Master Thesis Behavioral Economics and Game Theory
Music Piracy Among International Students:
The Effects of Risk Perception and Social Norm on the Decision to
Download Songs Illegally
Name: Martín Ossandon Busch Student Number: 11806915 Date: 02/09/2018
Specialization: Behavioral Economics Subject: Economics
Statement of Originality
This document is written by Student Martín Ossandon Busch who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
This thesis studies the change in the illegal download of music of international students once they have moved to a different country. The focus is on two determinants of music piracy that are different between countries: the perception of the risk of getting caught and the view peers have about this activity. The findings show that students that have more peers that discourage music piracy in their current country of residence compared to their previous one are likely to decrease the amount of unauthorized songs they obtain. Regarding risk, individuals that believe the probability of getting caught for music piracy to be higher in their current country of residence than in their previous one are more likely to increase their amount of songs illegally obtained, a result that challenges theoretical predictions.
TABLE OF CONTENTS
1 INTRODUCTION ... 1
2 LITERATURE REVIEW AND HYPOTHESES ... 2
2.1. COUNTRY VARIATION ... 2
2.2. PREVIOUS RESEARCH ... 4
2.2.1. INTENTION BASED THEORIES ... 4
2.2.2. RISK ... 5
2.2.3. SUBJECTIVE NORM ... 6
2.3. ETHICAL DECISION MAKING ... 7
2.4. DEMOGRAPHICS ... 8
3 RESEARCH DESING ... 8
3.1. DATA COLLECTION AND SAMPLE SELECTION ... 8
3.2. VARIABLES ... 9
3.3. CONTROL VARIABLES ... 10
4 SAMPLE ANALYSIS ... 11
4.1. DEMOGRAPHIC VARIABLES ... 11
4.2. MUSIC CONSUMPTION ... 12
4.3. CHANGE IN PIRACY BEHAVIOR ... 13
4.4. RISK ... 13
4.5. SOCIAL NORM ... 13
5 ANALYSIS AND RESULTS ... 14
5.1. RISK ... 15
5.2. SOCIAL NORM ... 16
6 GENERAL DISCUSSION ... 17
7 LIMITATIONS ... 19
8 CONCLUDING REMARKS AND FUTURE RESEARCH ... 20
1
1 INTRODUCTION
With the growth of the Internet at the end of the last century came music industry’s biggest treat: digital piracy. This problem does not only steams from the illegal copy and distribution in physical form, such as CDs and DVDs, but also from the digital sharing of material through peer-to-peer (P2P) programs in the world-wide web. Digital piracy can be defined as “the infringement of copyrighted content (such as music, films, software, broadcasting, books, etc.), where the end product does not involve the use of hard media, such as CDs and DVDs” (Stryszowski & Scorpecci, 2009), and it is an activity in which individuals with computer access can easily engage to obtain licensed material for free.
The damage this practice has done to the music industry has been worldwide. Only in Europe, a report conducted by the European Union found that in 2014 the record industry lost nearly €170 millions of sales revenue in the region due to the consumption of music through unauthorized channels, corresponding to 5.2% of the sector’s revenue from both digital and physical sales (EUIPO, 2016). This illegal activity is of public concern since it has direct effects in terms of employment losses and indirect ones on governments due to loss in potential revenue and the costs implied in persecuting piracy. Adding these different costs the same report concludes that the infringement of intellectual property rights costs the EU economy around €336 million in lost sales, leads to a loss of 2,155 related jobs and of €63 million in government revenue (EUIPO, 2016).
For decades’ record labels have tried to stop music piracy by taking legal actions against unauthorized providers of music, and websites and platforms that facilitate online file sharing such as Napster, Megaupload and thePirateBay. Nonetheless, the consumers that have obtained music through these channels have remained safe due to the incapacity labels have to prosecute music pirates individually. The problem faced by the music industry is exacerbated when considering its international dimension: thanks to the Internet, digital files can easily travel from one country to another, and since copyrights are protected under national legal frameworks, what constitutes infringement, or fair and private use, differs across borders, making it harder to enforce corrective actions.
Research on this subject has identified several reasons that lie behind the decision to pirate music, such as the price of music, attitude towards the music industry, the low risk of getting caught and peer influence. Through data on international university students, the present study seeks to understand an aspect that has not yet been addressed: the impact that moving from one country to another can have on a person’s consumption of music through illegal channels. In
2
particular, this report focuses on two channels that can alter an individual’s decision to download and that are expected to differ among countries: legal risk and social pressure. The first one relates to the different legal structure a person encounters in a new country and which can be perceive as more or less harsh against digital piracy. The second one deals with how accepted music piracy in one’s social circle is and how it intervenes in the decision to acquire music illegally. The next section provides a concise account of music piracy and how it has been addressed on previous research.
The next section offers an overview on previous research on music piracy, highlighting those aspects that are of special interest for the present study. Section 3 presents the research design, the methodology used and the construction of the variables used in the posterior analysis. Section 4 presents the main characteristics of the sample while Section 5 consists of the multinomial regression model used to predict the change in individuals’ download behavior and is followed by a general discussion in Section 6. The last two sections correspond to the limitations this study faced, and the concluding remarks and aspects that future research on the subject should take into consideration.
