(Caution: here lies crazy speculation. For a backgrounder, the casual reader should probably read my Common Misconceptions about Publishing series of essays; otherwise you're going to fundamentally misapprehend what I'm talking about.)
Trade fiction publishing is a supply chain business. At the back end, out of sight, a trade fic publisher takes raw inputs from a large number of small businesses (mostly sole traders). It transforms these inputs, packages them, and then—at the other end of the business—distributes them via wholesale and retail channels. You or I then buy the products, which are micro-branded and highly idiosyncratic. The author is the micro-brand; despite centuries of striving there are few sub-sectors of trade fic publishing where a reader might go to a store and buy half a kilo of a particular publisher's product range without reference to the authorial brand.
Like all supply chain businesses, trade fiction publishing is dominated by contracts—contracts with suppliers (such as authors, copy editors, typesetting bureaux, print shops, cover artists), and contracts with customers. (You and I are not these customers: I'm talking about Amazon, Barnes and Noble, Waterstones, et al.)
These contracts lock in certain types of business practice. And the first contracts in the chain are author/publisher contracts. And so it occurs to me to ask: what new business models might be possible if author/publisher contracts were drafted differently?
An author sells the publisher a limited license to reproduce their work in a given language, in specific territories, usually for an open-ended but finite (and terminable) period. (That's misconception #1 right there: I never sell my copyright to a publisher. Period. In fact, I would view any publisher who tried to buy the copyright to a piece of my work as negotiating in bad faith. This point differs in some other sectors—journalism and technical writing come to mind—but is pretty much standard in trade publishing.)
The contract enumerates the sales channels through which the publisher may reproduce the work, and specifies how the author will be paid for each copy sold through each given channel, and also how the publisher will account for sales (including the author's right to audit their books on demand to ensure they're not hiding anything).
The big sales channels are: hardcover (shipped to booksellers on 90 or 120 days' rolling credit—at the end of that time the seller must either return the product to inventory undamaged, or pay for it), mass market (sold like newsstand magazines, i.e. essentially disposables: at the end of the credit period unsold stock must be destroyed and the stripped covers returned as proof, and sold stock must be paid for), trade paperback (sold like hardcovers), audiobooks, and so on.
(Ever wondered why mass market paperbacks are all the same size, and trade paperbacks are a different size? It's so that bookstore clerks don't accidentally rip the covers off trade paperbacks in order to claim them for credit—to get credit for a trade book you have to return it intact; if you destroy it, you bought it. Oh, and this channel imploded in the UK about 20 years ago; all British paperbacks are trade books now. Some are just smaller than others because that's what customers expect, is all.)
The big new sales channel is: ebooks. These are sold as trade. There's no credit, as I understand it, because there's no physical print run and no stock: a customer clicks "buy the book", the retailer takes their money and forwards a request to the publisher's server to send back a DRM-locked copy encrypted with the customer's public key, and the retailer forwards the ebook to the customer before they have time to notice what's going on in the back office. Alternatively, some (Amazon springs to mind) act as re-publishers who license re-publication rights (on a specific platform) from the initial publisher; they handle the DRM themselves and account and pay for sales monthly.
Right now the ebook sales channel is still growing. It's eating away at the "disposable reading" niche that for three quarters of a century was occupied by the cheap mass-market paperback. But in another couple of years it will probably stabilize. The US mass market paperback will die, but trade paperback distribution will still exist for beach-side reading and technophobes; hardcovers will still exist for bibliophiles: but ebooks will dominate casual reading.
Author/publisher contracts specify royalty rates in the craziest way imaginable. This is because they consist of archaeological strata of legal boilerplate, accumulated over decades and haggled over by publishers' lawyers and authors' agents. Contract law is essentially a defensive scorched-earth battleground where the constant question is, "if my business partner was possessed by a brain-eating monster from beyond spacetime tomorrow, what is the worst thing they could do to me?"
And so we have constant re-use of legal boilerplate that's decades old. "For sales under 10,000 copies, a royalty of 10% will be assigned based on the undiscounted suggested retail price. From 10,001 to 15,000 copies, a royalty of 12% will be allocated ... from 15,001 up, a royalty of 15% will be allocated ... for copies sold at less than 40% discount off SRP, the full royalty will be paid; for copies sold at discount of 41-50% 80% of royalties due will be paid: from 51%-65% 50% of royalties will be paid: above 65% 40% of royalties will be paid." You can think of it as a stack of IF () THEN () ELSE () statements switched off the number of copies sold and the discount the wholesaler extorted for taking them off the publisher's hands.
Now, here's my question:
What if ebooks weren't just a sales channel?
