Handbook of Media Economics, vol 1A

of W’s who listen to radio is independent of the number of B’s. This is not because the W’s

do not listen to b stations. Rather, it is because the extra b stations crowd out w ones at

exactly the rate that keeps overall W listening the same. As noted above, this is an artifact

of assuming valuation symmetry. If instead other-side valuations were less than s, then

there would be fewer W listeners as M

B

rose. The reason is the dominant crowd-out effect.

Put another way, negative cross preference externalities go hand-in-hand with decreasing

listener shares. Regardless, the prediction for own-side presence is that more listen: the

presence of more W’s increases the equilibrium fraction of W’s who listen.

1.4. EMPIRICAL RESULTS: FACTS RELEVANT TO PREDICTIONS ...