Business TEDTalks

The numbers behind the Copyright Math

Posted by: Kate Torgovnick May

Were you intrigued by “The $8 Billion iPod,” Rob Reid’s short TEDTalk about the new science of Copyright Math (TM)? We were. We needed to know more. More numbers, Rob! we said. And Rob (whose comic novel Year Zero comes out in July) sent us this treatise, a master class in creative mathematics:

A few weeks back, I gave a short TED talk about “Copyright Math.” Since TED draws both Hollywood and Silicon Valley bigwigs, I thought it would be a great venue for raising certain rights issues that have been a sore point between the two industries for years. But January’s brawl over the proposed SOPA law was a raw and recent memory. So I decided to make my talk playful, rather than sermonizing. Everyone can laugh at silly infographics. And who DOESN’T want to deface a Leave-it-to-Beaver-like Christmas scene with pirate-and-Santa graffiti?

Since the talk was so short, I couldn’t dive deeply into the numbers and sources that I based it on (which would have shattered the whimsical tone anyway). But even my silliest numbers were derived from actual research, performed by an actual Copyright Mathematician (me, that is). So I thought I’d use this blog post to put my sources and calculations out there for anyone who’d like to nerd out on the details.

First, the Motion Picture Association’s claims of $58 billion in actual US economic losses and 373,000 lost jobs came from this press release[1] (which can also be found on Scribd[2]). These numbers originated at a think tank called the “Institute for Policy Innovation” – an organization that Businessweek once profiled in an article called “Op-Eds for Sale.”[3] In it, an IPI analyst freely admitted to taking payoffs from disgraced lobbyist Jack Abramoff[4] in exchange for writing “op-ed pieces boosting the lobbyist’s clients.” The IPI’s president supported this behavior, saying it was neither wrong nor unethical, and dismissing those who apply “a naïve purity standard” to the business of writing op-eds.

This doesn’t necessarily mean that MPAA lobbyists paid the IPI to conjure up these numbers. But whatever their genesis, they’re not easy figures to support. In a February New York Times piece,[5] an MPAA spokesman did his best by attributing the eye-popping $58 billion sum to “piracy’s impact on a range of tangentially related industries — florists, restaurants, trucking companies, and so on.” Florists? Really? Exactly how many bouquets go unsold whenever someone swipes a copy of My Sharona?

Ignoring improbabilities like pirated steaks and daffodils, I looked at actual employment and headcount in actual content industries, and found nothing approaching the claimed losses. There are definitely concrete and quantifiable piracy-related losses in the American music industry. The Recording Industry Association’s website has a robust and credible database[6] that details industry sales going back to 1973, which any researcher can access for a few bucks (and annoying as I’ve found the RIAA to be on certain occasions, I applaud them for making this data available). I used it to compare the industry’s revenues in 1999 (when Napster debuted) to 2010 (the most recent available data). Sales plunged from $14.6 billion down to $6.8 billion — a drop that I rounded to $8 billion in my talk. This number is broadly supported by other sources, and I find it to be entirely credible.

But this pattern just isn’t echoed in other major content industries. My movie industry figures (showing significant growth from the rise of Napster to the present day) came from a meticulously researched report by BMO Capital Markets[7] called “Perspectives on the Filmed Entertainment Industry” which is sadly not currently findable on their website (and BMO — if you’re listening, please do the world a service, and at least publish “Exhibit 9” publicly!). My TV, satellite and cable figures (showing spectacular growth during the same period) came from the same outstanding report. I didn’t have time to discuss them during my talk, but numbers from local media analysts BIA/Kelsey showed robust growth in radio[8] in the years immediately following Napster’s debut. This was followed by a brief, agonizing contraction in the 2007-09 timeframe[9], which the organization attributes wholly to the recession, rather than piracy.[10]

So where is the missing $50 billion in piracy? It’s hard to accept that it’s foregone growth in markets that have grown in line with, or (in the case of the giant TV/ satellite/ cable market) far faster than historic norms. So we’re left looking for a market that has no historic norms. Because in such a case, one can tenuously argue that but for piracy, it might have grown at such a blistering rate as to make $50 billion in foregone sales at least hypothetically possible. So what significant American media market literally didn’t exist at all in the ’90s?

