Women in Piracy. Experimental Perspectives on Copyright Infringement

1 Introduction

Piracy, defined as a form of copyright infringement, has become a major topic in legal and economic discourse in recent decades, garnering significant attention due to the growing importance of copyright in national economies.^[1] The rise of disruptive innovations, particularly the internet and related technologies, has prompted a re-evaluation of copyright and its infringement, leading to extensive scholarly research. Most studies have focused on assessing piracy’s impact on sales and industry profits, with fewer examining consumer choices from a behavioural perspective (Belleflamme and Peitz 2019; Domon, Melcarne, and Ramello 2019; Olivero et al. 2019).

Yet, to comprehensively address copyright infringement, it is crucial to understand its behavioural aspects, as these reveal the underlying reasons why individuals violate copyright laws. The behavioural law and economics literature highlights the complex nature of legal compliance and the many contingent factors such as risk perception, sanction severity, psychological determinants, and cultural values (Guerra and Harrington 2021). These factors collectively interplay with social norms, which must be understood to create effective legal frameworks.^[2]

Behavioural investigations can offer valuable policy-making tools and insights into specific phenomena.

In this paper, we conduct a behavioural investigation using experimental data. To our knowledge, this is one of the very few experiments on music piracy. Using experimental methods, we elicited individual preferences regarding legal and burned CDs. We used also hypothetical choices to measure further purchasing and consumption behaviours, including downloading music files from the internet. The respondents are university students who are considered a good sample for observing consumption of legal and illegal music, given that they have low spending capacity, while high demand for music (Chiang and Assane 2008).

Following the growing literature in economics, we focus on whether gender influences copyright infringement, examining differences between men and women and the specific characteristics that explain these differences.

Our results indicate that while women are less likely to engage in piracy, when they do participate, their level of consumption (measured by the number of pirated files) is similar to that of men, thereby reducing the gender gap in piracy behaviours. Both genders perceive market prices for original CDs as unfair, but women express stronger feelings of unfairness and value legality more. Women seem to refrain from piracy more due to ethical concerns, while men value the product’s originality more than women.

The paper is organised as follows: Section 2 reviews the literature on the law and economics of piracy. Section 3 summarises the existing evidence on law infringement and gender. Section 4 describes the experimental design, and Section 5 presents the results and discussion. Section 6 concludes the paper.

2 Copyright and Piracy

The term “copy-right” refers to the legal right granted to a rights holder to reproduce creative works. This exclusive right allows the rights holder to prevent others from copying and distributing the copyrighted work, creating a market entry barrier for the protected information goods and potentially giving market power to the rights holder. Piracy, in contrast, is the unauthorised reproduction of copyrighted material, which can harm the rights holder by reducing sales of legal copies due to the spread of illegal copies (Nicita, Ramello, and Scherer 2005; Domon, Melcarne, and Ramello 2019).

The scholarly discussion on piracy in economic literature began in the early 1980s, driven by technological advancements that made large-scale copying easier, such as photocopying for texts and music cassettes for audio recordings. The debate intensified in the late 1990s and early 2000s with the rise of digital technologies and the internet, which enabled widespread piracy on an unprecedented scale. Digital technology allowed for high-quality duplication of audio files, while the internet provided an efficient distribution channel, challenging traditional commercial delivery methods and changing the dynamics of the music market.

Technological changes since the 1980s have continued to reshape media markets, particularly in the audio and video sectors, also impacting the ability to infringe copyright law. Today, music can be (illegally) streamed or downloaded from many websites with virtually no technological barriers (Belleflamme and Peitz 2019; Perry and Sinnreich 2022).

Despite the shift to online systems, traditional formats like vinyl and CD records remain significant. In 2023, they accounted for 18 % of the German music market (about 190 million euros) and 17.4 % of the Italian market (about 62 million euros).^[3] In France, the latest data (limited to the first half of 2023) shows an even larger share: 22.23 % of the country’s market, amounting to 88.5 million euros.^[4] Globally, physical music records accounted for 17.8 % of the market in 2023, totalling 5.09 billion USD according to the Global Music Report 2024.

