How much evidence about the negative effects of online piracy can be ignored?

This week, the Canadian Intellectual Property Council released a report on The True Price of Peer to Peer File-Sharing. The report re-examined Don’t blame the P2P file-sharers: the impact of free music downloads on the purchase of music CDs in Canada, a 2007 report that cast doubts on the link between online piracy and music sales.

Using the same data from the 2007 report, analyst Dr. George Barker found the original conclusions to be incorrect.  Dr. Barker instead discovered:

1. three out of every four respondents said that if P2P were not available they would have purchased some or all of the music; and

2. almost two-thirds of the “hardcore” P2P downloaders (those who indicated in the survey that they acquired music by P2P only) said they would have purchased one-third of the tracks they downloaded if the songs were not available on P2P networks — this amounts to an average expense of $168 per person, adding up to hundreds of millions of dollars in extra revenue for the music industry per year from this group alone.

Based on this data, Dr. Barker concluded “that P2P downloads have strong negative effects on legitimate music purchases” and, contrary to the original analysis of the data, P2P downloading acts as a substitute for legitimate music purchases. Dr. Barker’s analysis infers that stronger copyright laws “would substantially increase music purchases and music industry sales revenues and, by implication, increase artist income and industry employment and contribute to both economic growth and higher government tax revenues in Canada.”

While it’s significant that a closer analysis of the original data reveals entirely different conclusions, this isn’t the first time the conclusions of the original report have been challenged. Noted economist Stan Liebowitz examined the original report’s conclusions when it first came out and found them “not only implausible” but actually “impossible to be true.”

Other studies independently confirm that reducing online piracy leads to increased sales.1 Taking a broader perspective, one economic study of stock prices found that “current and past efforts by the media industry to check illegal file-sharing over P2P networks through stricter copyright laws and lawsuits against violators have a significant positive impact on expected long-term profitability and economic viability of major media firms.2

The conclusions of the CIPC report — “people buy things they like unless they can get them for free — seem obvious enough not to need studying in the first place. But despite study after study and piles of evidence that show the harmful effects of online piracy, there are those who continue to insist that piracy is not a problem. The arguments are not that this approach or that approach to diminishing piracy is unwise for whatever reason, but that the harm doesn’t even exist in the first place.

Prove It

The academic consensus is that online piracy has had a significant negative effect on music sales. I listed a number of major studies on the issue in a previous post here. Olberholzer-Gee and Strumpf provide a similar table of major academic studies on the effect of online piracy in their paper File-sharing and Copyright (Table 5). Studies since then reveal similar findings; for example, Choi and Kim found that piracy has a negative effect on online music sales in Korea in a 2010 study.3 While a handful of studies have argued that online piracy has no effect, or even a positive effect, on music sales — most notably an earlier study by Olberholzer-Gee and Strumpf — these studies are in the minority.

The effect of online piracy on music sales around the world is easy to see. Countries which have improved legal responses to online piracy in the past few years, like South Korea and Sweden, have seen music sales rise. In contrast, countries like Spain and Brazil, where copyright laws are lax or ineffective against digital infringement, have seen music sales fall at a rate above the global average.4

And ineffective copyright enforcement around the world is not just a concern of major media industries. Jiarui Liu looks at the problem of piracy in emerging markets by examining the music industry in China. While it may seem that copyright enforcement in such countries is only a foreign problem, it turns out that rampant infringement has a profound effect on local artists:

In many cases, piracy of foreign works could be more devastating to domestic companies than to foreign companies. Because the competition from low-priced pirated works both online and offline undercuts stable income from royalties, Chinese musicians have witnessed the entire music industry becoming increasingly dependent on alternative revenue streams such as advertising, merchandizing, and live performance. The pressures of paid appearances and extended tours have started to squeeze the time that artists need to spend on music production. The alternative revenue streams also force many music companies to abandon traditional album contracts and operate in a way more like talent agencies that control all aspects of an artist’s career. Music companies are inclined to sign talents at a very young age with a long-term agency deal in order to exploit the full value of artists in the advertising market. In addition, the need to attract sponsorship opportunities puts more emphasis on non-musical qualities, such as a fresh appearance and healthy public image, which to some extent marginalizes “pure” musicians who have less value in those alternative markets.

