Study Says P2P Downloaders Buy More Music

An anonymous reader writes "Michael Geist posts to his site about a study commissioned by the Canadian government intended to look into the buying habits of music fans. What the study found is that 'there is a positive correlation between peer-to-peer downloading and CD purchasing.' The report is entitled The Impact of Music Downloads and P2P File-Sharing on the Purchase of Music: A Study For Industry Canada, and it was 'conducted collaboratively by two professors from the University of London, Industry Canada, and Decima Research, who surveyed over 2,000 Canadians on their music downloading and purchasing habits. The authors believe this is the first ever empirical study to employ representative microeconomic data.'"

Bias in the study?

[Guido del Confuso]

According to the study:

...among Canadians actually engaged in it, P2P file-sharing increases CD purchasing. We estimate that the effect of one additional P2P download per month is to increase music purchasing by 0.44 CDs per year.

However, it is important to remember that correlation does not equal causation. It seems just as probable, if not more so, that people who buy more CDs are more likely to engage in file sharing.

I find it curious that they would phrase their results in such a manner. From the data gathered in the study, I believe it is impossible to determine causation. To me, this throws their entire credibility into question.

Correlation != Causation

[johndiii]

It's not possible to do a controlled experiment in this context - to see if an otherwise similar group of individuals will buy more or fewer CDs if they do not have P2P access to music. So one cannot say whether or not such access reduces or enhances CD sales. It's quite plausible that the latter would happen, as a result of increasing immersion in the music culture, but it would seem to be very difficult to produce direct evidence.

However, this does reinforce the fairly obvious conclusion that the recording industry has chosen to use strongarm tactics on its best customers. It does not seem like the best of business models.

Re:Bias in the study?

[someone1234]

I agree that correlation does not equal causation.
But at least this study shows that people who download, will still buy CDs.
And people who don't buy CDs are less likely to use p2p too.
So, simply people who like music will get it, be it CD or P2P.

Well of course!

[Erpo]

there is a positive correlation between peer-to-peer downloading and CD purchasing.

Well of course. This study makes it perfectly clear that P2P downloading leads to CD purchasing, so P2P is obviously helping the music industry.

Wait a minute. Before P2P some people liked to buy a lot of CDs and some people didn't like to buy CDs at all. Those people who liked to buy a lot of CDs are now buying fewer CDs and downloading music illegally instead. Those people who didn't care much about music before are not downloading musically illegally because they don't want it very much. So P2P is obviously hurting the music industry.

Oh wait. I can come to two different opinions based on the same evidence depending on what mood I'm in and the people I listen to. Maybe I should recognize that it's totally possible to make a convincing argument for a statement that isn't true. Maybe I should re-evaluate some of the things I'm dead certain about.

Ignoring Causality

[keean]

Of course if we ignore what the causes are, and we believe this report, we are left with the fact that by going after P2P file sharers aggressively, the record industry is attacking its best customers... this does not seem sensible behaviour for any business.

Re:Bias in the study?

[Guido del Confuso]

I see what you're saying, but I'm not even sure how you could pose a question like that. At least, not one that would give you reliable information upon which to base such a conclusion.

I think what they've shown here is that P2P sharing does not decrease CD sales. That is, there is not a negative correlation. In fact, there may even be a positive correlation. But claiming that one causes the other strikes me as a politically biased conclusion. In other words, they set out to prove a position, and interpreted the facts in order to support that conclusion.

In addition, I think part of their hypothesis is flawed:

H2b. People who engage in music downloading and P2P file-sharing do so partly because they wish to hear a soundtrack or an artist before buying. Thus, there is a positive relationship between P2P file-sharing and music purchasing.

First, this is poorly worded. Are they asking whether everyone engages in P2P filesharing has a partial desire to preview music? Or rather that some P2P filesharers have such a desire? Second, the conclusion does not logically follow. Whether some people use P2P to preview music, it may or may not be enough to offset the number of potential people who use P2P in lieu of buying music. Therefore, proving that some people do use P2P to preview before buying does not prove that "there is a positive relationship between P2P file-sharing and music purchasing."

