- When assessing the P2P downloading population, there was “a strong positive relationship between P2P file sharing and CD purchasing. That is, among Canadians actually engaged in it, P2P file sharing increases CD purchases.” The study estimated that 12 additional P2P downloads per month increases music purchasing by 0.44 CDs per year.
- When viewed in the aggregate (ie. the entire Canadian population), there is no direct relationship between P2P file sharing and CD purchases in Canada. According to the study authors, “the analysis of the entire Canadian population does not uncover either a positive or negative relationship between the number of files downloaded from P2P networks and CDs purchased. That is, we find no direct evidence to suggest that the net effect of P2P file sharing on CD purchasing is either positive or negative for Canada as a whole.”
With CRIA and the Balanced Copyright for Canada group scheduled to appear before the Bill C-32 committee tomorrow, CRIA is releasing a new study it commissioned that tries to discredit the Industry Canada study and reach the opposite conclusion. The author is Australian professor George Barker, who has completed CRIA-commissioned work in the past. In the years since the original study, the authors of the original study have posted responses to some criticisms including this one in 2007 and a more detailed discussion from 2010 that counters the mud-slinging that has come from several sources.
Independence aside, one strives to please the hand that feeds it …
One has to wonder about studies that are backed by a group with a special interest. I am not privy to the motivations of the independent study sponsored by Industry Canada, but I assume it was less motivated by a polarized viewpoint.
and upon reading …
The CRIA backed study states that on average, if there were no P2P, that Canadians would spend $175 more per year on music. I cannot speak for others, but for myself I already spend as much as my budget can afford on entertainment (all forms). The music I do purchase I have discovered via services such as Pandora or independent shared artists (some via P2P). There may be some truth in their data mining, statistics are an art form after all, but I don’t think it reflects the reality of where todays music consumption and consumer expenditures are heading.
It’ll be interesting
to view the CRIA study, specifically to see what the assumptions and methodologies are.
Can I see people spending another $175 per year on music over and above what they currently do? No. What I suspect a better description would be is that the amount of money spent on works from CRIA represented artists would go up by $175, at the expense of the indie artists ($175 per year redirected rather than new spending).
This is entertaining ….
$175 a year per person on music. Talk about wishful thinking. The labels are competing against 5 million artists-bands on MySpace and faceBook, gaming, blogging, websurfing, texting, cellphones, CC music, social networking, youTube, e-mail, kindle and its ilk, expensive techno toys, all of which are competing for the same volume of time and same percentage of peoples money.
The share that is going to the record labels is shrinking and will continue to. Not because of infringement and p2p. But because they are competing in an ever growing “Entertainment” industry, that is no longer just music, tv, and movies.
Speaking of Statistics …
Stats can has just released the profit/costs of the music industry for 2009 (actual numbers, not doom and gloom soothsaying). Profit margins are in the healthy range and in most cases near the maximum average of the last decade, not bad for a year we experienced the 2nd worst economic depression in recent memory.
What I find really interesting though is the section for “Other sound recording industries”, I am going to make an assumption (open to correction) that this is independent and non traditional publishing, and if so it shows the biggest increases in percentage and total profits.
Interesting.
$175 / year
Of course, the finding that people used to spend more before p2p, doesn’t necessarily take into account that there are other services out there that help us reduce the amount we spend, such as itunes, why pay $15 for an album when we only want 1 or 2 songs off it. It’s a matter of coincidence vs causation. Is it just a coincidence that since the advent of p2p people are spending less or is p2p the only cause of this?
Crockett,
Notice how the StasCan data actually shows revenues and expenses dramatically decreasing along with salaries over the comparison period. That’s a nice cherry-picking of “profit” to come to your conclusion. I note that Professor Geist has done the same thing by tweating the profit conclusion only.
Looking at the whole picture, I see what looks like an industry in decline and real people losing their jobs.
Yep, those evil record companies — always twisting stats to suit their needs and pre-determined conclusions. Good thing we have ethical academics to set us straight.