Most importantly, as copyright piracy obstructs the communication of consumer preferences to musicians, an increasing number of musical works are created to accommodate the tastes of entrepreneurs (e.g., sponsors and advertisers) rather than those of average consumers, and this has caused a fundamental shift in the creative process of the Chinese music industry. Although entrepreneurs should arguably be willing to take whatever is popular among music fans as a draw to their own products, the expectations of entrepreneurs and consumers do not always meet in a dynamic market setting. For this reason, the interests of less commercial artists and new artists are more likely to be compromised.5

Commitment to effective intellectual property laws has ripple effects in a nation’s economy beyond the media industries. On Tuesday, attorney Lawrence J. Siskind, writing in The Recorder, asked Has Israel’s Approach to IP Law It Strong? In the article, he attributes some of Israel’s national growth over the past several decades to strengthening its approach to intellectual property. He quotes one report that says, “Israel, by 1990 was still mostly barren of technology and finance. … Israel generated few significant companies or technologies, no significant financial institutions, and little important science.” The country was known for wide-scale piracy at the time. But this changed by the turn of the new century (in part due to trade pressures from the US). “As Israel’s image morphed from IP pirate to IP protector, the country became a magnet for global investment.”

Courts, which are in the business of judging evidentiary claims, recognize the harms of online piracy. For example, the court in A&M Records v. Napster was convinced that “Napster use is likely to reduce CD purchases by college students.” Napster’s expert did not “provide credible evidence that music file-sharing on Napster stimulates more CD sales than it displaces.” Napster also tried to argue that the “sampling” of music its service provided (try-before-you-buy) stimulates retail music sales. The court was still not persuaded, calling the evidence used to support this argument “unreliable” and accepting “that the activity defendant calls sampling actually decreases retail sales of their music.”6

Yet, some still find ways to rationalize away the conclusions revealed above.

Yes, But…

These ways include focusing a lot of attention on less-than reliable studies or distorting the literature that is out there.

As mentioned above, one of the favorite studies of the piracy-is-not-a-problem contingent is Olberholzer-Gee & Strumpf’s 2007 paper, The Effect of File Sharing on Record Sales: An Empirical Analysis. But the study looked only at a small period of time (17 weeks at the end of 2002) and uses some curious methodologies (“Our most important instrument  is the number of German secondary school kids who are on vacation in a given week”). Stan Liebowitz has been especially critical of the study, and has published several responses worth a read here, here, here, and here. But the biggest surprise comes from Olberholzer-Gee and Strumpf themselves, who backed off from their earlier claims in a 2010 paper.

Another report widely cited as evidence against the harm of piracy is the US Government Accountability Office’s April 2010 Observations on Efforts to Quantify the Economic Effects of Counterfeit and Pirated Goods. After the release of this report, many declared the debate over the effects of piracy over — most notably a number of interest groups in a full page ad that said “Content Industry Piracy Claims are Bogus.”7 But is that really what the report said?

No. Far from looking at all or even most piracy claims, the report only examined a small percentage of independent reports — and even then, it merely highlighted some of the limitations inherent in those reports. It’s most damning conclusions were aimed at only three government agency figures that have been used — an FBI estimate used in 2002, a CBP estimate also used in a 2002 memo, and an estimate used by the Motor and Equipment Manufacturers Association in 2005 attributed to the FTC. The GAO’s conclusion? It couldn’t substantiate these estimates. The broader takeaway of the GAO report reflects the consensus: piracy has significant negative effects, but it’s difficult to nail down a specific number or dollar figure on the harm caused.

