« iepriekšējāTurpināt »
4.2 THE QUESTION OF ENFORCEMENT
There are situations involving copyright that concern the fundamental issue of what Arrow referred to as "exclusion." At the present time, some of these situations are occurring because of the availability of the technologies of high-speed photocopying and of copying digitized information by computer.
Persons with easy access to machines employing these technologies can become low-cost publishers, legalities aside. Thus, these persons are not easily "excluded" from ownership of copies upon their failure to pay a royalty. The question of enforcement then arises, and the cost of enforcement must become an issue. Concern with efficient allocation of resources as well as the deleterious effects of easy evasion of law must prompt the question of whether there is any value in issuing copyrights that cannot be enforced with any reasonable allocation of effort.
Hurt and Schuchman have theorized about strategies an original book publisher might employ in the absence of any copyright at all.25 According to one scenario, the original publisher must produce enough books in his first edition to saturate the market. If a copying publisher enters the market (probably with a similar number of copies), the first publisher must be prepared to compete by lowering his prices. Many unsold books can be expected in this situation. A second strategy is for the first publisher to be prepared with an extremely low-cost edition as a retaliatory measure.
Similarly, in a 1970 article in the Harvard Law Review opposing copyright protection for computer programs at that time, Professor Stephen Breyer proposed a strategy that could be employed by program developers in the absence of such protection. 26
"One may wonder, for example, whether, without protection,
Professor Breyer wrote. 27
Although Professor Breyer did not extend his scenario, it is possible to theorize about protective behaviors available to the originators of computer programs to protect themselves in such a hypothetical situation. One such strategy could be for an originator to produce programs for sale in object code only, with minimum documentation, thereby making it very difficult for a potential copier to know exactly what he had in hand. In fact a proposal for "sealed-in software" that might be protectable by either trade secret or copyright has been made recently by Calvin Mooers. 28
4.2.1 Transaction Costs Even If No Copyright
A conclusion that can be drawn from both these examples is that there
are transaction costs regardless of whether the imperfect protection of law exists or does not exist. To repeat from Arrow, "transaction costs ....are attached to any market and indeed to any mode of resource allocation." In the Hurt and Schuchman example, among the transaction costs that might be expected are the extra books left over, the poor quality of merchandise required to prevent financial losses, the extra secrecy required to prevent future plans and the first copies from being prematurely revealed, and the extra efforts that would be needed in merchandizing strategems to thwart a competitor's sales outlet possibilities. In the Breyer example, assuming the protective strategy of object code dissemination only with minimal documentation, among the transaction costs to be expected are the reduction in information dissemination about program content to everyone including disinterested observers who might benefit in another context, the reduction in ability to recognize mistakes in programs and to correct them, and the lowering of incentives to produce new programs that are genuinely novel or original.
Thus, in both examples which assume no Government copyright protection, we have postulated that cut-throat competition, losses in information flow and increases in secrecy would result. In a society in which the market protection of copyright is available, Government regulation has its cost and some infringement from imperfect exclusion can be expected to result, but we suggest that in addition, a more open society with greater opportunities for creativity exists. Thus, the choice is not just between the size of transaction costs inherent in the alternatives, but in the kinds of costs and their effects which a society is willing to tolerate.
4.2.2 The Optimal Level of Enforcement and Its Consequences
Hopefully, a society will select that set of resource allocation mechanisms that maximizes its satisfactions. However, a difficult state of affairs for a society to accept is that it cannot achieve the complete maximization of its satisfactions with any set of mechanisms because of the limited resources it can apply. A reasonable strategy is to achieve an optimum level of satisfaction from resources available, permitting a certain amount of dissatisfaction to remain. Professor Edwin Mansfield has demonstrated that there is an optimum level of crime whose cost ought to be tolerated, based on the finite resources of enforcement which a society is willing to allocate.29 This concept can be easily adapted to copyright infringement.
As shown in Fig. 1, the probability of apprehension and conviction of infringers increases with increasing expenditure of resources devoted to enforcement; but the costs to society of infringements increase as fewer resources are devoted to enforcement and the probability of conviction goes down. A minimum total cost results from the sum of infringement and enforcement costs, at a particular probability less than 1.0 of apprehension and conviction. This leaves some infringers unapprehended or unconvicted.
