Facts and friction
Many people believe that the Internet will make
middlemen extinct. It is more likely to help them
Middlemen have never been popular, except in economic
theory. The theory says, pretty incontestably, that by
intermediating between buyers and sellers, middlemen lower
the costs of transactions and make most goods and services
cheaper. And yet many consumers still begrudge travel and
property agents their "unearned" cut. Hence the cheerful
reception given to the claim of a growing band of amateur
economists that middlemen are about to be tossed into the
dustbin of history. Their exterminator? The ubiquitous
The Net's fans argue that its expanding influence -- along
with the growth of other forms of cheap communication --
will eventually allow consumers to bypass intermediaries
while increasing their choices and finding better bargains.
One influential proponent of this idea is Bill Gates. In
his recent book, "The Road Ahead" (Viking Penguin, 1995),
the founder of Microsoft talks about a "friction-free
capitalism", in which a modem in every living room will
enable consumers to locate and buy whatever they want with
no need of middlemen.
At first glance, such prophecies can seem quite plausible.
Intermediaries, after all, perform two main services:
providing information and distributing goods and services.
Both functions are likely to be changed by the Internet.
Many intermediaries earn their keep by collecting
information about products and their prices and thus
helping consumers to get a better deal than they could by
searching unaided. For example, travel agents specialise in
finding the most convenient flights, car rentals and hotels
at the best possible prices. They use this knowledge to
book better trips for their clients, charging a premium for
the service. Estate agents (realtors) also profit from
information. They help buyers and sellers find each other,
and charge a commission for doing so.
With the Internet, buyers of holidays and houses can gain
cheap, direct access to this information without having to
pay money to a middleman. It provides a place where
potential sellers can list what they have on offer, and
potential buyers can learn easily about bargains, locate
new products and receive advice about everything from legal
problems to new medical treatments. For those middlemen
whose edge is in information, the outlook may look ominous.
Less obviously, many of the Internet's boosters argue that
it will also do away with intermediaries' other main
function: distribution. Distribution networks exist mainly
to provide a cheap and convenient way to shop. By
co-ordinating transport and buying in bulk, retailers lower
However, once enough people have learned to let their
modems do the walking, they should be able to shop without
visiting the store. As this happens, customers will no
longer need retailers to carry huge stocks for them --
although there will be some need for shops, so that people
can inspect goods for themselves. Distribution of
everything from jeans to pharmaceuticals will then go
straight from the factory to the customer's home. And
informational goods -- from videos to newspapers -- can be
delivered by the Internet itself. Once again, the
implication is that middlemen will disappear.
The middle frontier
Compelling as they seem, however, these arguments do not
prove their case. They overlook two forces pushing in the
other direction: falling barriers to entry and rising
wages. Both point towards a bright future for middlemen.
Start with entry barriers. By slashing the costs of
communicating, the Internet will indeed make it harder for
a privileged few to hoard information. That, however, will
provide opportunities not just for consumers, but for
potential new middlemen as well. The result, points out Hal
Varian, an economist at the University of California at
Berkeley, will be greater competition among intermediaries,
and maybe many more of them.
Mr Varian uses travel agents as an example. He agrees that
as the Internet's reach continues to expand it will become
much easier for people to learn where the cheapest
aeroplane tickets are. As a result, those who prosper at
present by knowing how to navigate around special
electronic booking systems will lose out. Mr Varian points
out, however, that without the need to invest in those
special skills and technologies, it will become far easier
for anyone who is interested to become a travel agent.
Instead of competing on the basis of inside information,
those agents will compete by adding value in other ways.
Some may excel at combining a series of flights, hotels and
activities into a creative holiday package. Others may be
better at collecting information about the beauty (or
ugliness) of resorts and conveying it to customers. Thus,
intermediaries in travel and other sectors will still be
able to earn a profit by gathering and selling information
that people value. And the Net will help them.
Moreover, if the Internet does succeed in making
transactions tar more efficient, middlemen will benefit in
another way. This is because the fall in the cost of
gathering and transmitting information will boost
productivity in the economy as a whole, pushing wages up
and thus making people's time increasingly valuable.
Economic theory suggests that people will respond to this
in the same way that they have for centuries: by asking
middlemen to do more things for them.
For example, in recent years new intermediaries have
already arisen in many industries as people have got
richer. Food is a good example. Consumers can pay people to
deliver pizzas or buy their groceries. Ever more
specialised magazines and newsletters have proliferated to
filter people's news for them.
The Internet's admirers are right to say that it will put
some intermediaries out of business. But it will create
more work for others. And among these will be not only
suppliers of familiar services, but also a new breed
monitoring Internet sites, collecting news and information,
and repackaging it -- folk, in other words, like Mr Gates.
In the end, he and his competitors may be glad to see their
predictions miss the mark.