A particularly telling example of distortion can be seen in an article on Torrentfreak, Pirates are the Music Industry’s Most Valuable Customers. The site draws that conclusion from an IFPI commissioned survey carried out by Jupiter Research. Torrentfreak highlights the survey’s results that, compared to music buyers, music pirates are “31% more likely to buy single tracks online, 33% more likely to buy music albums online, 100% more likely to pay for music subscription services, and 60% more likely to pay for music on mobile phone.” That sounds like a big deal until you look at the actual report. There, you’ll see that Torrentfreak played a little numbers game — the percentage of either group of participants to engage in any of these activities is small, so that even a few percentage points difference results in an impressive-sounding “more likely” figure. For example, only 2% of “music buyers” paid for a music subscription service compare to 4% of “music pirates.” Not as impressive when you put it that way.

While that kind of statistical shenanigary might be overlooked, what’s truly misleading is that if you look at the total figures provided by the report, the total music spending by “music pirates” is less than the total of “music buyers” (and every other segment surveyed).

Well What About…

Maybe there are causes besides piracy of decreasing music sales, the final argument goes.

Not so, according to Stan Liebowitz, who has examined data over the past 30 years to arrive at his conclusion. While some factors may have played some role, that role is at best secondary and at the very least only negligible. Nothing else — the general economy, shifts in consumer spending to other forms of entertainment, etc. — has caused as big an impact as the advent of online piracy. The data doesn’t support the idea, and neither do economic theories.8

Is it possible that piracy has positive benefits? The GAO report considered this question, as have several other studies. Maybe network effects increase the value of copyrighted works. Maybe the “sampling” of digital files that file-sharing allows lead to more purchases.

Again, the consensus among researchers is that any possible positive benefits of piracy are far outweighed by the negative effects.9 Going back to Liebowitz, “sampling” is more likely to decrease sales rather than increase them. This seems like common sense if you think about it: if you can “sample” an entire work, over and over again, whenever and wherever you want, why would you ever buy it? Never mind that if piracy’s positive benefits were greater than the negative effects, we wouldn’t be having this debate, since creators and media industries would be doing better rather than worse after a decade of file-sharing.

Most creators don’t need to be convinced that piracy is a problem — the real question is how best to deal with it, whether that involves legal, technological, or business model solutions (or, most likely, some combination of the three). But despite all the evidence about the harms of online piracy, there will still be those who keep believing it is not harming creators and the media industries that invest in them. New research and studies will continue to be ignored or distorted. Alternative theories about piracy will be made that may sound nice, but without evidence to substantiate them, they are mere speculation.


  1. David Blackburn, On-line Piracy and Recorded Music Sales, Working Paper, Harvard University (2004), finding 30% reduction in files available online leads to 10% increase of industry sales in 2003; Kristina Groennings, An Analysis of the Recording Industry’s Litigation Strategy Against Direct Infringers, 7 Vanderbilt Journal of Entertainment & Technology Law 389 (2005), finding RIAA’s litigation strategy led to initial decrease in P2P filesharing and increase in album sales in 2003 and 2004 after several years of decline. []
  2. Sanjay Goel, Paul Miesing, Uday Chandra, The Impact of Illegal Peer-to-Peer File Sharing on the Media Industry, 52 California Management Review 6 (2010). []
  3. Dongook Choi & Yeonbae Kim, Effects of Piracy and Digital Rights Management on the Online Music Market in Korea, TEMEP Discussion Paper No. 2010:72 (December 2010). []
  4. See IFPI Digital Music Report 2011 and IFPI Recording Industry in Numbers 2010. []
  5. The Tough Reality of Copyright Piracy: A Case Study of the Music Industry in China, 27 Cardozo Arts & Entertainment Law Journal 621 (2010). []
  6. 114 F.Supp.2d 896, 910-914 (ND Cali 2000). []
  7. Other examples include an article in Ars Technica titled “US government finally admits most piracy estimates are bogus.” []
  8. Liebowitz’s own site summarizes his research in file-sharing and links to his papers on the subject. []
  9. Note that these questions of “benefits” are mostly limited to economic benefits. Not addressed in this discussion are the equally, if not more, important personal considerations of creators: their free speech interests and the chilling effect of piracy, to name just two. []


  1. Wow, Terry. Excellent article. It’s amazing how much debate has had to take place to establish something that simple economic sense should tell us is obviously true.

    Thanks for distilling it so well. Though I’m sure, as always, for many, it will fall on deaf ears.