2 LITERATURE REVIEW AND HYPOTHESES
2.1. COUNTRY VARIATION
Different countries have different approaches to music piracy. They can have conflicting views regarding the legality of making backups, downloading vs uploading files, the role of internet service providers in copyright infringement, and the evidence required for prosecution, to name a few dissonances. Moreover, the music industry’s efforts to fight piracy have been mostly deployed in Western countries, where the biggest share of their revenues are made. In addition, some governments have made little effort or shown a lack of interest in prioritizing the protection of intellectual property rights, especially the ones of foreign firms.
The consequences for being caught pirating music can also be more or less strict among countries. Japan, for example, has one of the harshest laws against copyright infringement, with consumers risking a 2 million Yen fine (USD 18.000), 2 years’ jail time for downloading and up to 10 years for uploading content. In Germany, downloading a single file illegally can lead to a €300 to €1000 fine, and is an activity that is highly persecuted. France and New Zealand have a three-strike system, where users are warned by email and mail before having to appear before a judge which can impose fines or limit their internet access.
3
Other European countries have more tolerant laws in this matter. In Switzerland, the downloading of copyrighted material is allowed for personal use, however, uploading files to share is illegal. In the Netherlands, up until recently it was permitted to download pirated content. That changed in 2014 when the downloading of copyrighted material was made illegal after a ruling of the European Court of Justice. However, in practice individual downloaders are not prosecuted and do not face criminal charges. Only sites that provide access to illegal material are.
A case worth mentioning is that of developing countries, which according to the industry’ reports hold above-average levels of digital piracy. In their “2017 Special 301 Report”, the Office of the United States Trade Representative recommended for countries such as Brazil, Colombia, Indonesia, Peru, Thailand and the United Arab Emirates to be put in the organization’s watch list of countries with problems in their protection of intellectual property. Digital piracy in the developing world provides access to a variety of media goods, from music to software, to a population that does not have the means to acquire them through licit channels. According to the 2011 “Media Piracy in Emerging Economies” report, the most important predictor of copyright infringement is the relative price of media (Social Science Research Council, 2011). In countries such as Brazil and South Africa, for example, the price of music and movies relative to income is ten times the one in Europe. Regardless of the major benefits that globalization has bring to media culture, a democratization in access has not taken place, leaving some with no better choice to obtain it through illegal means.
This result is exacerbated by the predominance of international companies in their markets, that by hampering national competition, increase the prices. The report also points out the problem of enforcing the law. Even though record and film companies have been successful at lobbying for more strict laws, governments have shown little willingness to put them into practice.
The argument behind this study is that due to these country differences, moving from one country to another will affect a person’s decision to engage in music piracy. A population that is susceptible of being affected by such differences are international students, that while pursuing studies abroad have to adapt to distinctive legal and social norms. Those who download unlicensed music can either decrease, increase or keep constant the amount of songs obtained illegally. Previous research has dealt with the determinants of music piracy, which can explain why a person would change his online behavior when living abroad.
4
2.2. PREVIOUS RESEARCH
An extensive literature has sought to explain the motivations behind a person’s decision to acquire unlicensed music. In addition to studies finding different behavioral dynamics based on demographic determinants such as gender, age or income, others have found that the attitude towards the industry; an individual’s music consumption; and ethical, legal, and social considerations are all key in a person’s predisposition to pirate songs.
Research on this field has been by two currents of thought: one emphasizing the intention to engage in a certain behavior based on psychological factors such as attitude and subjective norm. The second current posits that decision-making is mediated by self-ethical choices, which are a key explanation of an individual’s actions (Dilmperi et al., 2017).
2.2.1. INTENTION BASED THEORIES
Intention based theories are founded on the theory of reasoned action (TRA) developed by Fishbein & Ajzen (1975), according to which a person’s decisions are conscious and controlled. The intention to do a certain action depends on intention, which is determined by two cognitive elements: attitudes towards the comportment and the subjective norm that prevails in his social circle. While attitudes relate to the consequences of actions, subjective norm reflects an individual’s expectations about what those close to him believe he should or should not do. According to Ajzen (2002), the two main predictions of the TRA are that:
(i) the more positive the feelings towards an action, the greater the intention to perform it.
(ii) the greater the social pressure to behave in a certain way, the more likely it is for an individual to act accordingly.
Through the TRA, studies have been able to identify various determinants of music piracy. Dilmperi et al. (2017) built a conceptual model to explain the intention to acquire unauthorized music through five antecedents of attitude: the perceived quality of music (PQM), perceived benefits of piracy, price of legitimate music, perceived likelihood of punishment (PLP) and idolatry. Through a survey on university students in the UK and Greece, the researchers found that the intention to obtain music legally can be explained by idolatry, music quality, risk and social pressure. That is, individuals that believe the probability of being caught in this illegal
5
activity to be higher, that value music quality more, and/or that feel that such an unlawful behavior is not supported by their close ones, will show a higher intention to buy music from both physical and digital stores. Regarding idolization, the study suggests that even though consumers with a strong connection with artists do not like illegal CDs, they do show a higher tendency to download music through peer-to-peer platforms, a result that goes against the hypothesized predictions about fans. Out of the different attitudinal factors, the study indicates that the perceived benefits of piracy are the main predictor of the decision to acquire illegal music. These benefits include aspects such as collection, rarity of music, and time and money saved. In the same line, Kos Koklik et al. (2016) found that knowledge about how to obtain pirated files and the perceived benefits of piracy compared to their legal counterpart are a significant predictor of an individual’s attitudes and intentions towards music piracy.
