Broadcasting versus narrowcasting

Broadcasting versus narrowcasting
of 17
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Information Economics and Policy 10 (1998) 41–57 Broadcasting versus narrowcasting *Suchan Chae , Daniel Flores  Department of Economics -  MS  22,  Rice University ,  Houston ,  TX   77005  - 1892,  USA Abstract In this paper, we identify the characteristics of TV programs which make the programsmore likely to be broadcast than to be shown on pay channels. One characteristic we focuson is the degree of extensiveness of the market. An extensive (intensive, respectively)market is a broad but shallow (narrow but deep, respectively) one.We find that a program ismore likely to be broadcast as the market is more extensive. We also investigate which onebetween broadcasting and pay TV is the more desirable outlet for programs of differentcharacteristics from the welfare point of view. Key words :   Broadcasting; Narrowcasting; Cable TV; Pay TV  JEL Classification :   L82; L12; D42‘‘The last big fight to be shown on pay-per-view, Evander Holyfield versus BusterDouglas, also brought in over $35m in revenue – which is far more than networkscould ever hoped to raise from advertising’’ The Economist, March 23, 1991. 1. Introduction Some TV programs are shown on broadcast channels and others on paychannels. The main difference between the broadcast and pay channels lies in themethod of generating revenues. Broadcasters use advertising as the primary sourceof revenue while pay TV or ‘‘narrowcasters’’ rely primarily on payments made byviewers. Thus, if there are identifiable characteristics of programs which makeprograms more likely to be broadcast than to be shown on pay channels, their *Corresponding author. Tel.:  1 1 713 7375752; fax:  1 1 713 2855278; email:$19.00  󰂩  1998 Elsevier Science B.V. All rights reserved. PII   S0167-6245(97)00025-5  42  S  .  Chae ,  D .  Flores  /   Information Economics and Policy  10 (1998) 41 – 57  explanatory power must be traceable to this difference in the mode of generatingrevenues.From the 1950s to 1970s, some authors including Steiner (1952), Rothenberg(1962), Wiles (1963) and Beebe (1977) argued that a monopolistic broadcastertends to offer ‘‘lowest common denominator’’ programs based on models where 1 consumers’ preferences over alternative programs are lexicographic. As activeprogramming for cable television began in early 1970s, people immediatelyrecognized the potential for pay TV to offer a variety of programs. Noll et al.(1973), for instance, make the following observation without a formal analysis:‘‘In the present commercial system, television is geared to advertising revenuesand viewers influence programming only by their numbers. With a charge perprogram, both their numbers and the intensity of their desires could be reflected’’Spence and Owen (1977) compare the welfare generated by broadcasting andthat generated by pay TV in two different market structures: monopoly andmonopolistic competition. They conclude that both broadcasters and pay TVoperators have biases against ‘‘minority-taste’’ programs. From the methodologicalpoint of view, their innovation is to introduce an aggregate utility function which 2 depends on the consumption of all available programs. The cardinal approachallows the authors to use total surplus, or the sum of consumer surplus and profit,as a measure of welfare. Wildman and Owen (1985) extend the Spence-Owenmodel by allowing ad-supported and pay services to compete against each otherand by making a programmer choose the amount of time for commercials.Waterman (1992) analyzes a monopoly model with two types of viewers andtwo programs for which the value can be improved by investment, and reaches aconclusion similar to that of Spence and Owen, that is, both broadcasters and payTVoperators have biases against ‘‘narrow appeal’’ programs. But, whereas Spenceand Owen say that the bias is weaker with pay TV,Waterman says that the bias canbe greater with pay TV.Regarding the welfare implications of selecting either broadcasting or pay TV,Spence and Owen (1977) argue that the first best solution requires broadcastingbecause consumers pay zero price, which is the marginal cost of supplying the 3 program to a viewer, when the program is broadcast with advertising. Then in asecond-best framework they argue that welfare increases if the television industrymoves from a situation dominated by broadcasters supported by advertisement 1 Owen and Wildman (1992) Chapter 3 provides a good survey of the literature. 2 The aggregate utility function, however, is difficult to justify as is often the case. Authors’derivation of aggregate utility maximization from individual utility maximization only highlights thedifficulty: ‘‘Given a set of offerings, each viewer will select his preferred program . . . . We add up . . .dollar benefits for all viewers to arrive at a measure of the gross dollar benefits for all viewers . . . .When confronted with prices . . . for the . . . programs, viewers will react by allocating themselves toprograms so as to maximize the net benefits to them . . . ’’ 3 The argument is not entirely valid as explained in Section 4 of the current paper.  S  .  Chae ,  D .  