An Experimental Market for Public Goods: The PBS Station Program cooperative
研究了美国公共广播服务（PBS）在1974-75赛季开始的三年实验，通过市场机制（节目合作社）让约150家参与电视台用美元投票选择全国播出的节目，以检验分散化市场能否替代集中决策并影响节目类型。
Beginning with the 1974-75 season, the Public Broadcasting Service (PBS) undertook a three-year experiment to decentralize through a market process the selection of programs to be broadcast over the national noncommercial television network. The experimental market, called the Station Program Cooperative (SPC), enables each of the approximately 150 participating stations to vote with dollars for the programs they prefer, based on information about the subject matter, quality and costs of the alternatives. The SPC had two purposes. The first was to determine whether a decentralized market mechanism for acquiring national programming can be established that, first, has a reasonable cost and, second, retains the basic features of a network. A network is a means of centralizing the acquisition, transmission and promotion of programming. Although the advantages of networking are several, probably the most important is that it saves substantial transactions cost in comparison to the two other alternatives: syndication (in which each station negotiates for nationally distributed programs with the owner of the rights to the program) and localism (in which each station produces its own programming). (See Bruce Owen, Jack Beebe and Willard Manning.) The second purpose of the SPC is to determine its impact upon the types of programs that are broadcast. Centralized programming authorities such as the present commercial networks and the Corporation for Public Broadcasting, because they are few in number, are relatively easy targets for political pressure with respect to program content, may at least tacitly collude in programming decisions, and in any event, have limited experience concerning the nature of the demand for programs in all markets in the network system. (See Noll, Merton Peck and John McGowan, Ch. 8.) Because of the cost advantages of networking and the technical barriers to entry of new networks, in principle a network can produce an inefficient programming mix and still meet minimum standards of profitability or, in the case of the noncommercial sector, public acceptability. Concomitantly, station managers may be too unsophisticated about judging programming quality and sensing the structure of demand in the local market to select programming that, in benefit-cost terms, is than that selected by a centralized decision maker. While direct evidence on which system works better is not possible to acquire, two useful questions can be answered, at least in principle: (1) Do the SPC and centralized decision making lead to the selection of different programs, and if so, is that difference systemmatic? and (2) XVhich system produces higher audience * California Institute of Technology. Financial support for preparing this paper was provided by the National Science Foundation RANN program, grant #APR76-01920. We are indebted to Richard Beatty and PBS for providing us with remarkably complete data for all three SPCs. We are, of course, solely responsible for the contents of this paper.