On January 28, 2010, John Sargent Jr. and Brian Napack, the chief executive and the president of Macmillan Publishers, flew to Seattle to give Amazon.com some news they knew the online retailer wouldn’t like. A couple of days before, Macmillan had made a deal to sell its titles in Apple’s iBookstore, just as Steve Jobs was set to introduce the company’s new iPad. Rather than sell titles to Apple on a wholesale basis and let the company set a retail price, as it did with Amazon and other bookstores, Macmillan would set the price for its digital books itself and give Apple a 30 percent commission for selling them to consumers. Apple had also made similar deals with four of the other five major U.S. publishers.
But Amazon wanted to set book prices as well.
Since Amazon started selling downloadable e-books in the fall of 2007, it had priced best sellers at $9.99—usually $2.00 or $3.00 less than the wholesale price it paid publishers—in order to drive demand for its Kindle digital reading device. When Jobs introduced the iPad, Amazon controlled about 90 percent of the e-book market, and publishers worried that its ability to sell e-books at a loss would discourage other retailers from entering the business.1
Amazon seemed to be looking at Apple’s iPod as a model for its Kindle: instead of giving away a razor to make money selling blades, the online retailer wanted to drive down the price of media and make money selling gadgets to consume it. The cheaper it became to buy e-books, the more value the Kindle would have to consumers. Because the e-books Amazon sells have digital copy protection, Kindle owners looking for a new reading device would find it easier to stick with the same brand. That’s why Amazon didn’t mind losing a few dollars on some sales.
Publishers didn’t like this. Since Walmart also offers heavy discounts on some new books, they feared consumers would resent paying more than $20 for a new hardcover—and, more important, that stores selling them at that price wouldn’t be able to compete. But until Apple introduced the iPad, no individual publisher believed it had enough negotiating power to change the nature of its deal with Amazon. Antitrust law prevented companies from acting together. In the end, what kept Amazon from becoming the next Apple was none other than Apple itself. By offering an alternative to the Kindle, the computer company gave publishers the leverage to change their deals with Amazon.
In Seattle, Sargent and Napack met with Russ Grandinetti, Amazon’s vice president for Kindle content, and gave the company a choice: sell e-books under the same terms as Apple—using what came to be called the agency model—or Macmillan would hold back some forthcoming titles for several months.2 Amazon, which controls about 20 percent of the mainstream U.S. bookselling business, is more accustomed to making demands than hearing them.3 (The online retailer declined to comment on the negotiations.) Still, even the tense meeting didn’t prepare Sargent and Napack for what happened next. By the time they got back to New York late the next day, a Friday, Amazon had removed the “buy” buttons from Macmillan’s print books, as well as their digital versions, halting their sales.
In the fairly genteel publishing world, this amounted to a declaration of war. By all appearances, Amazon had Macmillan outgunned: The publisher depended on it more than the other way around. Amazon had the power to change the prices of e-books and their physical counterparts to make the former more appealing. And Macmillan couldn’t predict the reactions of agents and authors, who might be more inclined to worry about losing sales in the short term than the economics of publishing in the future.
And yet, Amazon folded. On Sunday, January 31, Amazon announced, “We will have to capitulate and accept Macmillan’s terms because Macmillan has a monopoly over their own titles, and we will want to offer them to you even at prices we believe are needlessly high for e-books.”4 The line about Macmillan having a monopoly made little sense to most readers, but people in the technology world understood it to mean that Macmillan had too much control over its own products. By that logic, so does every other media company.
Four of the other five major publishers had also been planning to ask Amazon for more control over their pricing, according to the New Yorker, and within a week Simon & Schuster, HarperCollins, Penguin, and Hachette announced that they too would move to agency deals.5 The exception was Random House, which continued to sell e-books to Amazon at a wholesale price until early 2011. (Random House owns my publisher, Doubleday.6) Even then, negotiations didn’t proceed without incident: in early April, Amazon temporarily removed the “buy” buttons from the e-book versions of a few titles from Penguin and the Hachette Book Group.7
Since Amazon’s negotiations with Macmillan played out in public, bloggers followed along. Henry Blodget, the former equity analyst who now runs the Web site Business Insider, told Sargent to “take your $15 ebooks and shove them!” “The marginal cost of an incremental ebook is pretty much zero,” Blodget wrote. “So why on earth should we pay $15 for it?”8
But books, like all media products, have never been priced according to their marginal cost: it doesn’t cost anything close to $25.00 to print a single copy of a hardcover book. And although many bloggers implied that publishers were greedy, the new agency deals actually called for them to receive less money for each book they sold. Amazon had been paying publishers $12.00 or $13.00 for many new titles it sold for $9.99; under the new arrangement, publishers would get about $9.00 for a book priced at $13.00—and probably sell fewer of them at a higher price. The publishers were thinking about the long term, especially about how they could make sure other companies would be able to start selling e-books without having to do so at a loss in order to compete with Amazon.
