April 26, 2024

New Study of the Net

Eric Boorstin, a senior at Princeton, just filed his senior thesis, Music Sales in the Age of File Sharing. The thesis includes a clever study of the impact of Internet usage on CD sales. This is a twist on previous studies, which have tried to correlate CD sales to usage of filesharing. The tradeoff here is that although Internet usage is one step removed from filesharing, the data on Internet usage are much more detailed and much more reliable than the data on filesharing usage.

Eric worked from two datasets. The first dataset came from SoundScan, and gave him aggregate sales of CDs, on a week-by-week basis, for many separate metropolitan areas in the U.S. The second dataset came from the U.S. Census Bureau, and contained data on population, income, and Internet usage, broken down by age group and geographic area. The census data came from 1998, 2000, and 2001. Combining these datasets, he ended up with data for CD sales, age group demographics, income, and Internet adoption, at three different points in time, in ninety-nine separate metropolitan areas in the U.S.

Eric took these datasets and did a regression to determine the correlation between Internet adoption rate and CD sales, broken down by age group. He controlled for differences in personal income. (For more methodological details, see the thesis.)

For people in the 15-24 age group, he found a significant negative correlation between Internet adoption and CD sales. For people in all of the age groups older than 25, he found the opposite

Comments

  1. Music is Hell

    The Recording Industry Association of America’s shrill crusade to end music piracy is making in-roads in Congress. But is the industry’s war of lawsuits and intimidation destined to become a quagmire? Despite Wired magazine’s report that CD and album s…

  2. Cypherpunk,

    My reasons were:
    1 – Many car players and many home players will not play burned CDs. They won’t play the DRMd discs either. The majority of OEM car players can only play greenbook discs.
    2 – I’ve had hard drives fail, I’ve seen burned CDs become unreadable after time. One of my oddest discs I bought at a concert was from an Alabama Surfer band, a burnt disc, but it became unreadable after about 2 years.
    3 – If I like some songs from band X, I will probably like more, so I get the CD to find out.
    4 – Purchasing the album theoretically ends up supporting the band, not a whole lot if you ever get around to studying how mechanical royalties are used to stiff bands.
    5 – The manufactured CDs tend to last forever with little care or maintainence.
    6 – In the past, some co workers were from Columbia and Equador, and the albums they downloaded were frequently things unavailable at any price in the US. So to get the albums, I had to hand them a shopping list when they flew home to visit family (or family came to visit).
    7 – Watching the DivX fiasco was a huge insight to how the minds of the DRM creators/proponents work. If your DivX player broke, you had to buy your movies all over again even though the discs never left your house, only the broken player. I will not be trapped into any scheme locking something I purchased to a single device. I don’t have to pay for a book all over again when I take it into the bathroom. If I do read it there, the book does not stop working in the living room.

    So to your altruism/charity charge, please add self interest/safety/security.

    I consider the net to be an equivalent distribution channel to the radio. Finding the stuff back then was a chore, now it is about impossible.

    I think I have rather eclectic tastes, ranging from classical to jazz to folk to techno to ethnic stuff. Its not hard to spend $1k and get >95% of what is played on NPR’s classical repetoire. Just how many CDs of Holst’s Planets do I really need? Dominico Scarlatti wrote about 550 sonatas, I can buy the sheet music for all of them, but I cannot get recordings of all of them. Between 13 or 14 albums there are 50 unique ones and about 13-14 renditions of K.380.

    Before I moved from South Florida, I would have to get in my car and drive a half hour north to get in range of 1 station that played techno only on Friday evenings. Or get in my car and drive 15-30 min in the other direction to get in range of the only station playing Prarie Home Companion on Saturday.

    I don’t mind dropping $15 or $20 on a CD if it is good music. That is merely cutting out 1 or 2 pizzas a week. It is not a hardship. It is hard to find new stuff worth buying. That appears to be the point the industry cannot or will not get.

  3. Cypherpunk,

    I buy the CD’s for a few different reasons.

    – Quality: I like to make my own rips. I’ve got about 40gb of music I’ve ripped myself, in the format I like, at the bitrate I like, which I can play on the device of my choice. (I refuse to buy DRMed music thanks to the restrictions). Big value to me. If the industry steps up the crippled/copy protected/DRMed CDs, I’ll stop buying them immediately. (Don’t buy ebooks for the same reason).

    – Ethical: It’s just the right thing to do. If I really like the artist, I want to support him/her/them and encourage more music. (Okay, so the DVD of The Wall won’t encourage Pink Floyd to write more music… heh)

    – Easy: My time is valuable, and it’s more cost-effective and just plain easier to drop a few CDs into the amazon shopping cart and pick the super-saver shipping, rather than hunting around online for an iffy rip. (Probably makes a difference that the only peer-to-peer ap I’ve ever used is DC++ — I stay away from Kazaa and the like thanks to the spyware).

    Do I pay for everything I download? Heck no. A lot of stuff I delete after trying it out, some I keep, and some I buy. When I buy one, I’m very likely to buy more from the same artist (without a download “demo”).

    Here’s a great example: I heard a song on an mp3 “radio” stream a few times and just loved it (Sarah Brightman, Hino de la Luna). Came across an album on a newsgroup and gave it a whirl. Now I have 6 of her CD’s (including the one I downloaded) and three DVDs. All thanks to filesharing. My last amazon order included four CDs — all of which I’d sampled as a download, and three of which I’d never even heard of before downloading.