The best I could come up with was downloadable ringtones, which were first launched in Japan and Finland in 1998, [11] and didn’t appear on these shores until later.[12] Sure, citing ringtones was a punch line. But if the MPAA can document $50 billion in other pirated media, I’d love to hear about it.

And in case you’re wondering, at 30 seconds per ringtone, and $1.39 a pop (this was the lowest price I could find for ringtones anywhere, and I figured we’d get a bargain by buying in bulk), we’re looking at 34,218 years worth of ringtones — which, laid end to end, would stretch clear back to the late Neanderthal period.[13] And for you astrophysicists, the penny is ¾ of an inch in diameter.[14] 5.8 trillion of these suckers would therefore stretch for 68,655,303 miles, which can easily connect the Long Beach Westin to Mars when we’re on a close approach.[15] And for a mere $128, we could extend that journey clear to the auditorium that contains the TED stage. Meanwhile, my agricultural crop values all came from Wikipedia.[16]

As for the MPAA’s employment numbers, I compared them to data reported by the US Bureau of Labor Statistics in its 2010-11 “Career Guide to Industries.” This put the motion picture and video industry’s total employment at 361,900 jobs.[17] The 2000-01 edition of the same guide put employment at 270,000 in 1998.[18] The 2000-01 Guide simply uses “Motion Picture Production and Distribution” as the industry descriptor, but a close reading of both Guides seems to indicate that they’re talking about the same industrial sector, so I infer that the 1998 Guide used “Motion Picture” as shorthand for the broader filmed entertainment sector.

As for music industry employment, I took the average revenue per employee at Universal Music in 2010 (roughly $852,000),[19] at EMI in 2009 ($300,000),[20] and Warner Music Group in 2008 (about $875,000).[21] This gave me an average revenue-per-employee of about $675,000 throughout the industry. Applied to the industry’s 1999 revenue, this ratio implies total employment of about 22,000 at record labels, which I doubled to account for the retail side of the business as well. This is clearly an imperfect estimate, but even if it’s off by 100% (and I’m quite certain that it’s not), it doesn’t undermine my bigger point. In any event, I combined my music industry number with the Bureau of Labor Statistics’ filmed entertainment number to get my starting-point content industry employment. I then subtracted the claimed 373,000 in job losses to infer in the (obviously playful) “negative employment” statistic.

To me, the most depressing number in the presentation is the $150,000 maximum fine that Congress designates for “willfully” pirating a single copy of a single song under the Digital Theft Deterrence and Copyright Damages Improvement Act of 1999.[22] This number is grotesquely divorced from the actual damages and harm caused by a single instance of piracy. As such, it represents a naked perversion of “The Law” — turning it from a source of justice into a bludgeon for a powerful and cynical lobby. The music industry has sued more than 30,000 US citizens under this law. Since the consequences of losing would be bankruptcy in almost all cases, the crushing majority of defendants settled without daring to challenge the industry. As a result, the maximum $150,000 per-song fine has never actually been imposed (although one student is currently fighting a verdict of over $20,000 per song,[23] and a single mom was hit with an $80,000-per-song ruling,[24] which was later reduced, but is still being debated in appeal).

In determining a given device’s maximum capacity for infringing material, I assumed an average song length of three minutes, and an encoding rate of 128 kilobits/second. I went with 128 kbps because using the AAC codec,[25] this is the rate at which music achieves “hi-fi transparency[26] — which is to say, it becomes indistinguishable from CD quality in most listening environments. This rounds very closely to 1 megabyte of data per minute of music.[27] At 32 megabytes, the Rio (1999’s Christmas hit) therefore had room for about 10 songs, which, if pirated, could represent up to $1.5MM in liabilities under US law. Today’s iPod classic, with its 160GB capacity, can hold 53,333 songs, which at $150,000 a pop is precisely $8 billion. Incidentally, Apple markets the iPod classic as having room for just 40,000 songs, but by my math, that’s selling it short. I meant to note this in the presentation, but I was running way over time by then, and spared everyone the convoluted math (so if the leap from 40,000 songs to an $8 billion liability confused anyone, I apologize — I had meant to take a quick detour through that 53,333 figure!).

FINALLY: the 75,000 jobs figure was just a joke. I think that was probably obvious.

And as for alien music liabilities … well, that happens to be major area of interest and research for me. But I’ll leave that for another forum.

— Rob Reid