Effectiveness of an advertising ban

the viewer prices and losing advertising market shares always dominate the increase in the advertising price as well as an eventual gain in viewer market shares. For the high quality broadcaster B, there are gains on the advertising market due to higher prices and a higher market share on the one hand, but potential losses on the viewer market on the other hand, as prices decrease and the viewer market shares may fall as well. If and only if the negative externality β is sufficiently small, the gains from advertisements outweigh the losses on the viewer market, andB’s profits increase inδ.

A similar reasoning applies to the profits ofB under the sym2-regime.

• decreases the overall amount of advertisements.

Proof. See Section A6 of the Appendix.

The effects of an advertising ban on all remaining equilibrium values are depicted in Table 3.4.

Table 3.4: Comparison across regimes (sym2 vs. asym)

Broadcaster A B

Viewer prices p^sym2_A Rp^asym_A p^sym2_B Rp^asym_B

Advertising prices τ_A^sym2 < τ_A^asym —

Viewer market shares n^v,sym2_A < n^v,asym_A n^v,sym2_B > n^v,asym_B Advertising market shares n^ad,sym2_A +n^ad,sym2_B > n^ad,asym_A —

Profits Π^sym2_A RΠ^asym_A Π^sym2_B >Π^asym_B

Note: This table illustrates the effects of broadcasterBnot being allowed to enter the advertising market by comparing the equilibria of the symmetric model with abstention (regimesym2) and the asymmetric model (regimeasym).

While the result concerning the reduction of the overall amount of advertising is in-tuitive, it may be surprising that medium B loses viewers by reducing its advertising level to zero. The latter result is due to the twofold nature of broadcasters’ incen-tives to attract viewers by low viewer prices: First, lowering viewer prices has a direct (positive) effect on viewer market shares. Second, there is an indirect (positive) effect on advertising market shares due to the positive externality the number of viewers exerts on the demand for advertising slots. Being prevented from advertising, the high quality medium B loses this second motive while the incentives of medium A remain unchanged. Hence, in the case of asymmetric advertising, the relative incentives to set low viewer prices are stronger for channel A than for channel B. Accordingly, in equilibrium, broadcaster A serves the larger part of the viewer market than B, which allows him to increase the advertising price.

Both broadcasters’ price setting behavior on the viewer market is subject to two oppos-ing effects. There is a negative effect onA’s viewer prices due to the ad-free platformB becomingceteris paribus more attractive to consumers, which leads to increased price competition that also puts downward pressure onB’s viewer prices. The positive effect onB’s viewer prices comes fromBlosing part of his incentives to attract viewers, which reduces price competition and thus enables both broadcasters to increase their prices.

Whether the positive or the negative effect dominates the evolution of viewer prices depends on the relative strength of the externalities between the two sides of the mar-ket. As shown in Figure 3.6, we can distinguish between three cases: If, for any given level of the viewer externality δ, the nuisance cost of advertisement β is sufficiently low (high), both channels set a higher (lower) viewer price in the case of asymmetric advertising compared to the case of symmetric advertising; for an intermediate range of β, channelA lowers its viewer price while channel B raises it.

Figure 3.6: Comparison of both regimes (sym2 vs. asym)

Note: This figure illustrates the effects of an advertising ban for broadcasterB on equilibrium viewer prices by comparing the equilibria of the symmetric model with market abstention (regimesym2) and the asymmetric model (regimeasym).

For broadcaster B, being prevented from entering the advertising market obviously is a disadvantage that lowers his profits. The analysis shows that for any given value ofδ, the advertising ban onB is beneficial for broadcasterA if and only if the nuisance cost βis sufficiently small. If the nuisanceβ is high, the intensified price competition on the viewer market dominates from broadcasterA’s perspective, and his profits decrease as well.

In our analysis, the levels of quality are exogenously fixed at maximum differentiation.

Though analytically hardly tractable, the framework at hand also allows for modeling an endogenous decision on quality levels. The results with exogenous levels of quality may already give a hint on how these levels would react to an advertising ban on broadcaster B, if the decision on quality was endogenous. As we have emphasized above, broadcasterB loses a part of his incentives to attract viewers. The choice of an endogenous quality level is, besides viewer prices, a second instrument for attracting viewers. Hence, one might expect quality levels to evolve analogically to viewer prices:

On the one hand, channel A has an incentive to raise its quality level in order to regain shares in the viewer market. This is due to the direct effect of no advertising at channel B making B ceteris paribus more attractive to viewers. An increase in A’s quality levels intensifies competition in the quality dimension and exerts upward pressure on channel B’s quality level. On the other hand, the indirect effect of not being obliged to please any customers on the advertising market lowers channel B’s incentive to attract viewers by offering high quality. This mitigates quality competition in the viewer market and gives room for decreasing quality levels, even to channelA.

Whether the direct or indirect effect dominates is again determined by the relative strength of the externalities between the two sides of the market.

3.6 Equilibria in free-to-air systems with and