Less attention has been given to the question of whether there are substantial cost
benefits to operating stations across different markets, either because they can share pro-
gramming or some aspects of management (e.g., programming or news directors). These
benefits would be interesting to understand given the growth of large, national radio
companies described in the introduction and the emergence of firms that specialize in
particular formats, such as Radio One in the Urban format or Univision Radio in
Spanish-language formats.
8.6. EXCESS ENTRY
While most of the empirical literature has focused on understanding the effects of con-
solidation, a separate literature has asked whether, from a social welfare perspective, there
are simply too many stations or too many stations of a particular type. As well as the fact
that there are many stations in most radio markets, two features of the radio industry lead
one to expect that “excess entry,” in the sense of
Mankiw and Whinston (1986), is likely:
radio listenership is relatively inelastic, so new stations typically gain audience at the
expense of existing ones, and many station costs are fixed, in the sense of not varying
directly with the number of listeners served.
The first papers in this literature,
Borenstein (1986) and Rogers and Woodbury
(1996)
, used limited data on station listening in local radio markets to show strong evi-
dence of business (audience) stealing in the listener market both at the aggregate level and
within specific programming formats.
Berry and Waldfogel (1999a) estimated an endogenous entry model, combined with a
model of post-entry competition where stations compete symmetrically for listeners and
advertising prices decline in the number of listeners, using a cross-section of data from
1993. Stations are treated as being independent, which is also a reasonable simplifying
assumption given that the data come from before the Telecommunications Act (although
after ownership rules began to be relaxed). They ignore the value of product variety to
listeners because, without prices, they cannot estimate listener welfare in dollars. Under
their assumptions and ignoring integer constraints, excess entry is implied, so the interest
in their results comes from the fact that they estimate that the degree of excess entry is really
large. For example, they estimate that welfare would have been maximized by reducing the
number of stations in the San Diego market from 31 to 9, and in Jackson, MS from 17 to 3.
Two limitations of this analysis are that it is assumed that, having entered, all stations
have the same quality and are symmetrically horizontally differentiated, and that all stations
pay the same fixed costs from entering. In practice, this is not an appropriate assumption
because while stations may pay similar prices for licenses (of a given power and signal cov-
erage) and transmission equipment, they may then invest very different amounts in pro-
gramming quality. A better model of the industry is therefore one in which fixed costs are
largely endogenous and determined in equilibrium. A recent paper by
Berry et al. (2013)
375Radio
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.