Again, I'm not disputing the statistical results of the study. I believe it is a reasonable conclusion that P2P file sharing does not have a significant negative impact on CD sales. What I am disputing is the conclusion that P2P sharing increases sales of CDs. It may or may not actually be true, but that fact simply does not appear to be supported by the evidence.

Statistician-speak

[joel.neely]

I believe that the language quoted is typical of statisticians talking about the data (think "graph") rather than the underlying observed system. In other words, I believe one should read it as short-hand for

The segment of the sample which downloads X + 1 times per month was observed to purchase 0.44 CDs per year more than the segment of the sample which downloads X times per month.

I believe that professional statisticians and researchers understand the difference between describing 'the effect" of moving around on the graph of results (correlation) versus claiming cause and effect in the underlying system.

However, quibbling over statistician-speak is irrelevant to the key point that people who were observed to download more music were also observed to buy more CDs. This result drives a stake in the heart of the RIAA argument that people download music instead of purchasing CDs.

seems rather obvious, and what should it prove?

[Racemaniac]

people who care for music are more likely to download and/or buy
i hardly download any music, but neither do i listen to it often, or buy it. i do download lots of anime, and i've also got a good manga and anime collection i bought :).
besides it being pure logic that people downloading more are also more likely to buy (you don't download crap you don't care about), i fail to see what it's supposed to prove. that downloading completely inhibits buying is obviously not true. the claims are rather that people downloading are buying less than they normally would, and that could still be true. it's not because they buy more than people who don't download (and don't care), that they're not buying less than they normally would.

Conclusion : Don't piss off your best customers.

[DrYak]

However, it is important to remember that correlation does not equal causation. It seems just as probable, if not more so, that people who buy more CDs are more likely to engage in file sharing.

The fact is : users who happen to download a lot, happen to buy a lot too.
No matter which causes which, there's an important conclusion to be drawn for media companies :

Stop harassing downloaders, because currently, you happen to be pissing off you best buyers.
Yes we know you **AA hate people who "illegaly steal" your stuff, but those people happen to be those who buy most of your CDs anyway, so be nice with them.

Re:Statistician-speak

[Guido del Confuso]

I just don't agree with you. The statement is pretty clear. If they wanted to say what you said, they could use phrases like "is correlated with". They did not. They wanted to show that P2P increases CD sales, and that's the claim they made.

Also, as I said before, this doesn't really prove anything. It could be that people who buy more CDs are just more likely to engage in file sharing because they are music lovers. Put in economic terms, P2P sharing and CDs can be considered substitute goods. People will choose either one, depending on which is more economically suitable at the time (taking into account such factors as a desire to own physical media, bonus material, cover art, etc.--in other words, I don't mean just the price).

Let's say there's a guy who lives in North Dakota, and it's constantly snowing. He gets sick of buying salt all the time, so he installs a de-icing system in his driveway. However, the de-icer is not 100% reliable, and it doesn't cover all the area he wants thawed. So he continues to buy salt in lesser amounts.

Compared to a guy in Florida, you could reach the same conclusion: People who install de-icers buy more salt than those who don't. That conclusion ignores the fact, however, that were it not for the de-icer, our North Dakotan friend would buy even MORE salt than he otherwise would. In any case, because he has a need or desire for thawing his driveway, he will certainly buy more than the guy in Florida. As I see it, the RIAA's argument is still valid.

I don't really agree with the RIAA's argument, but I don't think attacking the logic behind it is a constructive way of defeating it. I certainly don't think this study accomplishes that. I have the suspicion, however, that that's exactly what this study was designed to do. And that is why I doubt its methodology.

Well no shit!

[goldspider]

Compared to who? People who don't download music, including people who don't give a shit about it?

People who have a greater interest in music buy more of it than those who don't? God Almighty, I hope my taxes didn't pay for this "study".

It helps the non-RIAA labels more

[Andy_R]

The RIAA is well aware that p2p sells more CDs, their problem it it's often not their CDs.