Since the early literature, law and economics research has focused on whether piracy negatively impacts sales and profits (Silva and Ramello 2000). A few studies have explored additional nuances and complementary issues, such as the effects on musicians’ incentives and the production of recorded music (Handke 2012; Waldfogel 2012). Some pioneers have examined idiosyncratic features related to individual and social factors. Early research by Liebowitz (1985) on photocopying showed that some consumers might pay a premium for the ability to make copies, potentially increasing profits for rights holders. Studies by Conner and Rumelt (1991) and Takeyama (1994) highlighted the role of network externalities in driving demand for information goods, suggesting that illegal copies could sometimes enhance both producer profits and consumer surplus.

These studies shifted the focus to individual behaviours, making them crucial for comprehensive analysing copyright infringement and developing effective policy responses.

3 Piracy, Gender and Other Illegal Behaviours

The specific literature on music piracy is limited (Montoro-Pons, Caballer-Tarazona, and Cuadrado-García 2021), and experimental or behavioural research is almost non-existent, with a few exceptions.^[5] Most empirical studies suggest a gender gap in piracy, with women burning, sharing and downloading less than men (Kini, Rominger, and Vijarayaman 2000; Kwong et al. 2003; Kini, Ramakrishna, and Vijayaraman 2004; Chiang and Assane 2007, 2008; Hinduja and Higgins 2011; Olivero et al. 2019; Cox and Collins 2014; Brown and Krause 2017).^[6]

Several explanations are put forward for this difference; particularly, personal preferences and ethics are likely to play a role, there is an impact on social welfare and its distribution, and the setting in which the behaviour takes place (Seale, Polakowski, and Schneider 1998; Croson and Gneezy 2009; Guerra and Zhuravleva 2022; Guerra, Randon, and Scorcu 2022). Chiang and Assane (2008) find that men engage in file-sharing more than women, even at equal levels of PC ability. They also observe that women are more sensitive than men to copyright law enforcement, more risk-averse, and have a higher willingness to pay for legal copies than men (however, this might be due to the preceding differences – see also Chiang 2009). Additionally, women respond more to anti-piracy messages, curbing their illegal downloads, while men increase them in response to such messages (Whitman, Murad, and Cox 2024). Women’s higher willingness to pay for legal copies implies they retain a larger surplus than men when buying a legal CD at a given price. Therefore, they sacrifice less marginal utility and have less incentive to engage in piracy.

These observations align with the findings of crime studies, which generally report a gender gap in favour of women. In fact, research shows that men are primarily responsible for violent offences, while women are more prone to non-violent law infringements, such as property crimes. Recent studies (Britton, Jacobsen, and Howard 2018; Belknap 2023) have noted a convergence in crime rates between genders, mirroring the reduction in the gender gap in labour market participation. However, despite this trend towards convergence in overall illegal activities, women still commit fewer violent crimes than men. This difference is attributed to the varying pro-social cognitive skills that arise from early education and cultural roles in socialisation (Bennett, Farrington, and Huesmann 2005).

Other studies show there is also a gender gap in risk aversion (Arch 1993; Powell and Ansic 1997; Hinz, McCarthy, and Turner 1997; Fehr-Duda, de Gennaro, and Schubert 2006; Eckel and Grossman 2008; Charness and Gneezy 2012),^[7] social preferences (Croson and Gneezy 2009; Solnick 2001), ethical decision-making (Adams and Funk 2012; Frank, Lambsdorff, and Boehm 2011; Rivas 2013; Swamy et al. 2001) and law violations (Torgler and Valev 2006).

Empirical observations (Yagil 1998) have shown that differences in financial incentives and perceptions of the likelihood and consequences of an action help explain the gender gap in law violations, which could also be relevant for copyright infringement.

Regarding property law violations, Campaniello and Gavrilova (2018) recently showed that the gender gap in participation is partly due to differences in the anticipated earnings from the crime and the probability of arrest. Thus, women’s distinct behaviours may once again be influenced by their unique status in society.

