Motoko Rich was on the front page of Tuesday’s NY Times business section with a story headlined “Print Books Are Target of Pirates on the Web”. She documented growth in piracy at the front (a quote from David Young about increased vigilance from Hachette’s lawyers and from Wiley’s lawyer with stats about a 5-fold increase in takedown notices.) And after an explanation of why the increase might be (people can read ripped files on devices now), she closes with a quote from DRM’s most savage opponent, author Cory Doctorow, repeating the line (which I believe originated with Tim O’Reilly) that he preferred piracy to obscurity.
But what was missing from this piece was any evidence that piracy is negatively affecting sales.
But in the past 24 hours, two reporting efforts have surfaced that seriously looked at that problem.
One is by doctoral student John Hilton and reported on Bloggasm. Hilton tried to track sales before and after an ebook giveaway. He found a lift in sales on books Random House had given away but a decline in sales for books given away by the science-fiction publisher, Tor.
The other study was done by my colleague Brian O’Leary in conjunction with O’Reilly Media and Random House. The methodology was similar to what Hilton employed and was reported by O’Leary and Mac Slocum of O’Reilly at Tools of Change last February. Now they have published a Research Paper with O’Leary’s findings which is available from O’Reilly.
What O’Leary found, using Random House data on ebook giveaways and O’Reilly Media data on books found on pirate sites, was that there was a correlation between free distribution and a sales lift for the books in question. But O’Leary cautions, “correlation is not causality”; the fact that sales rose after piracy and giveaway doesn’t mean sales rose because of piracy and giveaway. Both O’Leary and Hilton say more data is needed to come to any definitive conclusions.
Whatever we find out about the link between free ebooks and the sale of printed books and ebooks will:
1. not necessarily be true for all titles or genres.
2. not necessarily be true for all time.
Although O’Reilly also has a very tech-sophisticated audience and O’Leary found no evidence that piracy damaged sales, one has to wonder whether the difference Hilton saw between a general audience (Random House) and a sci-fi audience (Tor) is tech-sophisticate-specific. Or could it reflect wider use of Kindles or Sony Readers or iPhones among the sci-fi audience?
Cory Doctorow has made a name for himself giving content away and, not incidentally, skewering people who don’t see the virtue to that approach. I share Cory’s opinion that — up to about this moment — sharing of content from narrative books would almost always result in a sales increase. I have always expected that effect to decline when ereading became ubiquitous. When people just don’t want to read on screens, giving them a screen sample could provoke a print-book sale. When people get comfortable reading on screens, then giving them an ebook satisfies their demand.
So the Rich story failed to deliver any analytical data which two other “reporters” managed to construct. But there was another failure in the story and it is really a more serious one.
Rich points to two web sites where pirated material might reside: Scribd and Wattpad. I don’t know Wattpad, but Scribd has been in the news quite a bit lately. Several publishers have made deals to post material with them? Why? Because Scribd is aggressively anti-piracy! If they have cached a copy of a copyrighted text, it will not show up as a pirated edition on their site. So the very best way for publishers to prevent unauthorized posting of their copyrighted material to Scribd is to give them a digital copy. Since several major publishers in New York have made deals with them, which have been reported in the trade press, it is a bit mystifying that this mistaken reporting, potentially quite damaging to Scribd, could have appeared in the Times.
On a mailing list discussion of this topic, Allen Noren of O’Reilly cited a Norwegian study that indicated that the most frequent downloaders of pirated songs were also the biggest customers for legitimate downloads. In the same list, Jane Litte made the point that pirated ebooks might be available when a legitimate edition is not and, in additon, some of the pirate sites make downloading — acquiring the content — easier than the legitimate vendors do.
The modern thinking about DRM, piracy, and sharing is that lower prices and greater ease of acquisition and use will keep honest people honest without technical barriers. Unfortunately for book publishers, the so-cheap-it-hardly-matters 99 cent unit doesn’t work for us. Record companies are selling albums as digital downloads for a mere 25-30% less than they would cost as CDs, which conveniently translates to 99 cents a song. Selling books for 99 cents a chapter, or 50 cents a chapter, would not produce a similar result because the unit of music appreciation is the song; the unit of book appreciation is not the chapter.
So the fair comparison to music would be books to albums, not to songs. That puts the price at about $9.99, which is exactly where Amazon decided bestseller pricing should be. And that’s a price that could work for publishers too, if they weren’t giving retailers half of it.
The pricing and distribution of ebooks is a complex and moving target, which I’ve discussed in prior posts on several different occasions. Clearly Amazon, which creates and sells a proprietary reader through a closed system is a different animal than everybody else. They start out with a stronger hold on their customers which could provide leverage to demand more margin than everybody selling epubs and pdfs that can come from a variety of retailers and play on multiple platforms. That’s perverse from the standpoint of people who believe in industry standards, but it’s probably not an enduring advantage. To exploit it, you have to be willing to pass up titles that don’t meet your margin requirements.
But people who read their Kindle books on an iPhone (and Michael Cader’s recent analysis suggests that the recent Kindle sales spike might be due to that being a large number of people) can readily access other ebook platforms as well. If they have the title and Kindle doesn’t, it loosens Kindle’s grip on that audience.
Sorry, I’ve strayed a far bit from the piracy question. But you can’t contemplate piracy and what to do about it without analyzing ebook pricing. And you can’t discuss pricing without discussing the mushrooming complexity of the ebook supply chain.
Defending the margins from ebook sales is a big current challenge f or publishers, but getting to a point where they monetize eyeballs rather than IP is the long-run challenge. I’ll be discussing that one in a talk called “Stay Ahead of the Shift” at BEA at Javits Center, 11 am on Thursday, May 28.