My CD purchasing has vastly increased since I've been able to try before I buy using p2p... but I've mostly been discovering wonderful but tiny non RIAA labels, and unsigned bands who put out their own CDs, instead of blindly buying whatever lowest common denominator act the RIAA cartel is pushing with a recoupable advertising budget in the millions.

Without p2p, I'd never have risked buying a CD by Kattoo for example, but after a recommendation on OiNK, I bought all 3 Kattoo albums (hear them at http://www.myspace.com/kattoo [myspace.com] - stunning classical/IDM crossover music, but sales figures in 3 digits). I'm concentrating on obscure indie CDs not because it's not because I'm ethically opposed to the RIAA (even thought I am) but because I prefer it.

The truth is that the cartel only want people to buy their heavily hyped CDs, not CDs in general. It's not p2p's loss of revenue they have a problem with (they know p2p boosts CD sales), it's p2p's loosening the stranglehold they have on the market thats their problem with it.

The same goes for net radio, it's less susceptible to payola and features indie labels too much, that's why the RIAA want to tax it into oblivion.

(Disclaimer: I do have 1 on my own tracks on a compilation CD released on a non RIAA label myself, but I'm not slashvertising it here, go try that kattoo link instead, his stuff is amazing!).

When I buy CDs...

[JumperCable]

These days, about the only time I purchase music is when I see new band at one of the local bars. I purchase the CD if I enjoy hearing their music. I would not have purchased this CD otherwise. Nor would I have purchased any other CD in it's place. My excess funds tend to purchase investments.

The last time I bought a CD without seeing a band was several years ago before they started this whole 'kill internet radio' game. Once these hobbyists stopped spinning their tunes, due to the government backed racket set up to collect fees for playing, I stopped hearing music I enjoyed. So I stopped writing down band names & songs I liked. So I stopped purchasing their music. I would say there is a strong causation that the RIAA causes me not to buy music.

Not in my case

[nurb432]

Well, it USED to be that way, as being able to 'preview' would let me know i wanted to buy it. The risk of wasting 15 bucks with the style of music i listen to was great enough to prevent many purchases on sight only.

However, with the way they have been treating customers, and now knowing how little the artists get, in my case i stopped buying anything that is tied to the industry, and only buy indie music.

Correlation and causation

[Opportunist]

Now, there are more reasons why buying music and using P2P could be related. Someone who has a high interest in music will most likely have heard of P2P and will most likely also use it. Of course, someone who has a high interest in music buys more CDs than someone who doesn't (who, in turn, might not use P2P for the same reason, it's no interesting tool for him).

Whether P2P boosts CD sales won't be proved or disproved that way. What this study proves without a doubt, though, is that the strongest buyers of music are also the heaviest users of P2P. In other words, the content industry is getting on the nerves of those that are their best customers. People who don't use P2P also don't buy many CDs.

So suing those people is a lot like slaughtering the goose that lays the golden eggs.

Misleading summary, as usual

[billcopc]

This study does not say that P2P downloaders buy more music, but it underscores a commonly-known (but commercially ignored) fact: music lovers will get their music by any means necessary. P2P, mail-order, and the local record store; they're all equal players, and the price of an item is usually not the primary purchasing factor (unless said price is abnormally high). Convenience, in my opinion, is the primary factor. I'm a music lover (big time), and I hate the music industry... where did they go wrong ?

If I'm looking for something popular, chances are it will be all over P2P and I can get it in a matter of seconds. If I'm seeking a full album, or something less mainstream like an older release, Amazon might be my best bet. If I don't feel like buying online (and waiting for the mail), I'll stop by the mall on my way home from work. Either way, the moment I get home, the disc gets ripped to MP3 (SQ freaks, get off my lawn!). Every player I own is MP3, heck I still have my old MPTrip in a box somewhere, god bless that piece of shit!

