Why Books Want to Be Free


Yesterday I sketched a model of pricing in the traditional book industry. The question I’d like to address now is what this model implies for the future of the eBook industry.

My argument leans heavily on the proposition that the price of content tends not to vary with the quality of that content. So before I get into my predictions, it’s worth commenting on how universal this “law of one price” seems to be. The price of content varies by medium (hardcover fetches more than paperback) and timeliness (first-run movie theaters charge more than second-run movie theaters). But in virtually every competitive market for mass-market content, prices tend not to vary with the quality of the content itself. New music CDs are almost always priced at $10-15, new movies at $15-20. Magazines sell for $3-$5 a copy at the newsstand. Movie theaters charge the same price for tickets whether the movie is a $200 million blockbuster or a $2 million indy movie. There’s every reason to think that the mature eBook industry will conform to the same pattern: a “standard” price will emerge, and publishers won’t deviate from it very much.

What will that price be? The obvious point is that the marginal cost of “producing” and distributing a new copy of an eBook is very close to zero. So assuming a competitive market, we should expect prices to be pushed down to zero.

3878640866_51f1281091There are two factors that will push the prices of eBooks down to zero. First, the supply of eBooks is likely to expand dramatically, as publishers will have every incentive to publish a lot of books that they would have judged to be not good enough for paper printing. This increased competition will put downward pressure on prices. Second, the market for eBooks has no natural floor. Almost any price—$10, $5, even $1—is still well above marginal cost. And so once eBook publishers start cutting prices to build market share, there’s no obvious stopping point.

This is a fairly simple—some would say simplistic—argument. And it’s an argument that tends to trigger strong disagreement. Matt Yglesias, for example made a version of this argument about music earlier today, and got a lot of comments like this:

Making a song requires time from a songwriter, skill from a performer to learn the parts, and the expertise of a recordist and the use of the recordist’s highly specialized equipment.

The distribution ought to be slightly higher than the cost of distribution, which via the internet is close to zero. But you are forgetting the costs to make the song in the first place.

Of course, Matt was not “forgetting” the cost of making the song. Rather, he understands that prices in a competitive mrket tend toward marginal cost (the cost of the last unit), not average cost.

Still, it’s not crazy to think that in a market where books cost $0, so few authors will be willing to write new books that consumers will be willing to start paying again. But I do think it’s misguided. There are a lot of people who will happily write books (and songs) for very little money. Publishers routinely reject manuscripts that represent hundreds of hours of an aspiring author’s work. Most of these manuscripts are terrible, of course, but a significant number are not—the publisher simply judged them insufficiently good to recoup the cost of printing on paper. That calculus will change dramatically when publishing costs next to nothing. It will be worth taking a chance on even books that have a very modest chance of success. And some of those, will, in fact be popular with readers.

And this, in turn, means that the John Grishams and Agatha Christies of the world will have a lot more competition. Thousands of talented writers who failed to persuade a publisher to print their books on dead trees will now have the opportunity to publish directly to peoples’ Kindles. Some fraction of them will catch the imagination of readers and find a large audience. And that puts downward pressure on Grisham’s book advances in two distinct ways: because publishers are cutting their prices to compete with other publishers publishers, and because publishers have many more popular writers from whom to choose.

And as we’ve seen earlier, the price of content almost never varies by quality. Avatar cost $200 million to make and has been widely hailed by critics, yet tickets cost roughly the same as tickets to Jennifer’s Body, which cost $16 million to make and was widely panned. So it’s hard to imagine a stable equilibrium where new authors’ eBooks go for $1 but John Grisham’s eBooks go for $10, because content consumers tend to be price-sensitive.

If that sounds like hand-wavy theorizing to you, consider that it perfectly describes today’s blogosphere. There are millions of blogs in the world. The overwhelming majority of them are not very good, but even the top 1 percent still represents vastly more content than any one person can read. And the competition has made it virtually impossible for bloggers to charge for copies of their work. Even the most popular blogs are available for free online. Every “A-list” blogger understands that that he’d lose 90 percent of his readers overnight if he tried to charge a subscription fee.

Notice that this is true even though some of the top bloggers are extremely talented and have large and growing audiences. The zero price of blog content isn’t a negative judgment about blog quality. It’s simply a reflection of supply and demand. Even at a price of zero, the supply of high-quality content exceeds the attention span of the average reader by a huge margin. Which means that the equilibrium price is zero. There’s no reason to think the economics of eBooks will be any different.

In my final post in this series, I’ll look at what this analysis implies for content creators. The short version: we’re not all going to starve to death.

