Well, they would do, wouldn't they?
With Hachette insisting on the Agency Model in the UK, and five other publishers expected to join in, Amazon has posted a letter to its UK customers on the amazon.co.uk Kindle forum. Most of its content you can guess and doesn't really affect B&N at the moment, but here's an interesting tidbit:
Unsurprisingly, when prices went up on agency-priced books, sales immediately shifted away from agency publishers and towards the rest of our store. In fact, since agency prices went into effect on some e-books in the US, unit sales of books priced under the agency model have slowed to nearly half the rate of growth of the rest of Kindle book sales.
It's always an interesting challenge to try to read the tea leaves hidden under the spin. [What kind of mixed metaphor was that?!? ] First, it should be noted that what they said that was halved was the rate of growth, not the total sales. They didn't say anything about the Agency 5 (A5) sales dropping, just that they're not growing as fast as the non-agency sales.
As we recall, the Agency Model was introduced simultaneously with the iPad. There are a lot of iPads out there, and iPad users can buy e-books from Amazon, B&N, Apple, and a few other sources. Of particular interest, they can buy A5 titles from Apple, but for just about any other titles the iPadders have to turn to Amazon, B&N, or the other sources. Early on, Apple was crowing about having grabbed about 22% of the Agency 5 sales (and failing to mention that they grabbed 0% of other sales). So it's possible that when Amazon sees weak A5 sales relative to other titles, some of those A5 sales are simply going to Apple. Also, Amazon's loss of overall market share to B&N and the other sources would make Apple's claimed 22% of the whole A5 market a bigger percentage of Amazon's sales.
Would Apple sales be enough to reduce the rate of growth of A5 e-book sales at Amazon? Heck if I know. Nobody wants to give out hard numbers. Over at TUAW (The Unofficial Apple Weblog), Dave Winograd has declared the iBookstore to be "a dismal failure." And author Joe Konrath says that he's selling 70 e-books on Amazon for each 1 on iBookstore.
[Reading between the lines, it looks like Konrath's B&N sales are about 1/5 of his Amazon sales, which is passably in line with B&N's claims of 20% e-book market share. And that's with Konrath distributing through Smashwords rather than PubIt!, which wasn't available during the timeframe being measured.]
My conclusion: none. A lot of smoke there, but it's hard to see what's going on behind it.