Walfogel-style instruments, interacted with exogenous rival characteristics (BLP instru-
ments). In her case, she uses consumer demographics in the other counties served by a
rival newspaper. For example, if a paper sold in a sparsely populated suburban county is
headquartered in a populous central metropolitan county, it is likely to have high levels of
quality and variety. The headquarters county is taken as exogenous so that the demo-
graphic levels in that county are available as instruments that shift the choice set in
the suburban county. More specifically, the population demographics of “other
counties” where the newspaper is sold are excluded from the mean utility of the paper
in a given county. Therefore, “other county demographics” are available as instruments
for endogenous characteristics and prices in the “own county” mean utility equation. In
Fan (2013) , the Waldfogel-style instruments include other counties’ education level,
median income, median age, and urbanization.
Fan (2013) considers extending the model of (3.9) to include random coefficients on
some of the characteristics. As noted, random coefficients allow for richer substitution
patterns than simple logit models. For example, if some consumers have a larger than
average taste for local news, they will likely substitute from one locally focused paper
to another one, whereas the pure logit generates substitution patterns that vary only with
demographic-specific market shares. It is not clear whether Fan has adequate instruments
for this purpose, and so she models only one random coefficient (on local news). Once
again, we consider Fan’s supply side below.
Continuing with a discussion of random coefficients,
Sweeting (2013) and Jeziorski
(2014)
consider listener demand models for radio that include random coefficients. In
addition to demographically varying tastes for formats, both Sweeting and Jeziorski have
normally distributed random coefficients on format, which in practice is very similar to
the nested logit model. Though the model with normal coefficients lacks a closed form
solution, it also lacks any a priori constraints on the value of the substitution parameters
(which in the nested logit must fall between zero and one). Jeziorski also adds a random
coefficient on the quantity (in minutes per unit time) of advertising, which introduces a
new dimension of substitution beyond the nested logit. Both of these authors consider
dynamic models on the supply side, focusing either on joint ownership (Jeziorksi) or
format choice (Sweeting).
Both Sweeting and Jeziorski want to model market structure endogenously, and so do
not want to assume that the numbers and formats of competing products are uncorrelated
with the demand unobservable ξ
jt
and so available as potential instruments. Instead, both
use a combination of panel data and IV assumptions. For example, Jeziorski uses a “quasi-
difference” in the unobserved product characteristic for station j at time t:
ξ
jt
¼ρξ
j,t1
+ ν
jt
(3.10)
and then assumes that the innovation ν
jt
is uncorrelated with past market and product
characteristics. These are the BLP instruments, but with the exogeneity (uncorrelated)
103Empirical Modeling for Economics of the Media
assumption defined relative to the innovation ν
jt
. This has the disadvantage of relying
heavily on a specific panel data “timing” assumption, but it makes up for a lack of instru-
ments that are uncorrelated with the undifferenced ξ
jt
.
Another source of data variation is repeated choice or (similarly) ranked choices,
including second-choice data (
Berry et al., 2004). If consumers give similar ranks to prod-
ucts that are similar in some product dimension, that is evidence of close substitution in
that dimension. This greatly aids in the estimation of substitution parameters. In many
(but not all) models of repeated choice, a tendency to repeatedly choose similar products
is also evidence of tight substitution. Clearly, some notion of diminishing marginal utility
or a taste for variety could possibly overturn that result, so a good model might allow for
both persistent taste (that generates repeated similar choices) and for some taste for vari-
ety. One common model has persistent (over time) random coefficients on product char-
acteristics, but product-specific E
ijt
that are independent over time. If consumers make
similar choices (in some product dimension) over time, then the random coefficients will
be found to be important and the model will imply tight substitution in that dimension.
The independent E
ijt
allows for a variety of choices over time and implies substitution
patterns that are not driven by product characteristics (although nothing in the model
accounts for actual diversity-seeking behavior.)
Goettler and Shachar (2001) carry this idea further in the context of television view-
ing. They seek to uncover the structure of product characteristics from the repeated
choices of television viewers. The approach is similar to the discrete-choice analysis in
political science that identifies dimensions of political differences between politicians
in Congress. Goettler and Shachar also allow for switching costs (the utility cost of chang-
ing the channel) as an apparently important alternative source of persistent choices. They
find persistent preference for shows along four dimensions. The first two relate to com-
plexity of plot and the “realism” of the plot (crime dramas vs. situation comedies, for
example). The last two dimensions map into similarities between the viewer demo-
graphics and the demographic pitch of the show.
Continuing with studies of television viewing,
Crawford and Yurukoglu (2012) use
data not just on channels watched, but also data on the time spent watching each channel.
This adds additional data on the intensity of preference and does allow the estimation of
diminishing marginal returns to watching a given channel.
3
This is a true taste for variety.
The paper applies its demand analysis to a study of cable television channel bundling.
Note that many cable channels are almost always bundled together (for example, in
an “expanded basic” channel tier) so without data on time spent watching each channel
it would be difficult to empirically distinguish the taste for individual channels within
3
The use of discrete choice plus intensity of use data goes back at least to the Dubin and McFadden (1984)
study of residential electric appliance demand.
104
Handbook of Media Economics
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.