Update: One point that I should have made explicitly is that this argument has absolutely nothing to do with illegal file sharing. Obviously, illicit file sharing has accelerated the decline of the recording industry, and may very well have the same effect on the eBook market. But prices would continue trending downward even if the recording industry figured out a way to completely stop illicit file-sharing, because lower prices are what you always get when barriers to entry fall and competition increases.

This entry was posted in Uncategorized. Bookmark the permalink.

13 Responses to Why Books Want to Be Free

  1. Good article, with the exception of what I think is your mistake about illegal filesharing. Contrary to what the old music industry claims, all the evidence I’ve seen implies a greater liklihood that illegal file sharing actually improves music industry’s profits… in the same way lending a book to a friend increases sales. If it’s a good book the friend goes out and gets their own copy. There are several books I’ve had to buy several times because I keep lending them out but not getting them back.

    The physical distribution costs of CDs have been pennies from the outset, and worse, the audience has known that almost from the outset. Yet industry collusion has kept CD prices at the inflated price of $15 – $20. We know that. Filesharing is cheaper still, because the only overhead is recording. Music industry distribution costs hit virtual zero long before the book industry because their primary product has been recordings.

    All the stuff about illegal filesharing is a red herring: they really don’t care about illegal file-sharing… its the LEGAL file sharing which threatens their existence. Recording artists no longer need the industry… which is why 30% of today’s recording industry is independent. THAT is the real threat to the music business.

  2. jrep says:

    I think you leave out two factors, though I’m not sure how much impact they might have on your basic claim.

    First, especially in a world of endless content, distribution cost needs to include advertising cost, and publishers will, I think, continue to make conscious decisions about which works are worth their advertising energy.

    Second, we’ve just last week seen a publisher arm-wrestle Amazon into charging vastly more than marginal cost for e-books. Many (including Amazon) have predicted that this position won’t survive market competition, but as yet there’s no evidence to support that thought; it might, to the contrary, create a consensus price floor. Your “tends-to-marginal” argument is a free-market notion, but this market may be far from free.

  3. Brian Rush says:

    While I have no doubt that the price of ebooks will, at least at the top, trend downward, and that the current asking price of $15 is absurd and unsustainable, a literal zero price is equally unsustainable, and it’s also unrealistic to expect that no premium will be charged for the work of very popular authors.

    If you want to see your concept in action, AND to see its natural limits, take a look at prices for self-published ebooks. These range in almost all cases from free to $10, with most running under $5, and quite large numbers of titles under $3. Which ones are free? They fall into two general categories: books published for some other reason than to directly make money (books that make a statement, that promote some other service that the author expects to make money on, etc.), and books that are offered free for promotional purposes, with the idea of selling other work by the same author for a price.

    Most authors expect to be paid for their work. Competition will keep ebooks from being priced very high, but there is a limit to how far it can drive prices down, and that limit is certainly above zero except in the limited cases outlined above.

    I would also say that a very successful ebook author will find himself able to charge at least marginally more than one who is an unknown. Steven King could self-publish an ebook next week, charge twice what I do, and sell a lot of books, because he has a lot of fans. It all comes down to what the customer is willing to pay. A fan of a big-name author is willing to pay more than someone who is just shopping and finds something by someone he’s never read that looks interesting.

    That said, we also have to recognize that the Big 6 are not presenting a serious model for pricing ebooks and aren’t even trying to. The fact that publishers will receive less per e-volume sold under their preferred model, not more, is the key to understanding this. They’re not trying to shore up their revenues from ebook sales, they’re trying to deter readers from buying digital editions of best-selling titles. So it’s not how much money goes to the publisher from such a sale that’s the point, it’s how much the customer has to fork over — publishers want that high. They want it to hurt. They don’t want people buying their ebooks. And the reason why they are doing this is to protect sales of hardcover print books, because that is the business model that big publishers are committed to.

    This is, in my opinion, also unsustainable. The downward trend in sales of hardcover books is not going to be reversed, and by committing themselves to this form of publishing, big publishers render themselves dinosaurs doomed to extinction. As they decline towards death, so will $15 ebooks.

  4. Tim, I’ve spent several hundred dollars in the past couple of years buying DRM-free MP3s which, I assume, are also available to me at whatever the opportunity cost of my searching is. Is your prediction that if I wait until a year, the same files will be cheaper, and that at some date in the future the same files will be offered for $.00? Just want to be clear about what your claim is. Do you have a theory about the rate at which prices tend toward zero, or that explains why the market price of MP3s is not zero now?

  5. Will: I don’t have a model about how long it will take. Economics gives us a theory about what the equilibrium will be but it doesn’t tell us very much about how long it takes to get there.