An economics Prof teaching economics once said…
In the first class at university, that if you give 10 statistician the same data you can get 10 conflicting results. For decades I have always thought that was strange coming from an X-Minister of Finance that was teaching the course who was a former statistician. Anyway, I had to do an analysis a couple weeks ago on a large set of data and by goodness, I wrote three different algorithms and came out with three results that all contradicted each other, so I could basically pick which method would highlight the report the best (I actually disclosed and let them decide) but basically they took the more balanced approach that took in the widest number of variables. It showed a real view instead of one artificially inflated or deflated to prove a point.
The only way a study can be taken as relevant in this day and age is if the RAW data and the method of analysis is published for scrutiny. Otherwise, it is just like all the other numbers that are spoken referencing studies that are biased to say whatever the study sponsor wants to say.
Let them give the raw data and analysis method and let it be a real analysis and not just a let the numbers fit the designed result.
Degen
I’m not sure what Record production and integrated record production / distribution, notice record publishers refers to(is this CD’s). I mean the album is dead we can’t ignore the impact of singles sales over albus. Notice he publishers Salaries, wages and benefits are up over 40%. How is this even possible in this economy. What am I reading wrong here?
Also sound recording studios while relatively flat in the last year are up over 26% since 2006.
I think most Canadians would kill for these kinds of increases in the same period.
Open Source Government 🙂
Yo,
What’s the problem, I give my ISP 1500/year in service fees. I’m not gonna waste money on CD’s and DVD’s, no chance!
Not when in Korea you get 30/month for 1000x faster speeds then what we get here in Canada:
http://www.reddit.com/r/reddit.com/comments/fubkl/i_just_got_into_a_korean_hotel_i_heard_the/
If I could, I’d downvote bad laws not suiting Canadian people:
http://www.reddit.com/r/worldnews/comments/fpjdq/open_source_government_open_voting/
Business 101
@Degen “Looking at the whole picture, I see what looks like an industry in decline and real people losing their jobs.”
As I said before, statistics is more of an art than science, you can ‘glean’ from it what you want. For example here are some possible reads …
1) Record production and integrated record production: – Decrease in revenue, decrease in expenses, increase in profitability.
It looks like there is some industry streamlining going on here. No one can argue that the album is in decline as a format and distribution costs have decreased with digital delivery. Thus one would expect a decrease in labor and infrastructure leading, yes, to lost jobs (sad but not limited to the media sector). This is no different than workers on the assembly line being replaced by more efficient manufacturing processes and a changing market. The key factor here is the increase in profitability, isn’t that what most businesses and stock holders strive for?
2) Music publishers: – Increasing revenue, increasing expenses, flat profitability.
Music publishers do not have the same problems as the record producers as they are not as tied to a particular media format (album). In this case revenue has increased, which can only come from increased sales, along with associated expenses. This does not point to an industry in decline, rather one that is operating well in a depressed economy.
3) Sound recording studios: – On average, flat income, expenses & profitability.
This result is in line with the two above; that being a decrease in label albums and an increase in song publishing which leads to a neutral over-all impact on the recording studios. One must also factor into that the technology for artists to self produce becoming more affordable and attainable.
4) Other sound recording industries: – Increased profits, expenses & profitability.
Not sure what sector this category refers to but it seems to be doing, on the whole, quite well with salaries & wages doubling since 2006. This leads me to suspect we are dealing with emergent entrepreneurs leveraging new technology.
So there John, is my take. Feel free to post your analysis.
@Degen >Looking at the whole picture, I see what looks like an industry in decline and real people losing their jobs.
Heh so said the horse carriage builder, lumber jacks and anyone from any other industry displaced by something new.
With the amount of games and stupid cheap entertainment sites out there the entertainment industry to go first will be the most expensive to the consumers pocket books, which is music.
Real people can go and get another job. Their jobs were already displaced by jobs in other entertainment industries that popped up.
Remember that money not spend on music doesn’t just disappear from from circulation in Canada, if goes into other industries which innovated enough to catch those dollars from consumers.