The fact that the record store is my last resort says a lot about the industry. The concept of piling a ton of albums in a store is just dumber than dumb; it's like a warehouse, because you can't glean much information from the sealed package to help you in your purchase. The kid at checkout is little more than a cash jockey, he/she doesn't know shit about anything older than last week. Even Costco at least tries to demo the goods before you buy that big bland bulk box. Those listening stations with a half-dozen rap albums don't help either! Amazon has preview clips for a large number of albums. Vinyl stores will let you audition just about any record in stock, on a good set of headphones too - not the dollar-store junk they have at HMV or Music World.

I like the concept of iTunes, but it's wrapped in DRM and Apple's megalomania and I don't have an iPod, so to me it's more trouble than it's worth. I play most of my music in the car, on an MP3 deck that I've owned for years, and spent dozens of hours setting up and tuning for the tightest sound. If someone were to make a high-end iTunes-compatible car deck, it would be a step in the right direction (to me).

I know there are lots of smaller MP3 peddlers on the net, but I'm not after the indie stuff (sorry!), I want the big labels to grow a brain and offer the products I want to buy. Lucky for me, I'm into house music and Beatport is a godsend for that stuff... it's pricey at $1.99+ per track, but their model is great, you can preview almost every track, and download as a 320kbps MP3 or even uncompressed WAV for an extra dollar. Beatport is great, but they only cover house/techno. If someone would apply that model to mainstream music, I would be all over it.

The RIAA is killing themeselves

[webmaster404]

This just proves a simple fact, people won't buy music unless they have heard if for "free" one way or another, be it radio, someone else's MP3 player, internet radio, a YouTube video, a P2P download or as a secondary band at a concert. People don't just go out and buy a CD without not at least remotely liking one song and if a P2P download or even a YouTube video they will be more apt to get a CD by that band. Its not rocket (or computer) science, the P2P networks, and YouTube has replaced radio at least for the "unknown" bands that don't get played on major radio stations and its boosting their CD sales by a lot.

Re:Statistician-speak

[ThePromenader]

Yes, but don't get side-tracked by your own desire to express your misgivings with the study - did you read it? - there's no need to "interpret" their conclusions to determine their "stance" as you have done - the whole point is that they have no stance. They have collected data on the exchange of music (in all its formats and forms) - a unique pool of data - then categorised that data. Thus your "salt and de-icer" reasoning/comparison is not a good one. Their conclusions are based on that data, and not the other way around as you would imply.

It is written in plain language "The primary objective of this paper is to determine the effects of P2P file-sharing on purchases of CDs and electronically-delivered music tracks" in both the introduction and conclusion of the study. They did indeed determine the effects, and their data showed that... people who download music (for free) also buy more music. If they or anyone wants to add their own level of hypothetical "cause and effect" reasoning to that, they can, and this can subsequently be questioned (especially if not proven through further study), but you can't use that second level of reasoning to discredit the data itself.

I would agree that many of the study's passages - and the order in which they appear - are badly arranged (their conclusion does indeed seem in places to be presented as the reason for their study), but it is obvious, after reading, that they went about collecting their data in a completely scientific and unbiased way.

Re:Bias in the study?

[h4rm0ny]

I believe the relevant quote starts something like "He who fights monsters..."

Re:Bias in the study?

[Anonymous Coward]

A great answer to the problem, and no stealing involved. Nicely done.

Another news: people who listen to music buy it

[roman_mir]

So P2P users (who download music supposedly) buy more music?

Isn't it obvious? Those are people who listen to music. I don't listen to music, so I am not a P2P user and I don't buy music either.

Re:Bias in the study?

[Nossie]

That is illegal too if you ask the RIAA, at least in their eyes.

CDs are obsolete. Who cares how many are bought?

[presidenteloco]

Miraculous as they seemed in the 80s,
they are outclassed on a number of fronts by simple digital files,
as far as a consumer is concerned.
1. The digital file isn't tied to any particular physical object,
or player, or location. It's simpler. If I know part of its name,
I can be playing it a few seconds later.
2. The digital files can be more flexibly arranged in groups to different
tastes and purposes.
3. They can be stored on the Internet and communities of people
can review them, collate them very flexibly.
4. They don't encourage the production of cruft to fill extra
tracks on a CD album.

