Lester G. Telser

Advertising remains a challenging problem for economic theory for
several reasons. First, the ordinary assumption that tastes are given
does not serve us well for advertised products. We are forced to pay
attention to the effects of what people know about products that they
buy. The simplifying assumption of elementary economic theory focuses
attention on price alone for commodities of known virtues or vices. To
understand advertising we must reckon with the fact that a person's
stock of knowledge about goods and services influences his preferences.
It is, therefore, useful to assume given tastes only if the stock of knowl-
edge is given. Under these circumstances, when the stock of knowledge
changes, the theory that takes this into account will furnish better pre-
dictions of behavior.

Second, although advertising is a joint product that goes together
with some physical good or service, it is not literally tied to the good or
service. To illustrate, buttons and coats are normally tied together to
make a joint product. Since different coats are made with different
numbers of buttons, one can calculate a derived demand for buttons and
a derived demand for coats separately. In these respects alone there is
no substantial difference between advertising and coat buttons. The
distinction lies in this. When someone buys a coat he pays for both the
coat and the buttons. With advertising this is not generally true. There
are people who may receive the benefits of advertising messages without
facing the burden of paying for the advertising. There are also people
who pay for the advertising when they buy the advertised good or
service who do not necessarily benefit from the advertising because
whatever information it contains is redundant to these buyers. This,
however, begs the question of why they do not buy some equivalent
good or service that is not advertised at all or is less advertised and in

-71-