Home > Uncategorized > The Industry Canada Music Study, Part II, or Who Are You Going to Believe?

The Industry Canada Music Study, Part II, or Who Are You Going to Believe?

June 1, 2010

“Who are you going to believe, me or your own eyes?”
Chicolini (Chico Marx) in Duck Soup
(C) 1933 Paramount Pictures

As we saw in The Industry Canada Music Study, Part I, the bizarro “music study” conducted by Industry Canada reached the rather odd conclusion that file stealing is actually good for you. (This will come as news to the courts in Napster, Aimster, Kazaa, Grokster, Thomas, Tenenbaum, Isohunt and Limewire.) The study omitted that the sun rises in the West, the Earth is hollow and clapping on one and three means you got rhythm, but I guess you have to make some editorial choices.

So we will pick up in this Part II from where Part I left off–so I suggest you read Part I first if you haven’t already.

Cue Geist

I know it’s a shocking development, but guess who was the first to blog about the Industry Canada music study? That old speedreader, Michael Geist. (Michael Geist (aka “he who shall not be named,” according to a prominent Canadian artist. Geist is advisor to the U.S.-backed Samuelson-Glushko Canadian Internet Policy and Public Interest Clinic, the Alcan of IP with its almost 100% American board, and the paid consultant to Industry Canada. SG-CIPPIC’s board includes Lessig and the EFF legal director, so it shouldn’t shock anyone that Geist’s views bear a striking resemblance to the EFF’s own anti-worker positions.)

I guess I’m willing to take on faith that it would be possible to read the rather lengthy report, study the datasets, formulate thoughts and write them down all in a matter of hours. I don’t think I could accomplish that intellectual feat, but then I’m just a country lawyer and I’m not as smart as these city fellers. But then again, isn’t it equally likely that Geist got a copy of it beforehand out the back door of Industry Canada and had the time to read the report at his leisure? Particularly since the report sat around for so many months before it was released? Maybe the “insider” slipped it to him, our friend “SB” who posts on Geist’s blog. Who knows?

Geist throws down his support for the improbable conclusion in unequivocal terms:

“A new study commissioned by Industry Canada … finds what many have long suspected (though CRIA has denied) – there is a positive correlation between peer-to-peer downloading and CD purchasing. … [T]his is not a study with a particular desired outcome or sponsor – it is the government commissioning independent research to help make better policy decisions.”

This is a very interesting statement. Geist thinks that the music study “finds what many have long suspected.” “Many”? Who are this “many”? Perhaps the brain trust at the Berkman Center? But nobody who lives on Earth. You know, when sailing across an ocean, the thought has occurred to me around when bored silly around day 20 at sea that I was actually cast in a version of The Truman Show. But I wouldn’t exactly brag about it.

But the most interesting part of Geist’s post is the prophylactic statement defending the study’s bona fides. Geist feels the need to assert that the study does not have “a particular desired outcome or sponsor”.

Who said that it did?

Of all the things that a dispassionate critic could say, would that statement ever come up? How would he know that the government commissioned independent research? Who told him to bring that up?

As I described in Part I, it is hard to tell from the disclosed documents exactly how Industry Canada bureaucrats found the study’s author, Birgitte Andersen, in the first place. The disclosed materials pretty clearly indicate that some professor someplace was asked for recommendations, but because so much has been redacted about this subject, it’s hard to tell which professor. Redacting may be legally justified, but it does tend to create some long shadows. And it is hard to know who is operating in the shadows – so far.

One thing is clear– Geist found out about that study QUICKLY after its release. Another coincidentally well-timed leak? The talk around Ottawa? And he sure plowed through all that data quickly. Must be his econometric training. Wait…does he have any econometric training?


This all sounds a bit stinky to me, does it you? Not surprisingly, the music study was pretty much immediately dismissed by more experienced academics such as Dr. Stan Liebowitz (Director of the Center for Economic Analysis of Property Rights and Innovation at University of Texas) and Dr. George Barker (Director of the Centre for Law and Economics at the Australian National University and the President of the Australian Law and Economics Association). To give you a flavour of the reaction from the academy, Professor Liebowitz found the conclusions of the music study to be “not only implausible but…actually impossible to be true … The reason it is so hard to square the reported results with actual numbers is that the reported result is simply incorrect and the numbers in the survey are also incorrect.” (Liebowitz, “Copyright Issues, Copying and MP3 File-Sharing”, November 2007.)

Professor Barker came to largely the same conclusion:

“[T]he error in the report appears to be so serious as to completely undermine the conclusion it draws which renders much of the additional commentary and interpretation derived from it in the media (and blog columnists) quite misleading.…Given the serious and easily identifiable weaknesses in the paper outlined below, Industry Canada should accept responsibility for the consequent confusion caused by the results. We recommend the report be removed from circulation by Industry Canada pending its own independent review of the study. It is in the interests of Industry Canada’s reputation that this review be conducted by reputable researchers (e.g. the editors of a major economic journal such as Econometrica), and that the results of the review be published by Industry Canada…We recommend that greater care should be taken in the selection and subsequent publication of research that may have policy implications.” (Barker and Booth, A Review of “Impact of Music Downloads and P2P File-Sharing on the Purchase of Music: A Study for Industry Canada”, ANU Center for Law and Economics, Working Paper No. 2, November 2007.)