2.2.2. RISK
One of the determinants of attitudes towards music piracy that any research must consider are the risks and penalties that illegal downloading entails. The industry has tried to tackle piracy by taking legal actions against website owners and by pressuring governments into taking a stronger hand against those infringing intellectual property rights. They have opted for a more coercive approach in their fight against piracy.
According to expected utility theory, an act is the result of the preferences over a range of possible outcomes, weighted by the probability of each of these outcomes taking place. It follows, that the decision to download music is a function of the net gain between benefits, such as getting it for free or obtaining it immediately, and the probability and severity of the legal punishment for getting caught, which can range from a warning to having to pay damages to right owners and even jail time. Furthermore, for the generations that have grown up with easy and fast internet connection the illegal downloading of files has been the norm, leading to an unclear perception of the legal risks and economic damages that this activity entails. Studies have shown mixed results regarding the impact of risk on the decision to engage in music piracy. Research conducted on college students has found that they perceive a low probability of getting caught and a small punishment in case of being found guilty for music piracy (Borja et al., 2015; Pryor et al., 2008). Gopal et. al (2004) run a test in two populations of students, one that had been informed of the consequences of piracy and one that had not, and found no difference in their intention to pirate music. Lalović et al. (2012) found no significant effect of prosecution risk in the decision to obtain music illegally, which they
6
attributed to the fact that the Internet hides one’s identity providing a sense of invisibility, and to ineffective fines that have been a threat only on paper.
Other studies have indeed find a negative impact of risk on the decision to pirate music (Pryor et al., 2008; McCorkle et al., 2012). Bellemare & Holmberg (2010), found that a 1 percent increase in a respondent’s believe of the likelihood she would get caught pirating music, decreased the probability of her last song having been obtained illegally by 0.5 percent. Such a result favors the view that the threat of legal action can indeed have an impact on a person’s decision to incur in music piracy. Bhattacharjee et al. (2006) found that the risk attitude of individuals who share a considerable number of files online was different than the one of “soft” users and that their propensity to pirate was altered under stronger penalties.
Based on this theoretical background the following hypothesis was developed:
H1: the perceived likelihood of getting caught for music piracy negatively affects the amount
of songs illegally downloaded.
2.2.3. SUBJECTIVE NORM
The effect of social influence on music piracy has also attracted the attention of researchers. Yang & Wang (2014) examined five different sources of social learning that could favor the formation of a positive attitude towards music piracy. According to them, parents, peers, traditional media, the Internet itself and the music industry are the socialization agents that impact music piracy attitudes and behavior through two modes of learning: imitation and reinforcement. The first one referring to the adoption of a comportment after observing a similar behavior in others, while the second one dealing with a behavior being determined by the balance of expected rewards and punishments from society or subgroups (Akers & Jensen, 2006).
Regarding music, the reference group can be as much as a determinant of the kind of music people listen to, as of the channels through which music is obtained. While a positive attitude towards piracy from a person’s social group increases the likelihood of downloading music illegally (Borja et al., 2015), peers also determine the decision to pirate while serving as a benchmark (Rochelandet and Le Guel, 2005). In that sense, this behavior seems to be more accepted if “everyone else is doing it” or if “I download much less compared to other people”. Furthermore, studies have found peer pressure to be the strongest predictor of piracy out of a set of explanatory variables (Chiou et al. (2005), Ingram & Hinduja (2008). Lastly, the
7
influence of the social sphere seems to also be determined by the kind of it is relationship. It has been found that peers and the Internet have a direct impact on attitude and behavior towards music piracy, while parents and the music industry only exert an indirect effect on the behavior by shaping people’s attitudes (Yang & Wang, 2015).
Supported by this research, a second hypothesis was formulated:
H2.: Peers’ views about music piracy positively impacts the likelihood of engaging in music
piracy.
2.3. ETHICAL DECISION MAKING
Since piracy is often considered an illegal or unethical activity, the perspective of theories of ethical decision making is an addition to better understanding its determinants. Regarding music piracy, most theories agree on that the decision to pirate will present an ethical dilemma at the individual level (Chiou et al., 2015). Even when consumers can view unauthorized downloading as something different than stealing, nearly all of them would agree that music piracy is a wrongful act, and that can carry legal costs in terms of fines or jail time.
Lysonski & Durvasula (2008) studied the role of ethics in students’ decision to engage in illegal file sharing. The results show that even though people with higher moral standards are more aware of the social costs of digital piracy, they do not believe the unauthorized downloading of music to be unethical or that it should be criminalized in the same way that the theft of CD’s is. In their study, respondents that said they would not steal from a music retail store also showed a disposition to obtain music files illegally through the Internet.
Several studies have found a negative influence of moral intensity in the attitude towards music piracy (Gopal et al., 2004; Chiu et al., 2005; Kos Koklic et. al., 2012; Cesareo & Pastore, 2014). More precisely, Bellemare & Holmberg (2010) found that a 10% increase in their morality index decreased the likelihood of a student having her last song obtained illegally by 0.2%. Nevertheless, ethical predisposition is something that is hard to change by itself, but it can be altered through socialization. Hence, it can be expected that a change in the view of piracy will take place through peers, which is captured in the second hypothesis.
8
2.4. DEMOGRAPHICS
Studies have shown different attitudes and piracy related behavior depending on age, gender, location, income, among other demographic variables. In a study conducted in Taiwan, Pai & Chie (2017) found that men were more prone to download music illegally and that disposable income had a negative impact on pop music piracy. Bellemare & Holmberg (2010) found that the decision to purchase music is largely determined by a student’s full income, which considers the amount of money parents make: the lower the annual family income, the more likely a respondent was to have acquired her last song through illegal channels. The results from Bhattacharjee et a. (2003) also show that young males and people with fast internet connection are more likely to pirate music. Based on this literature, the present report includes a series of demographic indicators, such as gender and income, for their moderating effects on an individual’s decision to download music illegally.