Flores  /   Information Economics and Policy  10 (1998) 41 – 57   43 toward a new industry structure consisting of competitive pay TV operators.Holden (1993) argues by a numerical example that a consumer surplus generatedby a pay-per-view (PPV) operator can be lower than that generated by abroadcaster.In this paper, we identify the characteristics of programs which make theprograms more likely to be broadcast than to be shown on pay channels (that is, tobe ‘‘narrowcast’’) and investigate which one between broadcasting and pay TV isthe more desirable outlet for programs of different characteristics from the welfarepoint of view. One characteristic we focus on is the degree of extensiveness orintensiveness of the market. A market is said to be  extensive  if the audience size islarge but individual viewers’ willingness to pay is relatively low, and  intensive  if the audience size is small but individual viewers’ willingness to pay is relativelyhigh. We show in Section 3 that a broadcaster (narrowcaster, respectively)generates higher profits and welfare than a narrowcaster (broadcaster, respectively)if the market is extensive (intensive, respectively). The result contrasts in spiritwith Spence and Owen (1977) and Waterman (1992) who argue that bothbroadcasters and narrowcasters have biases against minority-taste programs.Our innovation consists in dissociating the broadness or narrowness of a marketfrom the economic size of the market. What the literature calls a ‘‘broad (narrow,respectively) appeal’’ program is in fact a large (small, respectively) audienceprogram. In contrast, what we call an extensive (intensive, respectively) market isa ‘‘broad but shallow’’ (‘‘narrow but deep’’, respectively) one. When we compareextensive and intensive markets, we control for the economic size of the market,which is the product of the breadth and depth. This conceptual innovation allowsus to associate broadcasting with the extensiveness of the market and narrowcast-ing with the intensiveness of the market. As the market becomes more extensive,advertising becomes a more attractive means of collecting revenue, while a moreintensive market makes pay TV more attractive.We also investigate the first best and second best solutions in Sections 4 and 5.We find that the first best solution is broadcasting with some amount of commercials whatever the characteristics of programs may be. The reason for this,however, is not that zero price associated with advertisement-supported broadcast-ing is efficient as argued by Spence and Owen (1977) but that broadcasting createseconomic value captured by advertising revenues. As for the second best solution,we find that the effects of the various parameters on the second best solution arequalitatively similar to those of the profit-maximizing solution.We begin with a model where a broadcaster cannot collect any direct paymentsfrom viewers and a pay-TV operator cannot advertise. In this case, the pay-TVoperator is in fact like a PPV operator. In Subsection 6.1, however, we willconsider a firm which can both advertise and collect payments from viewers. InSubsection 6.2, we will consider a variation of the basic model where the nuisanceeffect of advertising takes a form different from the one considered in previoussections.  44  S  .  Chae ,  D .  Flores  /   Information Economics and Policy  10 (1998) 41 – 57  2. Consumers Potential consumers are identified by points (represented by  x ) on the interval[0,  ` ). We will consider programs with different characteristics affecting consum-ers’ willingness to pay. For all programs we consider here, it is assumed that aconsumer with lower  x  has a greater willingness to pay. By way of easyintroduction, we first consider a situation without commercials. In this case, thereare two characteristics of a program which affect consumers’ willingness to pay:the quality of the program and the ‘‘extensiveness’’ of the market for the program,represented by parameters  q . 0 and  u  . 0, respectively. The willingness to pay of consumer  x  for a program with characteristics ( q ,  u  ) is q x ] ] max H S 1 2  D , 0 J  (1) u u  The willingness to pay is proportional to quality. The extensiveness parameteraffects the willingness to pay in two ways. First, it determines the size of theaudience by identifying the marginal consumer (i.e.,  x 5 u  ) who is just indifferentbetween viewing and not viewing the program. Second, it determines, for a givenquality  q , the willingness to pay  q  /  u   of the highest-value consumer (i.e.,  x 5 0).Notice that the maximum willingness to pay  q  /  u  , is inversely related toaudience size  u  . Thus the potential consumer surplus, which is a measure of theeconomic size of the market, remains constant at  q  /2 when one varies theparameter  u  . In Fig. 1, two willingness-to-pay curves, which also serve as demandcurves, corresponding to two different values  u   and  u   such that  u   , u   are drawn. 1 2 1 2 A program with a lower  u   has a smaller but more intensive viewer base, while aprogram with a higher  u   has a larger but less intensive audience. In this sense, theparameter  u   represents the degree of ‘‘extensiveness’’ of the market for the Fig. 1. Potential willingness to pay for intensive and extensive programs.  S  .  Chae ,  D .  Flores  /   Information Economics and Policy  10 (1998) 41 – 57   45 program rather than just the audience size. A low  u   corresponds to an ‘‘intensive’’market while a high  u   corresponds to an ‘‘extensive’’ market. Intuitively, anextensive market is a ‘‘broad but shallow’’ one, while an intensive market is a‘‘narrow but deep’’ one.Now we will introduce commercials. With commercials, the willingness to payof consumer  x  for a program with characteristics ( q ,  u  ) and  t   minutes of commercials will be assumed to be q x ] ] max H H 1 2 2 n ( t  ) J , 0 J . (2) u u  The function  n ( t  ) $ 0 represents the nuisance consumers suffer due to commercials.The full willingness-to-pay curve with commercials is drawn in Fig. 2.Demand for a program comes from consumers who are willing to pay at leastthe price of a program. Thus, for a given price  p  and the length of commercials  t  ,active viewers are those whose characteristics satisfy q x ] ] S DH 1 2 2 n ( t  ) J $  p , (3) u u  i.e., u   p ]  x # u   1 2 2 n ( t  )  ;  x (  p ,  t  ) (4) H J q 4 as shown in Fig. 2. The measure of active viewers (or the quantity demanded) is Fig. 2. Willingness to pay with commercials. 4 Here, of course, we are using the Lebesgue measure. Intuitively, potential consumers aredistributed uniformly on the interval [0,  ` ]. The resulting linear demand function allows us to findclosed form solutions of the model.
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks