In April 2010 the U.S. Government Accountability Office (GAO) released a report to the Committee of Judiciary of the U.S. Senate as well as House of Representatives entitled “Intellectual Property. Observations on Efforts to Quantify the Economic Effects of Counterfeit and Pirated Goods“. Although the report is foced on counterfeiting it also deals with the infringement of intellectual property right (aka “digital piracy”). In the following you can read more about the key findings of the report.
Although the GAO-report primarily deals with counterfeited products such as branded clothes, branded accessories (handbags, purses, etc.), medicines, tobacco, jewellery, and watches, etc., it also highlights the violation of trademarks, patents, and copyrights. As potential negative effects for producers and distributors of original works the report names lost sales, lost brand value or damage to public image, cost of IP protection, and decreased incentives to invest in research and development. However, according to the report suppliers of these products are also bound to generate additional sales of legitimate goods because consumers who, for example, discover new music through file sharing networks in which they “sampled” pirated goods might end up purchasing them. On the consumer side, in turn, negative effects associated with product piracy – damage to health and safety, as well as the lower quality of the counterfeited product compared to that of the original – play a subordinate role, apart from the risk of malware infiltrating one’s computer through file sharing networks For the consumer, the lower price associated with pirated products, which in the case of file sharing of music is zero, constitutes an additional positive effect.
The economic impact of “piracy”
Apart from matters of copyright infringement, from an economic perspective the main task is to discover whether the net effect resulting from “digital piracy” is positive or negative. However, according to the GAO-report it is very difficult, if not impossible, to assess the exact impact “piracy” has on producers and consumers as well as on the economy at large because the available data is generally poor and incomplete: “Most experts we spoke with and reviewed the literature we observed that despite significant efforts, it is difficult if, not impossible, to quantify the net effect of counterfeiting and piracy on the economy as a whole.” (GAO-report, 2010: 15-16). Instead of hard facts, especially in case of copyright infringement, assumptions are used to compensate for the lack of data; thus, the slightest change in assumptions is bound to lead to significantly different conclusions. It is clear that under such conditions the floodgates are open to manipulations, which are also examplified in the GAO-report.
Two key assumptions are required to estimate the net effect of “piracy”. One is the substitution rate, i.e. the consumer’s willingness to switch from purchasing the genuine product to copying it. Numerous studies have often assumed a 1:1 substitution rate. This requires, however, three conditions: (1) that the copy is almost identical in quality to the genuine product from a consumer’s perspective; (2) that a consumer is willing to pay for the copy the same price as for the original; and (3) that a consumer is not aware of acquiring a copy, i.e. that the consumer is deceived. When only one of these three conditions is not met, the likelihood that the genuine product is replaced by a copy decreases. Therefore a 1:1 substitution rate must not be assumed in most of the cases. Especially in case of music file sharing, where the copy can be acquired at a cost equal to zero, it must be assumed that the bulk of transactions is no substitution for the actual sale of music. In other words: if there is no possibility of music file sharing, in most cases the music track or the physical music product would not have been purchased. Therefore, the substitution rate is significantly lower than the 1:1 rate.
In order to calculate the net effect of “piracy” we must take into consideration the value consumers attach to an illegitimate copy. Is the value equal, lower, or even higher compared to the retail price of the genuine product? Who knows? Answering this question raises methodological problems that cannot easily be solved.