3 RESEARCH DESING
This study examines how music piracy is affected among international university students when they move to a new country with different legal and social norms. The emphasis is on the risk associated to downloading music and the social view on piracy, since both determinants change from one country to another. Research on music piracy has focused on its determinants, such as perception of the music industry, income or the likelihood of being punished, but so far, they have not been any analysis on how people behave in this matter when faced with a change in their legal and social surroundings.
Given that countries have particular legal systems and that they persecute digital piracy more or less strongly, it is expected that international students will download less unlicensed music under a legal framework that they perceive as more strict. In addition, if students are confronted with a higher ethical judgment towards piracy from their peers, social norms can lead to a decrease in the amount of music illegally acquired.
3.1. DATA COLLECTION AND SAMPLE SELECTION
To test the main hypotheses, an online survey was developed and distributed to international students currently pursuing studies in the Netherlands and in France. The questionnaire was built on Qualtrics and was available for a week in July 2018. The invitation link was published
9
in social network groups meant for international students and the chance to win €10, from a total of €50 to be paid out, was offered as an incentive to complete the form.
The survey asked respondents about their online music consumption habits, both before and after changing countries of residence. It also included measurements of risk perception and expected punishments for getting caught downloading music illegally, and of the views peers have about music piracy in the two countries where they had live. Given that digital piracy is an illegal act people might not be willing to disclose that type of information. Hence, anonymity and confidentiality were assured by setting a different webpage where respondents could provide their email for being contacted in case of winning and that was not linked to their answers.
3.2. VARIABLES
To construct the indicators, most of questions were based on existing research, assuring that content validity was not a problem. To generate the dependent variable, respondents were asked about their monthly amount of songs illegally downloaded both in their previous and current country of residence. They could choose between 7 different ranges between 0 and more than 250 songs.1 These ranges were then assigned values from 0 (0 songs downloaded) to 6 (more than 250 songs), and after that the values for the previous country of residence (PCR) were subtracted from the ones for the current country of residence (CCR). The result were negative values for those cases were there was a decrease in the amount of unlicensed songs downloaded when moving to a new country, zero in case the amount remained constant and positive values if there was an increase in the quantity of songs obtained illegally. Finally, all the no changes were coded as 0, the positive ones as 1, and the negative ones as 2. An example can be found in Table 1.
Table 1. Example construction of dependent variable Monthly amount of songs
illegally downloaded PCR
Range Value PCR
Monthly amount of songs illegally downloaded CCR Range Value CCR Value CCR – Value PCR Downloading Difference
25-50 Songs 3 Less than 10 songs 1 -2 2
The perception of risk indicator was built by asking respondents if they believed the probability of getting caught for music piracy to be smaller than that of shoplifting, both in their previous
1 These ranges were taken from Medlin et al. (2015) and adjusted for a monthly amount of songs downloaded,
10
and current country of residence, an indicator used by Cox & Collins (2014). These responses were then given categorical values: 0 if they believe the probability to be higher in both countries, 1 if they believe it to be smaller than shoplifting in their previous country of residence but higher in their current one, 2 if they believe it to be higher in their previous country and in their current one, and 3 if they did not know.2 The details can be found in Table 2.
For measuring the effects that a change in the subjective norm can have on music piracy, respondents were asked if the illegal download of music was discouraged by their peers in each country, an indicator previously used by Yang & Wang (2015). The possible answers were constructed on a 5-point scale (none of them, a few of them, quite a lot of them, nearly all of them, all of them). After that, a similar coding procedure than the one done for the downloading difference was done. The details can be found in Table 2.
Table 2. Construction of independent variables
Variable Question Categorical Values
Risk Would you say that in your previous/current country of residence the probability of getting caught is…?
0 = believes the probability of getting caught to be higher than the one for shoplifting in both countries.
1 = believes the probability of getting caught to be smaller in the PCR, but higher in the CCR.
2 = believes the probability of getting caught to be higher than the one for shoplifting in both countries.
3 = does not know the probability in at least one country.
Social Norm
Is the unauthorized download of music discouraged by your close friends in your
previous/current country of residence?
0 = same amount of peers discourage music piracy in PCR and CCR.
1 = if music piracy is discouraged by more peers in the CCR than in the PCR.
2 = if music piracy is discouraged by less peers in the CCR than in the PCR.
3.3. CONTROL VARIABLES
Age, monthly disposable income and a dummy for preferred platform to obtain music (legal vs illegal) were used as control variables. Age was included as a continuous variable, income was constructed based on 3 categories, low (€0-500), middle (€500-1000) and high (€1000 or more). Contrary to what has been found by previous research, here age is expected to have a positive impact on piracy, since the age limit of the sample is of 28 years. It is expected that older students are more likely to use such platforms because they grew up with them and are more familiar with this way of consuming music. Younger ones are expected to pirate less,
2 There were no respondents that believe the probability to be higher than shoplifting in their previous country
11
because they preferred easy-access alternatives such as streaming platforms. Regarding income, a negative relation with piracy is expected. Buying legal music or using authorized channels represents less of an opportunity cost for those with more resources.