4 The Experimental Design

This research examines two types of music piracy: purchasing illegal CDs (“physical piracy”) and illegally downloading online files. Data on illegal CD purchases were collected through an experiment and via a survey, while data on illegal downloads were obtained via a survey. The data were obtained from the same sample with a post-experiment survey. We always tried to avoid any ethical connotations, including using words like “piracy” that convey negative meanings for most people. So, we decided to use general terms like “original CD” or “burned CD” that do not have a strong connotation and are perceived more neutrally. Moreover, as it will be further explained, the definition “burned CD” is also important, as it will be further discussed, for avoiding misleading participants.

To study individuals’ attitudes toward physical piracy, the experiment measured their willingness to pay for either an original CD or a copied (burned) CD, in an auction setting. We conducted an auction where both an original and a burned CD were sold in the laboratory. To strengthen the incentive mechanism, we used a sealed-bid third-price auction, which allows two participants to win by paying the maximum willingness to pay of the third-highest bidder (Kagel 1995). Each subject took part in two auctions, one for an original CD and one for a burned copy.^[8] Half of the subjects bid first for the original CD and then for the burned CD, while the order was reversed for the other half. The order of play of the two auctions in the different sessions of the experiment was randomly assigned.

Before the auction, participants gathered in the same room and were provided with paper and envelopes to submit their bids. To account for varying musical tastes, no specific CD was auctioned. Instead, participants were bidding for whichever CD they preferred at the time of the auction. To this end, subjects were asked to indicate three CDs in order of preference so that the experimenter could have an alternative in case one was unavailable. It is important to note that Italian copyright law has no exception in making the infringement possible, which is also considered a crime. So, to make the experiment viable without infringing the law and, what is more, without inducing the students to commit a crime while at the same time avoiding deception, after consulting an IP lawyer, a criminal attorney, and with the help of the student association, in the auction experiment where also burned CDs were to be sold for real, we bought the files over the internet. We did not disclose this information to our subjects; we deemed it preferable to hide a piece of information rather than violate Italian law. Additionally, as previously mentioned, we always referred to “burned CD” that has the only implication that the CD is not an original one but does not give any specific information about the production. As for ethical concerns, we decided to have some help from legal experts and student associations to agree with them on a fair experimental procedure.

The experimenter had a day to get the CDs and sell them to the winners for the agreed price.

At the end of the experiment, the participants completed a questionnaire. This included questions about demographic status (age, gender), disposable weekly income, and family income. They were also asked about their opinions, evaluations, and general attitudes towards piracy, including whether they considered copying a CD illegal, criminal,^[9] or unethical. Additional questions covered their habits regarding buying pirated CDs on the street, sharing them among friends, and illegally downloading music files from the internet, including the monthly frequency of these actions (once or less, twice, three times, or more than three times a month). This last part, of course, has the nature of survey data and was used for adding the variables on illegal downloads.

Participants were then asked to provide their monetary valuation of an original CD, whether they considered the cost of an original CD too high, the number of original CDs they bought, and the number of CDs burned per month (one or less, two, three, more than three). Finally, we asked about their perceived risk of being caught when buying a burned CD on the street or downloading music illegally. This probability was rated on a scale from 0 (“not at all likely”) to 3 (“very likely”).

Finally, participants were asked about the time and cost of burning a CD, to understand their familiarity with burning techniques, the actual cost they would incur, and to use the individual cost of burning as an additional factor in explaining their behaviour. Note that, despite variations in technology quality among participants (e.g. more or less obsolete PCs, different internet connection speeds, etc.), current devices for burning CDs are all based on comparable technology.

The experiment involved 120 undergraduate students from two universities in Northern Italy, divided into eight sessions of 15 subjects each. After testing the feasibility of aggregating the data from all sessions, we constructed a single database.

5 Results and Discussion

The findings reveal distinct and novel features, including gender differences. Table 1 presents the main descriptive statistics.