In a nutshell, these guys are making a very simple point—the music study makes a very simple error in assigning causation to a set of observed facts. Like if you said that we see people with umbrellas in the rain, therefore the umbrellas caused the rain.

Professor Hanel’s Critique

But it wasn’t just economists on the outside or experienced industry folk that criticized the music study. It appears that before releasing the study, Industry Canada had the study peer-reviewed by Professor Petr Hanel of the University of Sherbrooke–an actual economics professor. It seems that Professor Hanel found what I think can fairly be described as serious flaws in the study. He noted that “missing observations” in the data “seriously compromised the validity of the regressions” in the study. He says that the links between P2P activity and music sales (that the author seems to have tried so hard to show and Geist emphasized as the key finding of the study) “are not at all well established” in the actual data. Very plainly, Hanel says that the indicators that Andersen used to try to establish these links were “biased and inconsistent”. Professor Hanel’s work is a very candid and serious review of the music study as well as the literature, and a refreshing breath of air from the real world into the hothouse.

It is so refreshing, in fact, that it makes the reader wonder why Professor Hanel wasn’t commissioned to do the work in the first place. This is the message that seems to leap off each page of the detailed critique. And if Professor Hanel couldn’t have done the original work in the first place, why doesn’t Industry Canada publish his critique alongside the music study, thus offering two differing views of the data. At least one would be authored by a Canadian.

Professor Hanel suggested many, many changes questioning several of the study’s fundamental assumptions or methodology. Many of these changes would seem to require a do-over on the basic survey data on which the study is based. At this point it is difficult to ascertain how much of Professor Hanel’s suggestions made it into the final work product, but it does seem that Industry Canada itself got involved in revising its “independent” report. Judging by the email threads, Andersen seems out of time, out of patience and well past the work contemplated by her fee. And justifiably so, I might add.

On November 22, 2006, in a heavily redacted email she concludes with, “I try to simply do the changes you ask me to do, rather than going into long discussions.” Who can blame her? Not a penny more than $24,950, don’t forget—no matter how long the study takes to complete. People have lives they need to get on with, and eventually this thing has to stop. But the one message that comes through loud and clear to me after reading Professor Hanel’s work is that this guy is good–truly. Artists might actually have moved the ball if he’d been able to conduct this study. Another opportunity lost due to bureaucratic manoeuvring.

Office of Consumer Affairs Critique

In a heavily redacted email of April 30, 2007, the Canadian Office of Consumer Affairs also offers its observations on the music study. Almost the entire set of comments is redacted except for one line that questions whether students with limited income would have the ability to buy all the music that they download for free.


This is the kind of thing that the judges seem to think is “axiomatic”. Yet the music study did not take this axiomatic observation into account.

Let the Sun Shine In

So to sum up what we saw so far: Bureaucrats commission a study to prove that stealing music actually increases sales (which, by the way, is consistent with the views of Industry Canada’s go-to consultant, Michael Geist). They find a sympathetic academic to do their work for a bargain basement fee so that the work does not have to be put out to public tender. For inexplicable reasons, the bureaucrats commission a review of the study by Professor Henel — probably for the blessing of a Canadian economist who might have won the commission in the first place had it been properly tendered. No such luck. So the bureaucrats bury Professor Hanel’s devastating critique AND the study in the vaults and lie in the tall grass waiting for their chance to leak the study—a chance to embarrass their own Minister and the Canadian Government. Yet the critique remains buried. When the government finally plans to introduce copyright reform legislation in 2008, the study is quickly released to the public and Geist goes public with a post that is so detailed it’s hard to understand how he got there that fast without access to the unpublished study—which had then been laying around for quite some time.

Just a little too convenient not to attract attention.

The Sun Tries To Shine In

So it should come as no surprise that the disclosed documents show that on January 28, 2008, the Minister’s office contacted the Director of the IP Directorate (at the time, Susan Bincoletto), asking for an explanation and Bincoletto’s IP Directorate responds:

1. Minister’s Office Question: “[H]ow was the researcher and analyst selected?”

IP Directorate Response: “The contract to analyze the survey data was awarded to an economist with subject matter expertise in the relevant field of inquiry. The objectivity of the researcher was a selection criterion.”

“The objectivity of the researcher was a selection criterion.” What does that mean, exactly? It looks like it meant that there was a selection process with criteria, and that “objectivity” was a selection criterion. However, the IP Directorate does not provide the actual selection criteria. It seems that the only “objectivity” that was required was that the researcher not have worked for industry as opposed to an objective academic.

As noted in Part I, Andersen’s prior publications would probably lead one to the conclusion that she was something of a critical studies type and had a commitment to a particular point of view on these matters. She might be very good at being that type of academic, she might have a real talent for that kind of thing, but surely the prior work would not lead anyone to think she was objective. And to reiterate—there’s no evidence that anyone undertook any objective search for an author, much less more than one author.