The third control variable was constructed by asking students what their preferred way of obtaining music was from a wide range of legal and illegal options such as torrent, streaming services, YouTube or a YouTube to MP3 converter. Responses that privileged an illegal platform were coded as 1 and the ones for a legal one as 0.
4 SAMPLE ANALYSIS
Once the questionnaire was constructed, it was published online through Qualtrics. The invite link was shared in social network groups for international students living in Amsterdam or in Paris. Exchange, bachelor, master and PhD students were invited to participate and the website was opened for a week.
Out of the 93 respondents, 29 (25%) said they did not download any songs illegally in the last month before leaving their previous country of residence and in the month before answering the survey. This quarter of the original sample was not considered for the analysis since they were not music pirates. Afterwards, 5 other observations whose current country of residence was not Europe were taken out of the sample since the study is circumscribed to students living in Europe. Hence, the analysis that followed was based on an N = 59. Given the small sample size, the results of this study remain explorative and the different findings should be regarded as suggestive at most.
4.1. DEMOGRAPHIC VARIABLES
The main characteristics of the sample are the following:
I. Out of the 59 respondents, 71,2% were female and 28.8% males. II. The mean age of the sample was 23 years old.
III. The sample reflected high intra-Europe movement, as half of the respondents had previously lived in another European country. 15% came from Asia, 14% from North America, 17% from South America and 5% from Africa.
IV. Regarding revenue, 42% of the sample was in the low-income range (<€500), 39% in the middle one, and 19% in the high one (>€1000).
12
The summary of these variables can be found in Table 3:
4.2. MUSIC CONSUMPTION
Regarding the music habits of the respondents, their main features are: i. There is a noticeable difference between countries
in the percentage of respondents that said they downloaded 0 songs. 33,9% of students did not obtain unauthorized songs in the country they live in, compared to only a 3% of the sample in the previous one.
ii. Most of respondents do not download more than 25 songs per month in both countries. Besides implying a small amount of songs obtained, it
represents minor internet traffic, which also decreases the chances of getting caught. iii. It is also worth noting that students combine different services. For example, 43%
of those obtaining less than 10 songs a month in their previous country of residence and 27% of those downloading between 11 and 25 songs indicated a free streaming service as their preferred channel to obtain music. This choice can be due to different reasons. Since these platforms do not have contracts with some record companies or artist, consumers might try to obtain their music elsewhere. Moreover, having a digital copy gives a sense of ownership that the cloud does not, which could explain why some individuals still like to collect a few albums.
Table 3. Summary of demographic variables
Variable Mean Std. Dev. Min Max
Gender Fem = 0 Male= 1 .29 .46 0 1 Age 23,2 2.71 18 29 Income .76 .75 0 2
Prefers Illegal Platform No = 0 Yes = 1 .36 .48 0 1 Change in Download Decrease = 0 Constant = 1 Increase = 2 .56 .7 0 2 N = 59
Table 4. Amount of Songs Illegally Downloaded in Previous and Current Country of Residence by % of Respondents Amount of Songs PCR (%) CCR (%) 0 3,39 33,9 1-10 45,76 38,98 11-25 27,12 10,17 26-50 8,47 6,78 51-100 3,39 5,08 101-250 3,39 0 >250 8,47 5,08
13
iv. Still, 17% of respondents download more than 25 songs in their current country of residence.
4.3. CHANGE IN PIRACY BEHAVIOR
Once moving to a new country, the self-reported changes in downloading behavior were as follow:
i. 32% of respondents did not change their music consumption behavior when moving to their current country of residence, that is, they continue to obtain about the same amount of songs through piracy regardless of the change in their social and legal environment.
ii. While 56% of the sample showed a decrease in the amount of music obtained illegally after moving to a different country, a 12% started pirating more music in their new country of residence.
4.4. RISK
With respect to the risk perception and social norm, the differences shown by respondents are the following:
i. For 68% of respondents, the probability of getting caught for piracy in their previous country of residence is smaller than the one for shoplifting. A percentage that changed to 39% in the current country of residence.
ii. Comparing the believes about the two countries, 15% believes the probability of getting caught to be low in both countries they have lived in.
iii. 19% thinks that the probability is higher in the two of them.
iv. Only 13% thinks that there are more chances of getting caught in the country they are living in than in their previous place of living, a percentage that was expected to be higher.
4.5. SOCIAL NORM
14
i. For both countries, more than 50% of respondents said that music piracy is not discouraged by their peers
ii. Moreover, 36% thought it was discouraged by a few of their friends in their previous country of residence and 20% though it was the case in their current one. iii. Taken together, these results mean that for more than 80% of respondents, illegal downloading is not seen as something bad in their social circle, which gives an idea students aged between 18 and 28 have about this practice.
iv. It is also the case that a higher percentage of respondents believes more peers discourage music piracy in their current country of residence than in their previous one.
v. With respect to the difference between the two countries, 51% does not see a difference in the quantity of peers discouraging piracy, 24% thinks more of his peers in his current country discourage piracy, while 25% believes the opposite.
5 ANALYSIS AND RESULTS
Based on these responses to the survey, a multinomial logistic regression was specified to reveal the effects of risk and social norm on changing a person’s downloading behavior, which was measured categorically by the difference in the amount of songs downloaded in his current and previous country of residence (no change, increase and decrease). The results can be found on Table 6. The first model is the risk and controls only, the second one comprises the social norm variable and the controls, while model 3 is the full model. Each one of them is composed of two regressions, one for the group that decreased its amount of music downloaded after moving to their new country, and the other for the one that increased that amount, each one compared to the group that did not change its amount. Hence, the results must be interpreted
Table 5. Is the unauthorized downloading of music discouraged by your peers in your previous/current country of residence?