The sample comprised a majority of male students (65.13 %). Participants’ bids for each CD serve as proxies for their willingness to pay (WTP). The average bid for an original CD is higher than for a burned copy, a difference significant at the 99 % level (t-statistic = 15.84). Interestingly, the average bid for the original CD is lower than the fair price for the same good as deemed by the participants, and it is also significant at the 99 % level (t-statistic = −3.72). Only slightly more than half of the subjects consider piracy illegal, and only half think that this activity – though illegal – is also unethical.^[10]

6 Conclusions

In the past three decades, the increasing significance of copyright trade has drawn scholarly attention to piracy, defined as copyright infringement. Some studies have specifically examined music piracy, empirically investigating its impact on music sales and firms’ profitability. To our knowledge, only a few studies have explored consumer behaviour in this context, with one experimental study conducted to date. This paper aims to bridge this gap by conducting experiments to investigate whether gender differences exist in piracy akin to those observed in other types of law infringement.

Our hypothesis is that if, after controlling for other variables, women bid less than men for illegal copies and there is no gender difference in bids for legal copies, then women value pirated CDs less than men. This would suggest a lower inclination toward piracy among females. The results underscore a robust gender effect in piracy, influenced by each gender’s distinct responses to perceived unfairness. Additionally, our data suggest that effective enforcement against piracy, coupled with highlighting its ethical concerns, can help combat this illegal activity. Notably, our findings also reveal a significant positive price for burned CDs (see Maffioletti and Ramello 2004).

It is widely recognized that women and men tend to exhibit different behaviours in similar situations. Existing piracy studies generally indicate that females comply more with copyright laws than males. However, these findings often rely on surveys, which may be biased as respondents might not disclose their actual behaviour due to fear of judgment or stigma from interviewers or because they lack incentives to be truthful (see Whitman, Murad, and Cox 2024). To mitigate these biases and analyze more reliable data, we employed an experimental approach to elicit subjects’ willingness to pay for original and burned CDs using a market institution, specifically an auction.

The experimental results indicate that females engage in piracy less frequently than males; they are less likely to act illegally. However, once they do engage in illegal behaviour, that is, once they start to download music files illegally, the number of such files does not differ between men and women. Both genders perceive current market prices of original CDs as excessively high and unfair, with women generally expressing stronger feelings of unfairness compared to men.

Men value originality more than women, while both genders value the burned copy equally. Women value legality more than men, according to any measure.

For women, piracy seems to be driven in part by a desire to penalise recording companies perceived as selling products at unfairly high prices. However, females may refrain from burning and illegally downloading music, viewing these activities as more criminal or unethical than their male counterparts do.

Another noteworthy finding is that the awareness of the likelihood of detection while purchasing illegal CDs on the street deters participation in this black market; enhanced controls could effectively combat street piracy. At the same time, campaigns emphasising the ethical implications of online piracy could help curb illegal downloads.

This paper demonstrates further a gender effect in piracy, attributed in part to different responses to perceived unfairness between genders.

We are aware that there is a methodological limitation of our experiment – but also to all experiments of this kind, since the researchers have to deal with the problem of avoiding subject deception and respecting the law. There is no easy way out of it. However, we think that our choice was the only viable to reduce the potential negative effect on the results.^[19] For this reason, we also think that other experiments should be run to provide more insight into the topic and to improve the transparency of the design. Despite this methodological question, we think that our results are robust and do provide a new way to look at piracy. Our findings raise an important question: should the traditional models used in mainstream law and economics – and legal systems, particularly in the context of property and intellectual property – continue to rely solely on cost-benefit analyses to deter lawbreakers? Such models, which focus on creating expected costs for violators, may overlook the psychological and belief-driven components of potential infringers’ behaviour. It may be more effective to substitute or complement these models with behavioural policies that account for individual differences rather than simply increasing the penalties for violations. One-size-fits-all policies risk being ineffective, as they fail to address the diverse identities and motivations within society.