The bureaucrats seemed to know exactly who they were looking for and had no interest in trying to find anyone else. And—they made sure they didn’t have to “tender” by paying fifty dollars less than the limit of $25,000 that would have required them to do so. This after originally promised her $25,000 to the dollar, but inexplicably backtracked by fifty bucks. Perhaps to avoid the $25,000 requirement for open-tendering and really wanting to make sure this contract was not tendered.

2. Minister’s Office Question: “I assume we have a public process to access bids for this kind of thing?”

IP Directorate Response: “Did not apply to the music file sharing study.”

Of course it didn’t apply to the music study—the bureaucrats made sure that it wouldn’t because they got Andersen to agree to take fifty dollars less than the $25,000 threshold as an a prior restraint. The truthful answer is yes, there is a public process unless you purposely try to avoid it—exactly what the Gomery Report was trying to avoid. Exactly what the Canadian regulations were designed to prevent. And it seems obvious that it was exactly what the schemers were trying to accomplish.

In fact, the “Draft Terms of Reference” credited to Martin Islam dated November 11, 2008, the handwritten marginal notes in paragraph 7 regarding “Deliverables” shows that the payment to Anderson was $25,000 in the first draft, but someone was struggling with the math in the marginal notes. Why? Because the note taker wanted to get the price for the contract down below $25,000 to the magic number of $24,950. In other words—below the tender limit.

3. Minister’s Office Question
: “Was this [study] something Minister Bernier wanted info about?”

IP Directorate Response
: “Minister Bernier did not ask for such a study.”

This question needed to be asked because quite obviously the then-current Minister, Jim Prentice, had not asked for this study to be undertaken. So here we have the political staff asking the bureaucrats if Minister Prentice’s predecessor, Minister Bernier, had commissioned the work product. Which he had not.

So if Bernier didn’t, and Prentice didn’t, who did?

4. Question from Minister’s Office
: “Is there no other readily available information on the topic – I mean is there a gap I needed to fill?”

IP Directorate Response: “While there is growing literature on music file sharing activity via P2P and its impact on pre-recorded music sales, none of the existing studies have analyzed Canadian data, let alone focused on the Canadian case. This is one of the first analyses of its kind to analyze Canadian data.”

Actually—apparently not exactly true. It appears that Industry Canada provided Andersen with several Canadian studies for her reference that apparently had already investigated file sharing in Canada. Even if you didn’t like the studies, it is simply not true that there were no Canadian studies. Wouldn’t the truthful answer have been that there are XYZ studies, but we thought there were RST problems with them? (like perhaps the conclusion?)Not surprisingly, in each of these studies, the findings were consistent with those of research conducted in numerous other countries, namely that access to free, illegal music over the Internet decreases music sales. Shocking, I know, but true.

Professor Liebowitz, in summing up the economic literature that was available prior to the Industry Canada music study, says (“File Sharing: Creative Destruction or Just Plain Destruction?”, Journal of Law and Economics, Vol. XLIX, April 2006):

“[T]he evidence here supports the current findings from almost all econometric studies that have been undertaken to date, including those in this issue – file-sharing has brought significant harm to the recording industry. The birth of file sharing and the very large decline in CD sales that immediately followed is a powerful piece of evidence on its own. The 2004 increase in CD sales, temporarily reversing the decline, largely matches a reversal in the amount of file-sharing activity. Furthermore, analysis of the various possible alternative explanations for the decline in CD sales fails to find any viable candidates. This conclusion … should not be much of a surprise. Common sense is, or should be, the handmaiden of economic analysis. When given the choice of free and convenient high-quality copies versus purchasing originals, is it really a surprise that a significant number of individuals will choose to substitute the free copy for the purchase?”

It Tolls for Thee

The disclosed documents tell a pretty sad tale that is straight out of 1984—these bureaucrats selectively picked a researcher straight out of the deconstructionist crowd who had no demonstrable experience with the industry or the phenomenon she was supposed to study. The bureaucrats made sure she would not have the financial resources anyone would need to do a proper job, because they didn’t want to have to tender the contract publically—and they all knew that was the drill.

Plus, they picked someone who was studying the Canadian market from thousands of miles away.

Then they released the study in a way that would embarrass their minister, and then they obfuscated further by equivocating.

And no one would ever have known if it were not for the disclosure of documents demonstrating what happened.

So the answer is that copyright policy at Industry Canada is being shaped by people with an agenda–with no ministerial oversight–and that agenda is designed to undermine and obfuscate the very artists that they are supposed to be protecting.

The question is—will they be allowed to get away with it a second time?

Why do we care about this Canadian study? Because every time one of these things happens, the results get into the global bloodstream. For example–there was talk of the study making its way to the OECD. If the OECD gets its hands on this kind of thing, then governments around the world start using these studies without asking the critical questions–which would have been very difficult for them to even know they needed to ask.

That’s why we should all care about what The Shadow knows.

%d bloggers like this: