Another area that receives little exploration is the role of advertising. Although the book discusses the radio industry’s early opposition and later acceptance of advertising as a primary source of revenue (pp 74–77), it does not analyze the structural implications of advertising support or the role of advertisers as rivals for control. For example, radio networks often sold blocks of time to advertisers, who then hired advertising agencies to produce programs for these slots.146 This in effect forced networks to surrender control over their own schedules to advertisers, a fact that gave sponsors tremendous control over industry behavior. Advertisers thus represented important industry players who often served as important counterweights to Sarnoff.
Reliance on advertising support has several other structural effects.147 For example, it introduces an intermediary into the relationship between programmers and viewers. As a result, programming is likely to be influenced more by programs’ impact on consumers’ willingness to buy advertised products than by audiences’ desire to see particular programming.148
Furthermore, advertising support limits consumers’ ability to signal the intensity of their preferences in much the same way as voting regimes. As an initial matter, advertising revenue provides only an indirect signal of the value that listeners and viewers place on the underlying programs.149 Moreover, unlike pricing regimes, in which audiences can signal particularly strong preferences by paying more for programming, advertising support gives consumers only one way to signal the intensity of their preferences: viewing versus nonviewing.150 As a result, advertising responsiveness is generally regarded as understating the value that audiences place on those programs.151 The result is a reduction in the resources invested in program quality. At the same time, reliance on advertising support reduces the diversity of programming by increasing the breakeven audience size that programming needs to survive.152
These insights undercut the sharp distinction that Wu attempts to draw between these historical episodes’ impact on industry economics and their impact on the nature and quality of the content being created (pp 97, 303–05). By affecting the economics, these industry practices directly affect the quality and diversity of content being conveyed. In short, these practices and the quality and quantity of speech are inexorably linked.
Consider block booking, which Wu notes may be economically beneficial (p 96). (This debate remains ongoing, particularly in the modern context of allowing cable subscribers to select channels á la carte.) Beyond the works that Wu cites, there is a long scholarly tradition showing how bundling content from the same provider can promote economic welfare,153 either by allowing excess consumer surplus enjoyed by one consumer with respect to one product to fund any shortfalls in another product154 or by discouraging a single firm providing many channels from using its additional channels to offer content that simply cannibalizes audiences from offerings already on the air.155 Wu nonetheless claims that despite these potential economic efficiencies, block booking remains problematic because of its adverse effect on the nature of the content being produced. This argument disregards the fact that improving economic efficiency can also have a positive effect on the quantity, quality, and diversity of programming. Increasing program producers’ ability to appropriate surplus makes it easier for new films and programs to cover their costs. This favors special interest programs by enabling them to survive despite the fact that they draw relatively small audiences.156 This insight draws support from the fact that a diverse range of cable networks, including C-SPAN, Discovery, and a number of networks targeted toward African Americans, all opposed regulatory efforts to unbundle cable television channels.157
Wu also repeats the often-advanced claim that information industries are more horizontally concentrated than in the 1950s (p 255–56). This claim ignores the broader literature suggesting that this is not true empirically.158 Even more importantly, a rich theoretical literature exists showing that the relationship between horizontal concentration and program diversity is ambiguous.159 Some empirical studies have indicated that increases in horizontal concentration may actually improve program diversity,160 although the FCC found the empirical support too ambiguous to support a strong policy inference in either direction.161 Again, in addition to affecting economic welfare, structural features of the underlying industries have implications for the nature of the content being created.162 The ambiguity of the empirical record does not support attributing any simple relationship between structure and conduct.
But Wu reserves his harshest criticism for vertical integration, which he claims reduces content diversity and innovation (pp 130, 147, 295, 305–06, 311) and presumes that an open, vertically disintegrated structure will yield better content (pp 35–39, 72–73, 139–47, 297). At other times, however, the book concedes that vertical integration may actually benefit consumers (p 84, 162, 284 n *, 305, 306), epitomized by the seamless and high-quality end-user experience offered by Apple (pp 278, 291–92). Indeed, a vibrant theoretical literature exists identifying ways that vertical integration yields efficiencies.163 An empirical literature is emerging that explores these conclusions.164 Yet Wu concludes that the “Separations Principle” requires that these benefits be sacrificed (p 305). In the process, Wu also stops short of undertaking any detailed analysis of the literature exploring the impact that prohibiting these practices would have on the nature of media programming. The FCC’s experience with how restrictions on vertical integration in television broadcasting—such as the financial interest and syndication rules (finsyn) and the Prime Time Access Rule (PTAR)—actually restricted program diversity provide ample reason for caution.165
These theoretical and empirical debates are rich and hotly contested; resolving them far exceeds the scope of this Review. Engaging the literature that explores how structural features affect media content would offer an account that may be less straightforward, but would provide greater insights into the dynamics of innovation and technological change as well as provide some insight into if and when the balance might tip in the other direction. The measure of any media policy ultimately depends on the nature of the content that the public receives. One would thus expect an assessment of these impacts to be part of his proposal. Although Wu may well be right that the balance tips in favor of openness and vertical disintegration, without a clearer explanation of how to make the relevant tradeoffs, readers are left without a clear idea of why that is the case or the circumstances under which the balance might change (p 305).