So why are we talking about CDs at all. That was so 80s.

The discussion should be how music artists should be compensated
in the post CD world.

I think Radiohead demonstrated the way forward.

The traditional music industry, by fighting an inevitable change,
is driving a stake into its own heart by guaranteeing its irrelevance.

Re:Bias in the study?

[Technician]

I agree that correlation does not equal causation.
If you want evidence, just take a look into my past. When I was in high school, I lived in the country. There were no FM stations. The local AM station carried Country and Western. I didn't buy much music. LP's and 8 tracks were popular along with brick size mono compact tape recorders with fidelity good enough for voice letters. I managed to collect a couple LP's and 8 track tapes to play with, but nothing serious.

After high school, I went into the Navy. Here I became exposed to lots of great music much like a typical middle schooler or high school kid does now. I invested in great equipment, bought the best blank tapes, had a good linear track turntable with moving coil cartridge, etc and a pair of quality solenoid operated cassette decks. I made mix tapes, traded tapes, and bought albums of my favorite artists to put on tape to play in my car. Artists included Pink Floyd, Styx, Queen, Tomita, etc. My peak piracy days was my peak purchasing days. Without the peer to peer dorm life, I would not have had the exposure and would not have bought nearly as much stuff.

Now I am married and have kids and grandkids. Any band that needs to curse or have a screamer is not my idea of music contrary to what my adopted kids like. Other than volume to the wall distorted by heavy compression junk, I don't have much exposure to new music anymore. Most of my exposure to great music is often called illegal. For example, I caught the fantastic light show last Christmas with the house with the synchronized lights. By any RIAA rulings, that publication and distribution of the the song Wizards in Winter was a violation of copyright. Trans Siberian Orchestra on the other hand made a hero out of the guy and gave him VIP treatment to one of their concerts. Was he a criminal guilty of massive online copyright infringement, or a creative artist using and promoting another artist? The only reason he wasn't prosecuted was because the backlash would have been severe and swift. The artist that doesn't understand this is the artist once known as prince. See what happened when someone put a short video of a toddler dancing? The artist didn't get it. TSO and the RIAA could have been in the same boat but much worse for that Christmas light show.

I went to buy the album, but with the current litigation, I am directly avoiding RIAA labels. Unfortunately that album is on an RIAA label. Sorry TSO.

http://www.riaaradar.com/search.asp [riaaradar.com] Search for Trans-Siberian

Peer to Peer is how people find out about new great bands. It's advertising.

The band is coming to my local area this fall. Unless they drop their label, I am not going to the concert.
If my dorm tape recordings could have resulted in the same $222,000.00 settlement, than I want no part in the industry that is suing their best customers.

In a nutshell, I am not legally exposed to new good music. Stuff on the radio is payola and off limits RIAA, ASCAP, BMI on my reject list. I don't buy music I don't know about. I do know about the litigation. I am voting against it as much a possible. Anybody tainted by it is someone to not do business with. I don't pirate it. I simply don't accept it.

Re:Bias in the study?

[Artifakt]

And it's proof that the RIAA is fundamentally irrational. They've spoken out against used CD stores, tried to get laws passed limiting or eliminating the normal 'right of first sale', even lobbied against libraries carrying recorded music. They've lobbied congress in the wake of 9-11 to get copyright infringement added to the list of terrorist activities under the USAPATRIOT act. This is not just something superficially plausible as a business model but unworkable in the details, it's genuinely crazy behavior.
        To take your car analogy farther, if some guy, just because he had an MBA, told Toyota they could eliminate the used car market by act of congress, and their sales of new cars would go up, prices would remain as high as they are, and the extra money required would be produced by the customers with no losses to the industry, they wouldn't work for Toyota anymore. When do the stockholders of the RIAA members get the same clue?
      Since recorded music is very far from a necessity, at least compared to cars or housing, how did this industry ever entertain the nut-bar idea they could eliminate used sales and not hurt themselves in the process? Used sales help prop up their price structure, yet the RIAA is treating them as another problem instead of an ally, just like they did radio, the cassette deck and DAT.