By none of them By a few of them By quite a lot of them By nearly all of them By all of them Previous Country of Residence 50,85% 35,59% 6,78% 5,08% 1,69% Current Country of Residence 57,63% 20,34% 13,56% 5,08% 3,39%
15
as the likelihood of increasing or decreasing illegal downloading relative to no changing the amount of songs obtained, which is the baseline group.
5.1. RISK
The indicator of risk compares the three combinations of perceived probabilities of getting caught in the previous and current country of residence relative to the baseline group, which in this case corresponds to the group composed of respondents that did not know this probability in at least one of the countries.
With respect to the students that believe the risk of getting caught to be high in both countries, they are less likely to decrease their amount of songs downloaded than to keep it constant (Exp(b)=.27). Similarly, they are 8.1 times more likely to increase the amount of songs they obtain illegally compared to the baseline group. This result is in line with previous literature that have found that prolific downloaders are more aware of the law and the risks associated with it, which can explain why they would be more likely to obtain music illegally compared to a group that does not know what the risks are (Cox & Collins, 2014). Moreover, even though they might believe the chances of getting caught to be high, they might also be aware of the low probability of apprehension due to the lack of effectiveness that legal actions have had in the past.
In addition, given that this predictor consists of perceived risk and not risk attitude, it is possible that students that are more familiar with music piracy are also the ones more willing to take such a risk. If for them piracy is a common practice, it might be harder to change. In fact, having an illegal service as the preferred way to access music (Illegal download control) has a significant effect on the likelihood of being in the group that believes the risk to be high in both countries. Once controlling for this variable, the prospect of being in the no change group increases. Lastly it is worth considering the tools that facilitate piracy, such as Virtual Private Networks (VPN) that render the chances of getting caught close to zero. Even though in a country the probability of getting caught might be high in general, they can still be dodged through such platforms allowing people to continue or even increase the amount of unauthorized songs obtained.
The other two groups, composed by the students that believe the probability of getting caught to be small in both countries or higher in their current one, show a similar pattern in the decrease model. Compared to the group that is unaware of the risks, they are 1.3 times more likely to decrease their amount of songs downloaded (Exp(b) = 1.3) than to keep it constant. The group
16
that does not know the risk is conformed mostly by students that download less than 10 songs monthly, while for the other two groups the data is distributed more evenly between the different ranges. In that sense, it can be the case that it is harder to change the amount of songs obtained when this amount is relatively low, than when it fluctuates among higher quantities downloaded.
Lastly, students that believe downloading music illegally is riskier in their current country of residence than in their previous one are 4 times more likely to increase their amount of songs obtained than to keep it constant. Similarly, to the respondents that believe the probability to be higher in both countries, most of the students that comprise this group downloaded more than 10 songs monthly in their previous country of residence, which reflects more ease at using piracy platforms. Hence, this result is in the same direction of other studies that have found consumers to be indifferent to the legal effects that piracy can carried, despite of being aware of the illegality that this action entails (Lysonski and Durvasula, 2008; Altschuller and Benbunan-Fich, 2009). That is, the more files people download, the more in the known they are of the risks associated with this practice.
5.2. SOCIAL NORM
The indicator of social norm is a continuous variable measuring the difference in the quantity of peers that discourage piracy between the current and previous country of living. It takes negative values if more friends discourage piracy in the previous country of residence relative to the new one, positive values in the opposite case, and zero if the amount is the same. The model shows results according to theory for the effect of social influence on behavior. Having more peers that discourage music piracy in the current place of living compared to the previous one is negatively associated with piracy levels. In that sense, if in the new country of living piracy represents less of a social norm or is not supported by the social environment, the illegal consumption of music will decrease.
In terms of the results, once moving to their new country, having peers that discourage music piracy more than those in the previous one makes people 1.27 times more likely to download less than to keep on downloading the same amount of music. At the same time, it makes people .95 times more likely to keep their amount downloaded constant rather than increasing it. An interesting aspect of this results is the higher coefficient for the decrease group, which indicates that the view of peers can not only stop a person from increasing his download behavior, but it
17
can have a higher impact on piracy by making someone reduce their amount of songs illegally obtained.
Just as music piracy is a learned behavior by imitating or learning from others, it can be mitigated by peers the other way around too. This result has important policy implications as the record industry focused for too long on legal measures to fight piracy with little focus on the channels that were making this practice grow and that were more difficult to eradicate.
6 GENERAL DISCUSSION
This exploratory study aimed at shading light on a rather unexplored aspect of music piracy: how this behavior changes when an individual is faced with an abrupt change in his social and legal environment. Although the results of this study are not conclusive, it is worth mentioning how well they adapt to the theoretical predictions.
Regarding risk, a sound prediction was that when people moved to a country with more strict laws against music piracy, they would decrease their piracy behavior out of fear of getting
18
caught. However, the results show that those that believed that getting caught was more likely in their current country of residence were also likely to increase their downloading rather than keeping it at bay.
Several reasons can be behind this unpredicted result. It can depend on how used people in the high-risk group are to obtain illegal music or how risk averse they are. It is probably the case that more prolific pirates are also more aware of the risks associated with illegal downloading but that does not impedes them from stopping. Music piracy might be a behavior difficult to change among hardcore consumers that are used to obtaining it for free or that know their way around computers. Since piracy is already dangerous, the fact that they are more prolific at it might also be an indicator of a higher risk attitude and moving to a different country might not be enough of a reason to go legal.