  2. Pingback: Blisstunes | Blog | How Much More Evidence

  3. Yeah- The big pipe internet money is out there funding and promoting fringe people and numbers since legislation will eventually come to their doors. ( 17-24% of traffic on ISP’s are illegal file sharers or seekers)
    It turns out that MANY of these – copyright reform( their language) orgs and sites are funded in part or whole by the big tech money.
    Kind of like when the oil companies were trying to tell us that Elk and Salmon would actually increase if there was unlimited drilling.

  4. Pingback: With Many Thanks, O Noble Patrons | A Distant Soil by Colleen Doran

  5. *sigh*

    The problem with surveys is that people lie. And reading through just the first six pages of the IFPI is their foregone conclusion that P2P is to blame for being available to people. They directly asked these people if P2P wasn’t available, would they still buy music? 33% said yes. And yet… How are they supposed to do that? What does the music landscape look like? What is available from 2005 to now, that could increase sales? Hint: the music landscape is pretty sparse

    … (reading all 27 pages)

    “Economics predicts the decisions whether to acquire
    music illegally (participation decision) and how much
    (activity decision) will depend on access to relevant
    resources including computers and broadband. It
    will also depend on the legal sanctions for piracy and
    preferences or attitudes to criminal behaviour.”

    HA! That’s certainly not biased…

    “Pi = p*s
    Pi = expected sanction (price of illegal downloading
    or piracy)
    p = probability of being caught
    s = the sanction”

    Nowhere does it talk about legal availability of a good or anything else… And what is it about a consumer surplus over a good that seems to be such a taboo subject?

    [tangent]Reading this, I’m reminded of the Dictator Experiment:
    It shows one side of an experiment without really going into all parts. At first, the experiment seems to show the altruistic side of people. But the devil is in the details. The Foundations of Human Sociality shows that people will give. The experiments ran and showed that people would give every time. Enter John List who showed the fallacy of the research.

    What I’m critical about is the selection bias that’s inherent in just the first link. Like John List, if you have someone scrutinizing your actions, it’s going to weigh the data considerably in the wrong direction. Merely asking people “would you download if only legal options were available”, and someone taking your answer is NOT the way you can prove the data. Unfortunately, most of John List’s data was from interviews but I’ll put down some of the data here.

    When running the Dictator experiment changed variables:
    Classic game – Decide to give people some, all, or none of your cash. Result – 70% of people gave some money to another and the average donation was ~ 25%

    2nd version – variable the same, but you can take $1 from second person. Result- 35% gave in second experiment, 45% didn’t give a penny and the remaining 20% took $1 from second person.

    3rd version – Person A given same amount of money as person B. Options – Person A can steal B’s paycheck or give B a portion of her own money. Result – 10% give B money, 60% took from Person B and more than 40% took all of Person B’s money.

    4th version – Identical to the third but one twist – Both players had to work for the money. Now, only 28% took from Person B. 2/3 of the people neither gave nor took a penny.[/tangent]

    What the very first link shows is a selectional bias that I doubt anyone notices. You have a questionnaire and that’s it. What the cross sectional analysis shows, I doubt is the entire procedure. Most people try music before they buy, that’s been true since Napster. What isn’t coming up is what other music purchases people are doing instead? The entire project assumes that the ones that get free music would rather have spent their money on CDs or downloading elsewhere.

    And this:

    The Industry Canada commissioned 2005 survey thus clearly supports the view that stronger copyright laws that effectively reduce and deter free P2P music file-sharing would tend to increase music purchasing and music industry sales and, by implication, increase artist revenues and industry employment and contribute to both economic growth and higher government tax revenues. Whereas weaker copyright laws reduce music purchases, music industry sales, artist revenues, industry employment, GDP and government tax revenues.

    Is easily refuted. I won’t doubt that some people stopped purchasing music and now have the music that they want on mp3. But this absolutely ignores transformative use of music, and the music industry in general. Yes, you can tax the daylights out of an mp3 and create a black market. But nowadays, there’s more ways to make money, which this report absolutely ignores. Could it be that some of these people have done other things with their money, say… College students and concerts or kids buying guitars, that are really hard to track. And of course, I’d really love to see the music industry saying that the money they’ve gotten has increased. That’d really put this report to bed.

