We took a look at a study from March that may have played a role in selling SOPA and PIPA as a piece of essential legislation to pass. What we found were large holes in logic and gaps in seemingly essential research needed to prove its main points of SOPA being needed to pass for the good of society. So, we decided to go looking for other studies to find out what they’ve been saying over the years and figure out what the scholarly literature really says about filesharing.
When I looked at the Phoenix study last week, I was disappointed at how easy it was to poke holes through the paper. There were so many general assumptions in places where I expected there to be solid numbers and citations I hoped lead to solid numbers. Instead, those citations contained data that undermined the studies conclusions. It read more like an RIAA press release than a serious study on filesharing.
While, on the surface, this almost sounds like an indictment to the credibility of high end institutions, underneath, there’s actually a huge resource seemingly untapped full of valuable information within the scholarly community. In fact, when I went digging into these studies, I even found one study that was already covered by my colleague Chris clear back in 2004. So, my first impressions on all the studies I have gathered? Studies that seem to be more pro consumer is more the norm in the scholarly literature and not the exception. In fact, you can easily expect to hear some of the arguments expressed by seemingly random blog posters as well as see echos from the likes of Cory Doctorow. Yes, there is a science backing up these arguments against industry tactics that have been employed throughout the years.
Another observation I made was the surprising volume of studies. There were so many studies, the only way you could really review these studies is with a series. The total number of studies, including the study that was covered in 2004, was an even 20 studies spanning between 2004 and this year. We’ll see what we can do to cover all remaining 19 studies that may not have been either covered or widely known.
One question you might have right now is why are some of these studies seemingly unknown after all this time? There are a variety of possible reasons. One possible reason is that studies like this have a tendency to be locked behind pay walls, thus reducing the ability for the public to have access to them. Another possible reason is that a given study might not have had a publicity strategy that would have made it to the news. Another possible reason might be that the numbers and findings might not look favorably to organization like the RIAA, so they don’t get any sort of publicity backing by them and, thus, risk getting lost in the shuffle. Whatever the reason, we hope our little literature raiding operation will unearth some gems that haven’t been previously covered by anyone else. If they have been covered somewhere along the line, we hope it’s a nice reminder of the facts of the reality in today’s world.
Study in 2006 – Litigating Filesharers a Failure?
The first study we’ll be looking at is a study published in 2006 by The University of Chicago Law School. The study is entitled “Impact of Legal Threats on Online Music Sharing Activity: An Analysis of Music Industry Legal Actions”
A portion of the abstract points to what the study aimed to find at the time:
The focus now is on well-publicized legal threats and actions on a relatively small group of individuals to discourage overall music file sharing. To determine the resulting impact of these legal threats, we passively tracked online file-sharing behavior of over 2,000 individuals.
The study seemed quite meticulous in the data that was gathered. It broke down the process of file-sharing litigation into four, what the study called, events. Those events were:
- Announcement of Intention to Pursue Legal Actions (June 26, 2003)
- Lawsuits Filed against Alleged Music File Sharers (September 8, 2003)
- Court Ruling against Revealing Identities of Sharers (December 19, 2003)
- John Doe Lawsuits (January 21, 2004)
The study then offered an interesting point:
Earlier works have also explicitly incorporated constraints on resources (either time or monetary) that dictate that an individual solve an allocation problem—how much (time) to devote to legal versus illegal activities. One key difference in the environment we study is that such constraints do not naturally exist with online file sharing—participation in legal and illegal activities can take place simultaneously and can occur at large quantitative levels. A music consumer can purchase or listen to digitized music on an authorized retailer’s Web site and, at the same time, participate in illegal file sharing of the same or other music.
I would argue that this point blows a huge hole into an implied theory that people who download using unauthorized sources do so exclusively and that there are only two exclusive types of people – those who obtain content through unauthorized sources and those who obtain content through authorized sources. This has never been the reality and people have been known to download pirated material and obtain content through authorized sources at the same time. How many times have hardcore downloaders seen lines like “try before you buy”, or “If you like this stuff, buy it!” in the NFO of a given scene release? People in the filesharing scene, simply put, have been actively encouraged to purchase content legitimately for years.
Further down the study is the next section, the data. That section leads off with the following:
We developed an automated process to passively track sharing information from over 2,000 sharers on Kazaa, the most popular P2P file-sharing network at the time (Graham 2003). The process operates in the background, taking snapshot observations of the file-sharing activities of P2P participants. As no direct contact was established with the monitored individuals, the process provided no reason for individuals to alter their file-sharing behavior.
Keep in mind one thing here, this study was done a long time ago, so laws surrounding wiretapping were different in, say, 2003-2004 than they were now in 2012. Since this is also conducted on an open network, I don’t see how this is illegal in anyway. The study later explains the following:
The formal data collection started on the week of March 3, 2003, over 3 months prior to the first legal event. For analysis purposes, we report on data collected until the week of March 1, 2004, a date some 5 weeks after the final legal event.
This is what the study found:
In response to the RIAA’s initial announcement to pursue lawsuits, even though the sharers increased their file-sharing levels, they did lower their frequency of being online.
Even though the number of files shared exhibited an upward trend, this event actually represents a setback for RIAA in its legal strategy. While the frequency of being online did not exhibit a concomitant statistically significant increase in response to event 3, the frequency of usage levels did not drop.
While the preceding analysis suggests that a significant number of individuals altered their file-sharing behavior in response to legal threats from RIAA, the analysis does not indicate the magnitude of these shifts.
As a result of this, the studies authors did some further analysis and wound up finding that the litigation against file-sharing shares numerous similarities to the so-called “war on drugs” of the US government. Specifically, the study notes, “Drug enforcement agencies have targeted the supply side (large suppliers and large shipments—sharers) rather than the demand side (users—downloaders).”
The study then concludes with the following:
Our analysis indicates mixed success for RIAA’s strategy. On the positive side, before- and after-event comparisons suggest that over the course of the four events, the majority of substantial sharers decreased the number of files shared, typically by more than 90 percent. During this period, a majority of nonsubstantial sharers reduced sharing activity, typically to a third of their original levels. Further, a substantial number of sharers exhibited some risk mitigation behavior. On the other hand, some findings pose concern for the recording industry. We found an upsurge in the frequency of usage after event 3 from the sharers who continue to use the file-sharing network. These individuals are continuing to find value in accessing and using P2P networks.
While our results are consistent with the effect intended by RIAA, we feel it necessary to add the following caveat. It is possible that the observed reduction in file sharing on Kazaa may have been at least partially linked to a shift by sharers to other sharing networks. While we cannot rule this out, we do have information from another sharing network (WinMx)
In other words, the study found that while the targeted population may have reduced their activities, others simply picked up the slack and file-sharing kept going undeterred. When a whole network (in this case, Fast Track) was deemed unsafe by many, the users simply moved to another network.
The whole point of file-sharing lawsuits from the RIAA’s perspective at the time was to make a few high profile lawsuits happen and deter file-sharers – not just those targeted – from ever downloading ever again. That obviously never happened as we have years of general observations to go by here. It’s unclear how much the authors ever dreamed of their findings of file-sharers simply moving to other sources when something is determined to be too risky being a consistent theme for the six years after this study was published, but you have to give them credit for hitting the nail so squarely on the head on this one. I say that, given the initial goals of RIAA file-sharing lawsuits, this study alone showed that the litigation tactics failed spectacularly.