    I don’t like most of what’s on the radio anymore, unless they’re playing something that’s at least ten years old. 🙂 Before I started downloading music, I probably went five years without buying a single CD, and another five where I bought just a few each year.

    Incidentally, the same applies to DVDs. For example, I downloaded all five seasons of Babylon 5, and now that they’re finally coming out on DVD, I’ve picked up each season as they’ve been released.

  4. Felten’s Unified Theory

    On his Freedom to Tinker blog, Princeton computer science professor Edward Felten presents his “Grand Unified Theory of File Sharing,”…

  5. A Grand Unified Theory of Filesharing

    Recently we’ve seen several studies of the impact of filesharing on CD sales. We have enough data now to draw some (very) preliminary conclusions, assuming the studies are correct. Despite the apparent contradictions between the various studies, I thin…

  6. Chris Tunnell says

    Cypherpunk, I agree with you that there are a lot of people that unadmittedly download music instead of buying CDs, but here is my example to throw in the mix:

    File sharing has introduced me to many new genres of music, most recently Icelandic. I would never have bought an Icelandic-music CD pre-file-sharing. I bought the CD because it was cheap (since nobody else listens to this stuff), compatible across many devices (Read: car) and also to support the artist.

    File-sharing removes the doubt of whether you will like the CD or not before buying it.

  7. Cypherpunk says

    To Peter and Mason: you say you now buy many more CDs than you used to thanks to file sharing, because you find bands you like by downloading their songs. So why do you buy the CDs, exactly, if you can get the music off the net for free? Is it just altruism and charity, like giving homeless people money? Or do you gain some specific value from buying the CD that you can’t get from downloading, which makes it worth the extra fifteen bucks?

    I’ll give you another data point. My wife loves music, and she’s bought it for years, the whole time we’ve been married. I’m looking at probably 3,000 dollars worth of CDs filling the shelves in front of me. But in the last two years, since she discovered downloading, she’s spent about 50 bucks, tops.

    But hey, I guess she’s unusual, from all the stories I hear online. Everyone else seems to enjoy paying money for what they could get for free. Maybe most people just don’t care about money these days? I can’t help wondering whether we all inhabit the same world.

  8. Eric, you should make your data available for replication. And if you used {R} code to produce the tables, include the {R} code at the site.

    I probably have access to larger market research databases on consumer behavior that would be interesting to throw into the models.

  9. My experience is similar to Mason’s and appears to be corroborated in the thesis. I am currently 43 (and single if that matters to anyone). I find the references to Oh Brother Where Art Thou to be amusing, that was also one of my last album purchases.

    When I was able to find new bands on Napster and other online sites like x-radio.com, I purchased their albums. Before Napster had been branded as Outlaw Central, I used it to find and purchase 5-20 albums per month. Albums I would not hear on the radio stations. At that time in South Florida, Clear Channel purchased a number of stations (which all seemed to get the same playlist), the public radio stations switched to talk, the last commercial classical station sold out and changed formats to teenybopper stuff. Since the companies I worked for stopped allowing it (or similar products) around 2000, my album purchases dropped off dramatically. Litigation against the online “radio” stations dried up the remainder of sources of new music. The last album I purchased was in 2002, and I had heard part of it on All Things Considered. I don’t share music, nor use p2p due to the vigilante action of the record industry, but since I have somewhere over 700 CDs, I don’t think I need to buy another album as long as I live. Music distribution in the US is broken. Will I go back to buying 200+ albums per year? If there was a way for me to find new music, not the crap played on the top 40 stations, sure. If the record industry woke up to the fact that p2p is advertising, just like the payola they pay to the radio stations, they should be able to live with it. But they can’t. They want to pay Clear Channel $1,000,000 per week per song. They don’t want my money. Or rather they want my money but are unwilling to let me have anything for it.

    I guess if the RIAA wants a per capita consumption of ~5 albums per year, I have already purchased my expected lifetime of music. The word boycott comes to mind, but that implies the would-be purchaser is interested in purchasing the product or services. So far, I have heard nothing worth buying in 2003 and 2004.

  10. Here’s one anecdotal account. As a teenager, I spent crazy amounts of money on CDs. Once I grew up and acquired other responsibilities — and stopped enjoying the “latest” music — I spent next to nothing, barely a handful of them over several years.

    Now, just shy of 30, my CD purchases have risen dramatically thanks to discovery of new bands that would never get radio play, all via mp3s (usually from usenet).

    I’d be willing to bet my experience is duplicated many thousands of times.

  11. Eric Boorstin says

    The study includes a variable counting the number of kids aged 0 to 4 in a city, and finds that the variable is statistically insignificant.

    There are at least two effects at work. The first is the one you mentioned: people with young kids are more likely to spend their entertainment money at home. The second works in the opposite direction: people with young kids have less money to spend on entertainment because they must devote resources to caring for their children. I found that the aggregate effect of young children was insignificant.

  12. Honestly, I think the underlying variable is approximately “Likes to spend entertainment money on home”. Under 25, a greater part of the entertainment money might be spent on restaurants, bars, parties, clubbing. Over 25, more people are settling down and tied to home because of kids.

    “Has *young* kids” might be a very interesting variable to do regression on.