveling costs of buyers. The market may not give
the optimal number of sellers (and hence the optimal
distance between them), but the cost of gathering the
information that would let government decide the matter is
almost certainly greater than any possible welfare gain
(even ignoring the political
incentives that always go with government solutions).
Some economists have argued that the distance between
products is often phony, that firms differentiate products
to fool consumers. Their argument is surely correct in many
cases. In other cases, however, product differentiation
exists because it reflects differences in people's tastes.
Again, there is a cost to deciding whether in each case
product differentiation manipulates buyers or caters to
them. It is not clear that any government policy that tries
to eliminate "bad" differentiation will have benefits that exceed its costs.
When the cost of correcting a problem exceeds the
possible gain from the correction, is there any real welfare
loss by allowing the problem to continue? The model of
monopolistic competition shows that real market systems fall
short of theoretical constructs that assume away the
problems of information and of making agreements among
people. But any real-world system looks bad compared to
theory that assumes away problems.
Finally, the theory developed above can explain the decline of Main Street in small towns after World War II. When traveling costs are reduced, people become more price-sensitive, which means that the demand curve facing each seller becomes more elastic. As a result, the marginal revenue curve shifts upward and will intersect the marginal cost curve at a higher quantity (or greater distance). For individual firms to expand sales when the industry sales are constant requires some firms to disappear. Hence, reducing traveling costs reduces the number of firms if they are in monopolistically competitive industry. Development of the automobile and the highway system drastically cut traveling costs in the United States in the middle of the twentieth century. As a result, throughout America there are thousands of villages that have completely or largely lost their retail districts.
Next we leave transport costs and look at trade across political boundaries.
1 While the theory of monopolistic competition suggests that there may be too much product differentiation, Hotelling's model suggests that there may not be enough.
Copyright Robert Schenk