Lack of data and methodological problems
Based on this insight, the authors of the GAO-report conclude that in order to quantify the effects of piracy various methods have to be used simultaneously. They present the most commonly used methods to collect and analyze data on counterfeiting and piracy. (1) Extrapolation from a secured database. Since this kind of hard data is not available for “digital piracy”, estimations have to made that can lead to considerable differences in the extrapolation results if the assumptions are slightly altered. Therefore, this method is not useful to assess the effect of “digital piracy”. (2) Supply and demand studies: This method has the advantage of showing consumers’ attitude towards unauthorized copies based on their willingness to pay, the expected minimum quality of the copy, the number of copies acquired, the impact of the knowledge of sanctions on the use of e.g. P2P file sharing networks, etc. However, supply and demand studies are very labor-intensive and costly, and if they only focus on a specific social group, as is the case with P2P file sharing surveys among students, they are not generalizable and thus only of limited applicability. In this context, the GAO-report explicitly refers to the music file sharing study of Rob and Waldfogel (2004 and 2006), which will be evaluated in this blog in the near future. (3) Economic multiplier models: On the basis of an input-output modelling system, the induced effects from a one-time change in final demand are calculated. This methodology can also be adopted to measure the negative impact of “piracy”. The GAO-report, however, is skeptical about the benefit of this method in the context of “piracy” assessment: “Most of the experts we interviewed were reluctant to use economic multipliers to calculate losses from counterfeiting because this methodology was developed to look at a one-time change in output and employment.” (GAO-report, 2010: 23). However, such studies are frequently commissioned by administrative bodies to assess the impact of “digital piracy”. (4) Other data collection and modeling methods: In addition, the report also referred to alternative methods, explicitly citing a study by Felix Oberholzer-Gee and Koleman Strumpf (2004 and 2007), which this bog discussed in detail, and also Hui and Png’s (2003) study on the impact CD burning has on CD sales.
The authors come to the conclusion that there is no “rule of thumb” that measures the effects of “piracy” on an industry or even on the economy as a whole, since “piracy” rates completely differ across different industries. But even within an industry it makes a difference whether one looks at professional counterfeiting of CDs or at P2P music file sharing. Moreover, the use of only one method is not conclusive enough, and a triangulation of methods must be used in order to get reliable results. The more general the findings should be, e.g. the effect of “piracy” on the GNP, the harder it is to get accurate figures. The authors even go so far as to state that the macroeconomic effects of “piracy” cannot be estimated.
The GAO-report officially confirms that the effects of “piracy”, especially of P2P file sharing on social welfare in general and on the music industry in particular, cannot be easily explained. The interrelations are much more complex, and the ongoing decline in sales cannot be blamed (solely) on P2P music file sharing. There is a need, therefore, for other explanations and models aside from the music file sharing hypothesis (see e.g. The Recession in the Music Industry).
Although the GAO-report provides critical insights on the economic effects of “piracy”, the statements are not based on primary surveys and sources. Instead the authors conducted twelve structured expert interviews with government officials, academics, and other stakeholdern and also drew upon information gained from conversations with industry associations and other organizations outside the structured interview process. Moreover, the authors reviewed quantitative as well as qualitative studies published since 1999 of the impact of “piracy”. However, it should be noted that only three relevant studies on P2P music file sharing were included in the study, whereas there are many more relevant papers and studies available on this topic, which will be reviewed in this blog under the title “How Bad is Music File Sharing?” in the near future.
Rob, Rafael and Joel Waldfogel, 2004, Piracy on the High C’s: Music Downloading, Sales Displacement, and Social Welfare in a Sample of College Students. Working Paper, University of Pennsylvania, September 30, 2004.
Rob, Rafael and Joel Waldfogel, 2006, “Piracy on the High C’s: Music Downloading, Sales Displacement, and Social Welfare in a Sample of College Students”. Journal of Law and Economics, Vol. 49, No. 1: pp. 29-62.
Oberholzer/Strumpf 2004, The Effect of File Sharing on Record Sales: An Empirical Analysis. Working paper, Harvard Business School (March 2004).
Oberholzer-Gee, Felix and Strumpf, Koleman, 2007, “The Effect of File Sharing on Record Sales: An Empirical Analysis”. Journal of Political Economy, Vol. 115, No. 1 (2007).
Hui, Kai-Lung and Ivan Png, 2002/2003, Piracy and the Legitimate Demand for Recorded Music. Working paper, National University of Singapore.
United States Government Accountability Office (GAO), 2010, Intellectual Property. Observations on Efforts to Quantify the Economic Effects of Counterfeit and Pirated Goods. Report to Congressional Committees.