سرقت تجاری و سیاست های مالکیت معنوی

I discuss the competition between a copyright owner and several commercial pirates who sell copies of the same information good to consumers. I view the increased risk of a punishment that offering a pirate copy to a consumer causes as an advertising cost whose value is chosen by the government. The structure of the market for pirate copies is affected also by fixed costs that are caused by punishments or DRM systems. I present a systematic analysis of the effects of these policy variables and the quality of pirate copies on the market for the considered information good.

The distribution of illegal pirate copies of information goods might have a variety of motives. Such copies are distributed on the one hand by the members of peer-to-peer networks, who deliver digital goods on the Internet without monetary compensation and who are motivated by, for example, a feeling of identification with the other network members, and on the other hand by commercial pirates who are, more conventionally from the perspective of the economist, motivated by the revenue that results from their activities. Somewhat less obviously, the consumers of an information good might also form a club, which buys a single copy of an information good, produces copies of it, and distributes one of them to each club member. 1 Given that both commercial and non-commercial forms of piracy are illegal, there is no obvious way of estimating the extent to which pirate copies of information goods are sold rather than distributed for free or the effects of commercial piracy on the profits of copyright owners. Nevertheless, the International Federation of the Phonographic Industry (IFPI) has estimated that approximately 37% of all the [music] CDs that were purchased in 2005 globally were pirate copies.2 However, in the case of the software industry, it is more difficult to find estimates of the prevalence of commercial piracy. The Business Software Alliance (BSA) publishes yearly a piracy study that contains estimates for the piracy rate (i.e. the ratio of the number of pirated software units to the total number of installed software units) for different countries of the world, as well as for different regions of the world. For example, according to the BSA the worldwide piracy rate was 35% in 2005.3 However, such estimates do not make a distinction between commercial and non-commercial forms of piracy.4 Nevertheless, the other surveys of the BSA suggest that both the commercial and non-commercial forms of software piracy are of considerable economic significance.5 There is a relatively large economic literature on end-user copying. 6 Banerjee, 2003, Banerjee, 2006a and Banerjee, 2006b has recently put forward several closely related models of the competition between a monopolist (i.e., the copyright owner) and a single commercial pirate, but it nevertheless seems that until now economists have given much less attention to commercial piracy than to end-user copying. 7 Below I shall put forward a model of the competition between the copyright owner and several commercial pirates, to whom I shall refer as bootleggers. The production costs of pirate copies are low, and in the case of the pirate copies that are distributed in an electric form via the Internet they are almost zero. Accordingly, if intellectual property rights are not enforced, the prices of pirate copies can be expected to fall to zero via Bertrand competition. However, when bootleggers are in danger of being punished for their activities, it may be costly for them to inform potential consumers of their products, since this may increase the risk of being caught and receiving a punishment. For example, if an illegally operating Internet site that offers pirate copies of software products for sale informs its potential customers by sending e-mail messages to randomly chosen addresses, the risk of a punishment is increased by each message. In this case the expected cost from a punishment is analogous to an advertising cost, which explains the positive price of pirated information goods. Information goods can be protected not only by copyright and other intellectual property rights but also by digital rights management (DRM) systems. Digital rights management tools can, broadly speaking, be divided into cryptography (i.e. the distribution of information goods in an enciphered format) and watermarking (i.e. embedding information into a digital product in such a way that each copy of the good becomes different). 8 Watermarks can be used to track down the person who has originally bought the legitimate copy of an information good from which the pirate copies on the market have been produced, which makes it easier penalize commercial pirates. 9 Clearly, a cryptographic device causes a fixed cost for a commercial pirate, but the costs of watermarking can be either fixed or variable: if a bootlegger removes the watermark, its removal causes a fixed cost, but if she sells watermarked information goods, the risk caused by the watermark increases with the number of the copies that she sells. Below I shall analyze the effects of DRM systems and the policy instruments of the government on the profits of the bootleggers and the copyright owner. In the model the “advertising costs” that are caused by an increased risk of legal sanctions also keep the price of pirate copies positive when there are several bootleggers on the market. Accordingly, the model differs from previous work in the same field in so far that it provides tools for analyzing also the market structure of the market for pirate copies as well as its effects on the market for legitimate copies. 10

The role of government policy and DRM systems in preventing commercial piracy was analyzed in a setting in which the profitability of piracy was restricted not just by these factors and the competition with the copyright owner, but also by the competition between the commercial pirates. The apparently puzzling fact that the prices of pirate copies do not always sink to zero via Bertrand competition between the pirates was explained by drawing a distinction between the different effects that government monitoring and DRM systems have on the illegal business model of the commercial pirate. The costs of breaking DRM systems and the expected costs of a punishment were viewed as consisting of a fixed cost of production and an “advertising cost”. The “advertising cost” depends on the number of the consumers to whom a bootlegger (i.e., a commercial pirate) offers her products and it keeps the prices of pirate copies above their production costs. The markets for legitimate and illegitimate copies of information goods were studied both when the copyright owner chooses the price of legitimate copies optimally and when this is not the case, for example, because of lack of information concerning the pirate copy market. The comparative statics of these markets was studied with respect to the quality of the pirate copies and the policy variables, which were the fixed and the variable costs of the bootleggers. Some of the results of this analysis were to be expected, but others were more surprising. For example, it turned out that the revenue of the copyright owner is decreased by an increase in the quality of pirate copies both when the price of legitimate copies is optimal and when it is exogenously given, and that, as long as this does not affect the number of the bootleggers on the market, an increase in their “advertising costs” increases the revenue of the copyright owner and the prices of pirate copies. More interestingly, it also turned out that when there are several bootleggers on the market, an increase in the price of legitimate copies increases price dispersion in the market for pirate copies and decreases their minimum price. Since the pirate copies that have a very low price might be viewed as the counterpart of non-commercial forms of piracy in the current model, this result can be taken to mean that if the copyright owner chooses a sufficiently high price for an information good, it will simultaneously be subject to both commercial and non-commercial forms of piracy. Our analysis also revealed that the effects of the two policy variables on the revenue of the copyright owner were different: whereas it was always in the interest of the copyright owner that the fixed costs of the bootleggers were increased, this was not true of the “advertising costs” because an increase in the “advertising costs” might increase the profitability of commercial piracy and the incentive to enter the market for pirate copies. As the current model contains just a single information good without any network effects, this result is distinct from the familiar results that the producer of an information good might profit from piracy if this makes it easier to sell complementary goods or services to consumers20 or leads network effects that increase the popularity of the product.21 Since the fixed costs of the bootleggers can be caused by technical protection devices whereas, at least in the case of commercial piracy on the Internet, the “advertising costs” result almost completely from the increased risk of a punishment, this result can be interpreted to mean that an improvement in the DRM systems that make an information good technically difficult to copy is always in the interest of the copyright owner, but this is not necessarily true of an increase in the legal protection of information goods. I have not presented a detailed welfare analysis from which one could have deduced the optimal values of the policy variables of the model. Such an analysis is made problematic by the fact that it is not clear whether a social planner should aim at maximizing welfare that is obtained by illegal means, as through commercial piracy. I also have not analyzed the effects of piracy on the market penetration of the product, which would be quite essential if one wanted to include network effects in the current model. Inclusion of network effects would lead to several interesting questions that we have not considered. For example, above it was seen that an improvement in the quality of pirate copies tends to lower the revenue of the copyright owner, and it would be interesting to find out whether this effect could, in the presence of network effects, be balanced by the positive effects of the increase in market penetration that a quality improvement would cause.