Regarding the methodology, for building this indicator students were asked to compare the probability of getting caught for piracy to the one for shoplifting, and even though some respondents might believe the chances for the latter to be lower, it might still not be enough of a discouragement to take the risk. As mention, it was also the case that the baseline group for the dependent variable, the no change group, was largely composed by respondents that downloaded between 1 and 10 songs, without much variance. Hence, they are students that incur in piracy in counted occasions.
There could be other factors influencing the decision to increase music piracy even though the risks are perceived as higher. It can have to do with a change in the daily routine or with accessing to the internet through a public/university connection which is harder to track. It can also be the case that since most of international students stay for 1 or 2 years only in their foreign country, the fact that they will leave soon might contribute to pirate more. After all, once they have left, future sanctions are unlikely.
The results for the effects of socialization on music piracy are more straightforward. Students that surround themselves by peers that discourage illegal downloading more, negatively affects their decision to pirate. As students tend to emulate the conduct of their close social group, it makes sense that they would download less when surrounding themselves by other that oppose piracy. However, the causal mechanism cannot be fully elucidated. It can be that people give in to social pressure or that they are convinced by their peers about the dangers and damages that music piracy involves. It can also be the case that they introduce them to alternatives, such as streaming platforms, which can also be a way of discouraging piracy. Additionally, these services allow users to interact, through playlists or recommendations, which increases their social component.
19
7 LIMITATIONS
A few remarks about this study need to be addressed. Moving to a new country implies a drastic change in people’s routine and social life. Students can be spending less time on computers or listening to music, while they are discovering and establishing in a new city, or they might end up hanging out with peers that are not as interested in music. In that sense, if this sample is to be compared to other studies run on university students, it must be done with attention to these other differences.
Music piracy is already a behavior that is hard to measure. On the one hand, people are not prone to disclose information about illegal or activities that are ethically-questionable. On the other hand, due to its illegality and the multiple forms it can use, it is also difficult to track. In the survey students had to answer, even though anonymity and confidentiality were reassured, the self-reported amount of songs downloaded remained low, with 50% of respondents saying they download 10 songs or less in their previous country of residence, and 72% in the current one. When asked about the amount of songs downloaded, those that admitted to an amount higher than 50 songs per month were only a few, figures that is questionable. Since music piracy is an illegal and unethical action, some respondents might have indicated a lower amount of songs downloaded although they admitted having incurred in these actions. A second option is that they only recur to piracy on counted circumstances and are therefore unconcerned about possible consequences.
In terms of methodology, the information was gathered once respondents have already moved from their previous country of residence. It can be the case that some individuals did not remember their previous behavior perfectly, that their used their current amount as a proxy for their previous one, or that the questions about their previous behavior biased the ones about their current one.
Finally, in order to collect a larger sample, group homogeneity was lost. The respondents were university students in different fields and levels of studies, from different countries of origin and currently living in different countries too. Asking about their perceptions instead of more objective measurements and circumscribing the sample to respondents currently living in Europe helped mitigating this effect, but the sample remained diverse. The final sample was still not large enough to draw significant results and some theoretically relevant and fundamental variables, such as gender, had to be left aside. Out of the 17 male participants, there were 0 cases in the group that increased consumption, 11 in the decrease one and 6 in the baseline model, which did not allow for coherent results. The same holds true for other
20
predictors that were considered in the original project, and that had to be left aside such as the expected punishment for getting caught and ethical attitudes towards piracy, measured for example, by asking respondents if they considered piracy to be the same as stealing.
8 CONCLUDING REMARKS AND FUTURE RESEARCH
Since the beginning of the mass use of the Internet music piracy has represented a major challenge to record labels by allowing people to obtain songs without having to pay for them and with little or no risk of facing any consequences. In the first decades of this fight, instead of adjusting to technological change, the music industry pushed for strict laws and punishments against those involved in this practice. However, these attempts proved rather useless, as piracy was difficult to track and laws hard to enforce. If this report is to contribute to the fight against piracy, it is through its findings on the effect of social norm on piracy behavior. This is a hard task from a public policy perspective since it has to do with private, social interactions, but programs that aim at raising awareness of why piracy is bad might prove more useful than more coercive ones.
As mentioned in the general discussion, a relevant rival to piracy these days are streaming platforms, which can have an effect in the model presented through socialization. Figure 1
21
presents the interest over time in terms of google searches for three channels to obtain digital music, torrents, a video to mp3 converter and Spotify. The searches for the first two illegal channels were increasing 10 years ago, but over time downloading music online has become more difficult. Programs such as Napster and Kazaa are a thing of the past, sites to share files like Megaupload have been banned and using torrent sites is too complicated for some, specially with the growth of streaming services. The new platforms allow for an easy access to an almost unlimited amount of music for a low price and it makes sense then, that in our sample some people do not know what the risks associated with music piracy are, since they do not need to worry about them anymore. Instead, searches for streaming services have been on the high for more than 10 years. Unexpectedly, the decreased experience by music piracy in the past few years has not come from stronger laws against piracy and more enforcement, but from reinventing the model and adjusting it to new technologies. With streaming platforms people have an unlimited amount of songs literally in the palm of their hands, with or without paying for it, and without having to infringe the law. If music piracy has been decreasing in recent times, it is not because it has been forbidden, but because streaming services has been presented as an easier and more simple solution. The option to obtain illegal files is still there, but it implies a series of efforts such as finding the song, obtaining it, a risk of malware, the legal risk, an effort to transfer it among devices and an extra one when trying to share it with friends; all complications that streaming services rapidly avoid.