    • “The GAO’s conclusion? It couldn’t substantiate these estimates. The broader takeaway of the GAO report reflects the consensus: piracy has significant negative effects, but it’s difficult to nail down a specific number or dollar figure on the harm caused.”

      The GAO looked at at least three things in its report:
      Methodology of piracy

      Of those three things, if you add in the effects of destroying the goods of counterfeiting, you have quite a substantial number. This isn’t just medicines or hand bags that are being destroyed, it’s also generics that come into the US and are cheaper than the originals. Those cost… $19 million street value? And I know that to destroy those goods costs quite a lot.

      Still, we’re talking about piracy, so here’s the most relevant piece:

      There are also certain instances when IP rights holders in some industries might experience potentially positive effects from the knowing consumption of pirated or counterfeit goods. For example, consumers may use pirated goods to “sample” music, movies, software, or electronic games before purchasing legitimate copies, which may lead to increased sales of legitimate goods. In addition, industries with products that are characterized by large “switching costs,” may also benefit from piracy due to lock-in effects. For example, some experts we spoke with and literature we reviewed discussed how consumers after being introduced to the pirated version might get locked into new legitimate software because of large switching costs, such as a steep learning curve, reluctance to switch to new products, and search costs incurred by consumers to identify a new product to use.

      Some authors have argued that companies that experience revenue losses in one line of business—such as movies—may also increase revenues in related or complementary businesses due to increased brand awareness. For instance, companies may experience increased revenues due to the sales of merchandise that are based on movie characters whose popularity is enhanced by sales of pirated movies. One expert also observed that some industries may experience an increase in demand for their products because of piracy in other industries. This expert identified Internet infrastructure manufacturers (e.g., companies that make routers) as possible beneficiaries of digital piracy, because of the bandwidth demands related to the transfer of pirated digital content. While competitive pressure to keep one step ahead of counterfeiters may spur innovation in some cases, some of this innovation may be oriented toward anticounterfeiting and antipiracy efforts, rather than enhancing the product for consumers. – Page 18

      Moral of the story? We can’t trust the numbers in the game, but there’s better things to do than say piracy kills innovation all in a negative manner. Even though pirates are people too, there continue to be other ways to make money that don’t require litigation for egregious amounts (US copyright law -> $750- $150,000 maximum for statutory damages), having a PRO sue music venues, or overall, making the experience of entertainment more of a chore than doing its purpose of enjoyment for people.

  6. Pingback: ICE Seizures Criticism: Magic Words | Copyhype

  7. This is all pretty much irrelevant. No-one is seemingly even asking the correct question, so the ‘answers’ are worth almost nothing.

    The correct question follows trivially from the accepted economic structure of IP: is there an optimal trade-off between production and access?

    The purpose of IP is not to pay holders the most amount of money — more like the opposite. It is to supply the public with goods efficiently. If more of the public are getting these non-rival goods, for free, that is not a loss but a *gain*. And if these goods are still being produced healthily, reducing the payments to producers would have been the *right* thing to happen.

    This is not controversial, it is the normal orthodox view, and the most basic point in any informed addressing of the subject. But isn’t it remarkable how the real question is somehow forgotten, when it might lead to an inconvenient — i.e. less market protection — answer for the few loud-mouthed corporations concerned?

    And it does not take research to get the basic picture. Is there a terrible collapse in production of music, films, books — commensurate with the cataclysmic increase in ‘piracy’ so torturing these industries? No, pretty obviously not. Look around; do you have so much less music, films, books to enjoy than 10 years ago? It is a ridiculous notion.

  8. I don’t think it really matters whether piracy causes music purchases to go up or down. The purpose of copyright, and therefore its enforcement, is “To promote the Progress of Science and useful Arts”. Show us evidence that the public is being hurt (i.e. that progress is being affected), and maybe I’ll care. Otherwise you just sound like a whiny artist.