If a society is unhappy with this level of infringement, it can raise the resources allocated to enforcement. However, it might take unrealistically large resources to guarantee conviction of all infringers. On the other hand, abolishment of enforcement on the grounds of its ineffectiveness and the consequent large increase in what was formerly called crime might create new, unanticipated kinds of dissatisfactions which society is unprepared to accept.
Under the assumption that the benefits to a society of providing copyright protection and enforcement outweigh the costs, a question that arises is how the market for intellectual property should be structured to minimize transaction costs and to promote efficient pricing. The transaction costs considered here are Arrow's "costs of communication and information." A situation requiring special consideration for reduction of transaction costs is that which exists when there are a large number of users and a large number of producers. In this case, one of a number of different licensing schemes may be most effective.
4.3.1 A Comparison of Types of Licenses
Clearinghouse licensing and direct licensing are examples of licensing types that may be employed. With either of these situations, there is the possibility of a blanket license or a per-use license.
A clearinghouse is simply a multi-producer organization established for royalty collection. The advantage of a clearinghouse over direct licensing is that the user has a single point of negotiation, a single place to send royalty payments; and there is likely to be a reduction in the number of payments having to be made. The producer similarly has a reduction in transaction costs because he obtains his royalties from one place and with one payment. On the other hand, with a clearinghouse, there may be a blurring of individual producer considerations. The necessity of simple, all-encompassing contractual provisions may cause some producers with special situations to obtain less (or more) royalties than they would have if they negotiated individually. For each producer, the gain from the economy of scale of the clearinghouse would need to be traded-off against this loss of individuality.
Similar problems must be considered in the selection of the per-use or blanket license. With a per-use license, the major cost is collecting the information. This may be technologically dependent. For example, with uses that are associated with a computer, the capability of collecting use-related data may be high, particularly if it is the producer's computer that is being used and if "use," as opposed to memoryresidence, is easily defined. On the other hand, for mechanical photocopying, the collection of use-related data may be difficult, particularly data which might distinguish the various works being copied.
With blanket licensing (a single yearly fee for all use), the amount
of data needed to be collected is reduced. If the blanket license is in reality a substitute for a per-use license, simply because the cost of collecting per-use data is too high, then the reduction in data collection costs must be traded-off against the increase in inaccuracy and inequity in royalty collections and royalty distributions. Some reduction in inaccuracy may result from dividing users into classes dependent on expected use; and by sampling uses.
Appendix B presents some data from the British Lending Library (simply as an illustrative example) demonstrating that photocopying there is heavily skewed in terms of the frequency of photocopying from various journal titles. A survey indicates that of approximately 15,000 serial titles held by the British Lending Library, the top 200 titles accounted for 20% of the photocopying demand and the 6000 least-requested titles accounted for the last 10% of the demand. U.S. data will likely show a similar skewness.
As noted in Appendix B, this skewness can lead either to lower or higher payments to individual copyright proprietors, depending on the payment algorithm employed. In addition, for those journal titles little used, a larger amount of sampling conceivably coupled with more sophisticated sampling methods might be needed to accurately determine the true extent of photocopying.
At a time a new licensing scheme is to be established, producers may find it important to consider these various trade-offs so that the mechanism with the lowest transaction costs can be adopted. From the user's viewpoint, transaction costs include the value of time and effort as well as the dollar amount of royalties. That mechanism that is easiest to use, i.e. least costly in time and effort, all other things being equal, will probably generate the least amount of deliberate evasions and therefore the lowest enforcement costs as well.
4.3.2 Examples of Existing Clearinghouses
The Harry Fox Office is the mechanism through which many of the music publishers have issued licenses for the recording of individual compositions on phonorecords. (See Appendix A, Section A.4.6.3). Despite the availability, since the passage of the 1909 Act, of a compulsory license with the Copyright Office serving as a repository of ownership information30, licensees may find that better terms are available from the Harry Fox Office in return for greater assurance of precise information about numbers of records manufactured and delivered. Royalties owed are computed from this information.
Three clearinghouses now exist for the collecting royalty payments for public performances of musical works. These are the American Society of Composers, Authors and Publishers, Inc. (ASCAP), Broadcast Music, Inc. (BMI) and SESAC, Inc. The combined membership of ASCAP, BMI and SESAC comprise the copyright owners of virtually all music copyrighted in the United States. Licensees are required to pay only a lump-sum royalty