In that sense, this study might be coming too late, as music piracy has lost its momentum and students are not turning to it in mass as they used to. Aside from considering the role of streaming platforms, future research should pay special attention to the group of prolific pirates and the reasons they have to still prefer illegal channels instead of legal and user-friendly ones.
22
9 REFERENCES
Ajzen, I. (2002). Perceived behavioral control, self-efficacy, locus of control, and the theory of planned behavior 1. Journal of applied social psychology, 32(4), 665-683.
Akers, R. L., & Jensen, G. F. (2006). The empirical status of social learning theory of crime and deviance: The past, present, and future. Taking stock: The status of criminological
theory, 15, 37-76.
Bellemare, Marc F. and Holmberg, Andrew M. (2010), The Determinants of Music Piracy in
a Sample of College. Available at
SSRN: https://ssrn.com/abstract=1633430 or http://dx.doi.org/10.2139/ssrn.1633430
Bhattacharjee, S., Gopal, R. D., & Sanders, G. L. (2003). Digital music and online sharing: software piracy 2.0?. Communications of the ACM, 46(7), 107-111.
Bhattacharjee, S., Gopal, R. D., Lertwachara, K., & Marsden, J. R. (2006). Impact of legal threats on online music sharing activity: An analysis of music industry legal actions. The
Journal of Law and Economics, 49(1), 91-114.
Borja, K., Dieringer, S., & Daw, J. (2015). The effect of music streaming services on music piracy among college students. Computers in Human Behavior, 45, 69-76.
Cesareo, L., & Pastore, A. (2014). Consumers’ attitude and behavior towards online music piracy and subscription-based services. Journal of Consumer Marketing, 31(6/7), 515-525. Chiou, J. S., Huang, C. Y., & Lee, H. H. (2005). The antecedents of music piracy attitudes and intentions. Journal of Business Ethics, 57(2), 161-174.
Cox, J., & Collins, A. (2014). Sailing in the same ship? Differences in factors motivating piracy of music and movie content. Journal of Behavioral and Experimental Economics, 50, 70-76. Dilmperi, A., King, T., & Dennis, C. (2017). Toward a Framework for Identifying Attitudes and Intentions to Music Acquisition from Legal and Illegal Channels. Psychology &
Marketing, 34(4), 428-447.
European Union Intellectual Property Office. (2016). The Economics Cost of IPR Infringement in the Recorded Music Industry. Available at: https://euipo.europa.eu/tunnel-
web/secure/webdav/guest/document_library/observatory/resources/research-and-studies/ip_infringement/study7/Music_industry_en.pdf
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction to
theory and research.
Huang, C. Y. (2005). File sharing as a form of music consumption. International Journal of
Electronic Commerce, 9(4), 37-55.
Ingram, J. R., & Hinduja, S. (2008). Neutralizing music piracy: An empirical examination. Deviant Behavior, 29(4), 334-366.
23
Kos Koklic, M., Kukar-Kinney, M., & Vida, I. (2016) Three-Level Mechanism of Consumer Digital Piracy: Development and Cross-Cultural Validation. Journal of Business Ethics 134:1, 15-27.
Lalović, G., Reardon, S. A., Vida, I., & Reardon, J. (2012). Consumer decision model of intellectual property theft in emerging markets. Organizations & Markets in Emerging
Economies, 3(1).
Lysonski, S., & Durvasula, S. (2008). Digital piracy of MP3s: consumer and ethical predispositions. Journal of Consumer Marketing, 25(3), 167-178.
McCorkle, D., Reardon, J., Dalenberg, D., Pryor, A., & Wicks, J. (2012). Purchase or pirate: A model of consumer intellectual property theft. Journal of Marketing Theory and
Practice, 20(1), 73-86.
Medlin, D., & Vannoy, S. (2015). Music Piracy Among College Students. Twenty-first
Americas Conference on Information Systems, Puerto Rico.
Office of the United States Trade Representative (2017). 2017 Special 301 Report. Available at:
https://ustr.gov/sites/default/files/301/2017%20Special%20301%20Report%20FINAL.PDF Pai, C. L., & Chie, B. T. (2017). Illegal downloading of pop music among university students in Taiwan: An agent-based model simulation. Journal of Consumer Behaviour, 16(3), 197-211. Pryor, A., Dalenberg, D., McCorkle, D., Reardon, J., & Wicks, J. (2008). Buy or burn?: Empirical tests of models of crime using data from a general population. The Social Science
Journal, 45(1), 95-106.
Rochelandet, Fabrice & Le Guel, Fabrice. (2005). P2P Music-Sharing Networks: Why Legal Fight Against Copiers May be Inefficient?. Review of Economic Research on Copyright Issues. 2. 10.2139/ssrn.810124.
Social Science Research Council. (2011). Media Piracy in Emerging Economies. Available at: http://piracy.ssrc.org.
Stryszowski, P. and D. Scorpecci (2009), Piracy of Digital Content, OECD Publishing, Paris. Available at: https://doi.org/10.1787/9789264065437-en.
Yang, Z., & Wang, J. (2015). Differential effects of social influence sources on self-reported music piracy. Decision Support Systems, 69, 70-81.