    I think it’s likely to be quite a while before major labels free their back catalogs. Rather, I think we’ll see new business models emerge that gradually displace the labels and eventually drive the labels out of business. It’s hard to predict how markets will evolve, but if forced to guess, I would say that we’re likely to see two trends: the growth of ad-supported models like Pandora, and the emergence of new bands that choose to give their music away as a way of building their fan base and (therefore) their concert revenues. But markets are unpredictable. There are probably other ways to make money from free music that I haven’t thought about.

  6. “Economics gives us a theory about what the equilibrium will be but it doesn’t tell us very much about how long it takes to get there.” Is the economic theory neutral between reaching the equilibrium tomorrow and in billion year? Why is this bit of economic theory even relevant then?

    I will bet you $100 that the average price of e-book versions of the books on the NYT bestseller list 5 years from this week will be greater than $4.99.

  7. Mike says:

    The price of video games with online downloads available for purchasing don’t seem to be trending downward in price (though I might not be aware if they were). Given that online downloading of video games is expanding for PS3, XBOX, and computers, do you expect a similar effect to happen in the near future? That video games will converge to $0.00?

  8. Mike: software is a weird market for several reasons, and I don’t think it quite counts as “content” for purposes of my argument. One difference is that software is subject to rapid technological improvement—a new console gets introduced every five years or so, rendering existing video games obsolete. Additionally, console manufacturers work hard to prevent the commoditization of their platforms.

    Second, video game developers typically have to pay for the privilege of developing for a console, and this tends to exclude amateurs. The labels have been struggling to prevent the commoditization of music technology, but they’ve largely failed because the basic technology for music-playing isn’t very complicated and doesn’t change much over time.

    Third, modern video game development is orders of magnitude more capital-intensive than either writing a book or recording a song is. Spending $10 million on a video game will make it a significantly better game, whereas it’s hard to convert money into a better book or song. This means that my assumption of an ample supply of aspiring writers and musicians may not apply for video game development.

    Finally, networked video games are subject to network effects (literally!) to a greater degree than most content. If all of your friends are playing World of Warcraft, you’re going to want to play World of Warcraft, not a technically-equivalent game. This makes consumers dramatically less price-sensitive. It’s also true that people want to read the books their friends read, but the effect is much weaker.

    I haven’t paid close attention to the economics of online video games specifically, so I can’t comment specifically on trends in that market. But there are clear examples of popular zero-price video games. Second Life is one such example. The game itself is free, but you can pay for in-game currency and real estate. I wouldn’t be surprised if such business models became more common, but I think software is different enough that the arguments I made in this post don’t apply cleanly.

  9. Jim Crider says:

    I’m quite surprised to read on your bio page that you spent time as a freelance author, because here you display zero knowledge of how a book becomes a book (or, indeed, an ebook). Authors write, editors edit, art designers do layout, typesetters take the edited manuscript and format it into book/ebook form, artists create the covers and interior illustrations (yes, even those elaborate chapter headers), marketers construct marketing campaigns to get word of the book out to those who want to read it. Note that I haven’t said anything about printing, binding, and distribution of the physical copies. There’s a reason for that: printing, binding, and distribution of physical copies is less than 10% of the cost of producing a book. The remaining 90% are fixed costs, regardless of format (dead tree or electronic).

    That’s right: costs. Those people, who do those jobs *AS THEIR CAREERS*, do so for *MONEY*.

    Should you be able to take the painting off the wall of the museum (or the artist’s studio) and walk out with it gratis? How about the sculpture? When you do whatever it is that puts food on your table and keeps a roof over your head, and your internet connection live, do you do it for the love of doing it, or do you expect to get *PAID*?

    Why should an author, who puts considerable hours, not to meantion heart and soul, into their work, not be compensated fairly for it? You seem to think that the mere love of creating art should be enough for them. Love of creating is great, but when it’s 0F, snowing, and the pantry’s empty, fat load of good that love does. Even the authors who still have their day jobs (which is most of them) put in just as many hours, if not more, writing as they do in their day jobs — and they expect, if and when they *sell* (word choice very deliberate) a story to a publisher, to be paid for their effort.

    Even the innovators and free-sharing folks you admire in the tech industry aren’t adverse to making a buck or two. Think Steve Jobs lives in a cardboard box out back of Cupertino? Wozniak hasn’t done too bad, either. Ellison? Gates (not exactly free-sharing, but still, gotta admit he set the innovation bar pretty high early on)? Yeah, thought so.