There is stuff we can't do while this sort of rigid, mechanistic contract is in force: business models that are inapplicable to paper books, but that could work on electronic downloads.
* Books are sold today by reverse auction—the newer the title the more you pay for it. But the reverse auction only has a couple of price points, which correspond to sales channels: paperbacks are released a year after hardbacks, for example. And because the bindings are different, consumers mistake these for physical goods, not price points in a reverse auction. In contrast, we could do a direct reverse auction of ebooks: start at $20, then drop the price by 5% per month until it bottoms out at a floor of $2. For added fun: if you want my new novel but are only willing to pay $10, then I'll take your $10, give you the first half of the book right now, and email you the entire book when the price drops to match your payment.
* Publishers angst a lot about DRM and piracy. But a wired-up publisher could in principle be their own Kickstarter. At launch, you can buy a DRM'd copy of a book. Your payment goes into a pool. Once a set profit threshold is exceeded, the DRM on everyone's copy is released and copies sold thereafter are DRM-free. Or they could make the title 100% DRM-free but run it as a straight kickstarter, releasing the book when pledged pre-sales hit the profit point: implementing Bruce Schneier's Street Performer Protocol (for how to sell IP safely in a world with ubiquitous, instantaneous piracy).
* Books are the length they are because of binding and printing overheads. But in an ebook world it's perfectly straightforward to sell short stories ... or 2000 page doorstops. And to put different covers on them. In the world of ebooks, book covers persist as advertisements that show up first in thumbnails (Amazon's are 120x80 pixels) on storefront websites. Why not use animated covers?
* Why don't we do A/B testing on covers to work out what sells?
We don't see much A/B testing on the covers of physical paper books today because it costs money and time and can lose you a place on the bestseller charts. Books have occasionally been sold with multiple covers in the past—"Skyfall" by Harry Harrison springs to mind, as does the first Harry Potter novel—but they're usually on course to go bestseller before they get the high-budget launch. You can only really justify that kind of market research technique if the potential profits from, say, a 10% uptick in sell-through is going to justify the work of preparing a second cover and setting up the test. For most anything short of a novel from an A-list author that is hopefully going to go bestseller, the profits to carry the work just ain't there.
The snag with A/B testing such breakout-hopeful works is that books are sold as SKUs identified by an ISBN specific to the binding and distribution mechanism for the channel in question. That is, a hardcover has a different ISBN from a trade paperback or mass market paperback of the same book. Bookscan, from which bestseller charts are compiled, tracks book sales by ISBN of the item sold at the cashier's desk. To do A/B testing you'd have to issue two different ISBNs (or ASINs on Amazon). This splits your sales figures, thus almost guaranteeing that your hopefully-going-to-go-bestseller title won't make the charts.
However, ebooks need cover art which is legible even when iconized; this simpler graphical elements, which may reduce production costs. And if they're not being sold as SKUs through the normal distribution channels, either they're not being tracked by the bestseller chart compilers in the first place, or a new and hopefully more flexible sales capture system can be deployed. If we can combine cheaper covers with A/B testing we might actually end up getting somewhere.
* What about the first sale doctrine? Under the first sale doctrine, if you buy a paper book you own it and can do what you like with it—read it, warehouse it, burn it, give it away or sell it. Right now, most publishers are petrified by the question of what happens if the first sale doctrine is applied to electronic content, because ebooks are so easy to duplicate. This isn't an abstract concern; following a recent German court ruling, resale of second-hand software is legal in the EU. To avoid resale issues, early ebook publishers decided to license Ebooks as software products rather than selling them like physical books; looks like in the long term this gambit failed. But there may be ways to deal with the question of ebook sales that generate multiple readers if we can step outside the current licensing frameworks. Or if we steal a leaf from Artists' Resale Rights. (Authors to customers: "you own your ebooks. You can give them away to someone else with no strings attached. You can even re-sell them. But if you resell them for more than [insert reasonable amount here], you owe the creators a 10% commission on your proceeds.")
* What about libraries? This is a subset of the first sale problem; publishers (US publishers in particular) generally hate the idea of selling a single copy of an ebook to a library who then lend it to all their customers. As it happens, there's a solution to the library loan question in much of the world: Public Lending Right. Libraries track loans, so can in principle pay a [tiny] kickback to authors (or publishers) every time a book registered with them is loaned out. How you fund a PLR system is an open question—the UK one is paid for out of central government funds, but the Conservatives seem to be on track to abolish it (along with libraries).
Summary: The issues here are enormous and gnarly. But I suspect a chunk of problems currently facing the publishing industry may go away entirely, or look very different, if we could only tear up our existing contracts that view ebooks as a distribution channel and figure out a better way of accounting for success.