$ony news
http://www.theregister.co.uk/2011/02/28/jailbreaker_defies_sony/
Nap.
Hilarity…
The CRIA should be the LAST ones to talk about copyright infringement of any kind, since they are dealing with a lawsuit against themselves for… you guessed it… copyright infringement. Biggest one in Canadian history, as well.
http://www.thestar.com/business/article/735096–geist-record-industry-faces-liability-over-infringement
Maybe they should make sure their poop doesn’t stink before picking up their holy grail, ya know?
.
Everytime the music industry comes up with some half nutted sponsored research it’s debunked. The only think this industry knows how to do is put spin on numbers and hope the rest of the world (which they are at war with) will turn their heads look the other way, and take their misguided facts as truths. Nothing can be further from the truth.
Looking forward to seeing this report and how this industry makes its case to comittee. Its sad though that Ms. Anderson one of the authors of the industry canada study hasn’t been called to debunk this pile of dog poo.
At least we have this
The REAL Death Of The Music Industry http://ht.ly/44aIN
Probably something similar to Bain & Company’s report headed to comittee, which will probably be filled with “half truths”.
…
@Torinir: “The CRIA should be the LAST ones to talk about copyright infringement of any kind, since they are dealing with a lawsuit against themselves for… you guessed it… copyright infringement. Biggest one in Canadian history, as well. ”
Yes we should have bigger statutory damages for commercial scale infringements. Something like $500,000 per infringement.
Nap.
…
Looks like one of the CRIA members is infringing on patents too:
http://www.reghardware.com/2011/03/01/playstation_ban_europe/
They should seize all $ony’s corporate computer and data storage devices while this case is investigated. Just like what they did to those 2 “hackers” infringing on $ony’s PS3 “IP”.
Nap.
Barker’s Previous Work
Okay, having gone to the ANU website and seen some of George Barker’s most recent past work on this subject, I suspect that Birgitte Andersen has little to worry about. Granted, it’s only a working paper; but even a working paper should be free of cherry-picked sources, unexplained assumptions, questionable assertions, and glaring omitted sources.
Case in point: The Barker piece was written in 2009. The Andersen study was published in 2007. Yet the Barker piece quotes a 2006 piece by Stan Leibowitz stating that “all econometric studies that have been undertaken to date…[indicate that] file sharing has brought significant harm to the recording industry” without mentioning the Andersen study. It’s a rather glaring omission.
Nor, in fact, does it refer to any of the work that Oberholzer-Gee has done on the subject, even to refute it. This might be an understandable omission, except that Barker extensively quotes Liebowitz, and any reader of Liebowitz would have been immediately and rather thoroughly exposed to Oberholzer-Gee’s studies.
(He also rather completely misuses the concept of ceteris paribus when discussing book prices, but that may have been Stan Liebowitz’s error, instead of Barker’s. And one citation is to “various government, NGO and academic papers”, which I’d think Barker would flunk an undergrad for using.)
Now, obviously, this is just a working paper. It may be that he’s planning to tighten things up at some point, so the nonsense about “various papers”. The problem of selectivity in citation, however, seems irreparable, and suggests that Barker’s CRIA study will warrant close, close scrutiny. At the very least.
Possible Source of Rebuttal
Also, it should be noted that Barker has written a previous response to Andersen—which further raises the question of Andersen’s omission in the 2009 piece—which raises questions of its own..
First, its summary of Andersen’s main points is not only breezy and somewhat dismissive, but does not appear to include any direct citation whatsoever. Again, this is a working paper, but since it was written in 2007, I think it’s safe to say that any editing has already happened that WOULD have happened. This raises the question of whether Barker is in the habit of knocking down strawmen. (On that, see below.)
Second, Barker says the following:
We understand that in Canada music industry revenues have fallen 47% (from 1999 to 2006) 8 since the launch of ubiquitous peer-to-peer file sharing on the Internet. If there were the positive correlation between CD sales and P2P downloads that the study assumes, we should have seen a marked increase in music recording revenues in this period.