    When you’re ready to do your job “for the love of the work” alone, let me know, Mr. Lee. Until then, I suggest you do a wee bit more research. You might want to start with who was the aggressor in the Amazon/Macmillan spat this past weekend. Hint: it wasn’t the publisher who acted like a petulant 5-year-old, it was the distributor. Furthermore, despite what Amazon says, if you actually *READ* the open letter from the Macmillan CEO, you’ll see that the agency pricing model offered by the publisher would ultimately result in ebooks available for *less* than Bezos’ magic $9.99 bogey. Just not new hardcover releases. It’s not exactly a novel concept: new hardback costs more than later trade paper edition costs more than still-later mass-market paper edition costs more than remainders.

    And… you’ll start seeing big-name authors (your Grishams and Kings and Dan Browns) putting delayed-release-of-ebook clauses in their contracts (it’s already happening). Reason: certain retailers of large volumes of physical books have price-matching guarantees, and don’t want to take a loss on a new hardback that’s available for half the price as an ebook. Result: those certain retailers of large volumes of physical books simply *do not order* copies of those new releases, costing the authors many sales. Delay the ebook release, and you sell many dead tree copies at the large-volume retailers. THEN you get the boost from the e-books, and subsequent print editions at lower prices (because, remember, the sunk costs are essentially fixed, and resetting the typography for the smaller trade paper and mass-market paper page formats is a relatively small cost, as is the actual printing & distribution of the books themselves). I have a feeling you’ll start seeing this even among midlist authors in the next few years.

  10. Mike says:

    “Spending $10 million on a video game will make it a significantly better game, whereas it’s hard to convert money into a better book or song.”

    That seems right. Good call. Though there’s going to be some amount of not-arbitrary money that curves the quality up for books and music. (If only time as money for writing books.) It may be, say, $20,000, in terms of 3 months of writing or a recording studio, to get a significant jump in quality. In terms of bands, there touring and people paying for live music, that could recoups that in the zero-price world. How that gets recouped for writers is the big question, where I assume you are going next.

    Please don’t tell me everyone just becomes a policy wonk on a long enough timeline….

    I still think with search costs (even if only headaches), people would be willing to pay editors to filter writing for them, so in that sense I can see a non-zero price equilibrium for content magazines, for instances. Too many choices as a cognitive ‘cost’ and all that, behavioral rational choice as people with limited ‘bandwidth’ and all that, which I hear people are starting to play with models for.

  11. I will bet you $100 that the average price of e-book versions of the books on the NYT bestseller list 5 years from this week will be greater than $4.99.

    I will not take this bet because I think you are probably right! The reason is that in 2015 I expect most people will still be reading paper books, and so the economics of the paper book market will still be the primary thing driving the behavior of major authors and publishers. As long as paper is the dominant medium for books, publishers’ primary concern with e-book pricing will be to avoid cannibalizing paper book sales. And since paper publishing is likely to be more profitable than e-publishing, the best authors are likely to gravitate there as long as print contracts are available.

    As for why this is relevant: these posts were prompted by your request for “useful discussions of how authors make $ without DRM.” The point of this post is to explain why I don’t think selling copies of of e-books will be a major revenue source for authors in the long run. I also don’t think they’ll be a major revenue source in the short run, because I think that the vast majority of people will continue to buy paper books. I can’t predict precisely when paper book sales start to plummet, but my prediction is that with or without DRM, e-book revenues will never come close to replacing the revenue lost from falling print book sales.

    This is precisely the pattern we’ve seen in the music business. The labels’ revenues from CD sales are now falling precipitously, and while iTunes revenues are growing, they’re doing so far too slowly to replace falling CD sales. I think something similar will happen in the book industry if and when it moves to a predominantly e-book format.

  12. Tim Baker says:

    Does the time factor need to be factored in to the above arguments on the economics of ebooks?

    Books unlike music or films are “consumed” in parts over several days/weeks. It is interesting to note that although 30 years ago dead tree books were considered cheap, 180 years ago they were considered as too expensive for many works and many of them were published as a series in magazines. Charles Dicken’s “Olivier Twist”,was first published as a serial in monthly instalments over more than 2 years (February 1837 – April 1839). Only successful series made it to the dead tree version, as a signle physical book. As dead tree books return to having a status of being expensive to produce, perhaps literary works will be again constructed to be serialized, giving rise to opportunities of returning to a former business model where episodes of a book are supported by advertising. The better ones may still be raised to the dead tree status.

  13. Eorr says:

    I think you are starting to see the commoditization of games in two areas. One is internet flash type games which are enormously popular and free. The second is iPhone/android games where the price is constantly pressured toward $0. These types of games are much easier to produce and thus the competition is exponentially greater than console games and MMO’s.

Leave a Reply

Your email address will not be published.