This is just silly. Barker doesn’t need to “understand” anything. Either revenues have fallen, or they have not. His only source for that assertion is the CRIA, an interested party of dubious validity, but he doesn’t even provide a direct source for the assertion. The citation is simply “source: CRIA”. This defeats the very purpose of citation: independent confirmation of sources.
But more than that, this connection is completely invalid. It completely discounts the possibility of secular trends that might be reducing the number of CD sales that have nothing to do with downloading. An aging population, the decline of the vinyl replacement effect, increased competition from movies, television, video games and the Internet, quality of material, changes in the broader economy; none of these are even acknowledged as possible causes. Yes, Barker asserts that Stan Leibowitz supposedly proved that these aren’t factors; but once again, he provides no direct citation to Liebowitz proving it!
(Amusingly enough, Leibowitz’ own page links to a 2003 study that says that “MP3 downloads will annihilate the record industry. It, aha, appears to be a taking a bit longer than he had thought. Barker did not see fit to reference it, however.)
Thirdly and most disturbingly, though, it appears that Barker IS knocking down strawmen. Andersen goes into great detail about the market substitution and market creation effects that can help explain the behavioral foundation for the observed correlation between P2P usage and music purchases. Our host’s 2010 link features an excellent explanation and defense of same by Andersen herself.
But Barker discusses none of that. The phrases “Market creation” and “Market substitution” appear nowhere in the Barker piece. In fact, none of the behavioral foundations of the Andersen study appear anywhere in the Barker piece, or at least not as far as I can see. This strikes me as a MAJOR omission, especially in an article that is supposed to be about methodology and which attacks the Andersen study for supposed causation/correlation errors.
If this piece IS the foundation of the new study—and I suspect it is—then we could comfortably dismiss it out of hand. Though, honestly, I do hope that we get an Andersen response too. If it’s anything like the 2010 I mentioned, it’ll be epic.
More Barker Omissions
I’d forgotten about the 2007 piece that our host linked to. It also indicates a serious flaw in Barker’s work: he addressed Liebowitz’s various complaints, but in neither his 2007 nor 2009 effects did he address Andersen’s response. In fact, as far as I can tell, he hasn’t the faintest idea that it even exists.
Nor, judging by the press release, does he grasp the concept that the “hardcore downloaders who never purchase anything” may be a vanishingly small group. Since “hardcore downloaders” are also likely income-constrained teenagers and college students, using self-reportage of potential purchases is a dubious metric at best for proving his argument: while they may wish to do so, they may not have the means.
(The money they’re supposedly going to spend on music purchases has to come from somewhere, after all. If they don’t have it, they don’t have it.)
Also, the lumping together of “CD sales” and “digital download sales” in the CIPC release is very, very questionable. One of the reasons for P2P transfers that Andersen explored was the (understandable) desire of users to only pay for the songs they want; even if P2P disappeared tomorrow, CD sales would be unlikely to return to their former level simply because of the flexibility in song choice allowed for by digital downloads. It raises the question of where, exactly, that $175 figure came from, since the exact amount would depend entirely on the extent to which purchases are per-song or per-album.
In any case, we’ll know more when the study is published.
blog
Excellent .Thank you for your nice blog…………..
Compete with p2p?
What staggers me is that the big players in the music industry even two (?) decades after Napster first appeared, haven’t even attempted to compete and offer a cheap convenient alternative to p2p and downloading. Apple scooped them on the paid digital download because they couldn’t believe the glory days of tapes, LPs and CDs were gone. They suffer from a complete lack of foresight and deserve a slow death.
The only music lover that will ever suffer from this will be the top-40 listener. If you look at the indie or electronic scene, they take of themselves, thank you very much, without big label support. Nowadays, bands and DJs can produce high quality recordings without a fancy studio and a $100K advance from a record deal. Maybe this is exactly what we need. Music will be better when the big studios fade away.