“It was galvanizing,” says Dennis Loy Johnson, copublisher of the independent Melville House. “It was the first time anyone stood up to Amazon.” A week after Amazon started selling the publisher’s books again, a Macmillan representative at the American Booksellers Association Winter Institute program received a standing ovation from a crowd of independent store owners.9
Macmillan’s skirmish with Amazon was only one battle in the ongoing war between authors and publishers and the technology companies that want to use their books to sell other products. A few months earlier, the New York Times had run an opinion piece by Sergey Brin in which the Google cofounder compared his company’s plan to digitally scan all the world’s books to the Library of Alexandria.10 You can’t fault his ambition: as of early 2011, the company had scanned more than fifteen million books. Google Books will be the most accessible repository of human knowledge ever assembled, and one of the most remarkable accomplishments of the digital age.
The Alexandrian parallel may be more apt than Brin would like, however. According to historical accounts, the ancient library had an unusual acquisition policy: it forced city visitors to give their books to scribes for duplication—the ancient antecedent of begging forgiveness rather than asking permission.11 Google didn’t concern itself with niceties either. Copyright law allowed it to scan books published before 1923, which are in the public domain, and it made deals with most U.S. publishers to scan newer works. But it was impractical to negotiate with—or, in some cases, even find—the millions of authors who got back their copyrights when their books went out of print, or the publishers who retained rights if they didn’t. Many of the writers might have jumped at the chance to make their books available, since it was difficult to find new copies for sale (and they wouldn’t get royalties on used editions). But the statutory damages for copyright infringement—up to $150,000 per work—had scared other organizations that tried to build digital libraries into limiting their efforts to books in the public domain and those they could get permission to scan.
In true Alexandrian fashion, Google saw copyright law as a Gordian knot and cut right through it. In December 2004, the company simply announced it would begin scanning the contents of the Harvard, Stanford, and Oxford libraries, among others. When the executive director of the Authors Guild, Paul Aiken, realized how extensive the project was, he says, “It sent me scrambling to figure out how this could be allowed under copyright law.”
As with so many copyright questions, the answer wasn’t entirely clear. Google argued that its scanning and indexing qualify as fair use, since the company initially planned to display only a limited amount of text.12 But the project had a clear commercial purpose—giving Google’s search engine an advantage competitors couldn’t match, in addition to any direct benefits—and no one had ever used fair use to defend such an ambitious undertaking. Within a year, both the Authors Guild and the Association of American Publishers sued Google to stop the project.13
In the spring of 2006, in response to the lawsuits, Google started negotiating a settlement, which was submitted to the court in October 2008—and again, in an amended version, a year later.14 The settlement would have allowed Google to display limited portions of millions of books for free, sell digital books to individuals for prices set by copyright holders, and offer licenses to schools and libraries that would have made the entire database available on computer terminals. In return, Google would have paid $125 million, $34.5 million of which would have funded a Book Rights Registry to disburse 63 percent of Google Books revenue to publishers and authors according to the use of their works.15
Most authors and publishers supported the deal. “It was so much better than either winning or losing, for both Google and the authors and the publishers,” says Richard Sarnoff, who had served as chairman of the Association of American Publishers and as a senior executive at Bertelsmann, which owns Random House. If Google had prevailed in court, it would have had a library of scanned books but not permission to sell them. (The fair use defense it planned to use would have applied to scanning and indexing the books, but not selling digital copies.) Had the authors and publishers won, they would have received substantial damages but no way to sell out-of-print works. Perhaps most important, the settlement would have set an informal precedent that scanning books requires an agreement with publishers or authors. “The alternative was to take our chances on winning the lawsuit, and we probably would have,” Aiken says. “But if we didn’t, it would have been a catastrophe because [Google would have] millions of books scanned that authors and publishers would have no legal control over.”
Like Amazon and Apple, Google sees books as a means to an end—in this case giving its search engine access to more information. “Probably the highest-quality knowledge is captured in books,” Sergey Brin said.16 Like record labels, publishers have become arms suppliers in a cold war between technology companies. By bringing Google into the business of selling books—and giving it enough of a selection to make it a legitimate competitor to Amazon and Apple—the proposed settlement could have given publishers more leverage.
In order to do this, however, the proposed settlement essentially turned copyright law on its head. In the vast majority of cases, copyright law is “opt in”: creators or their representatives must give permission for their work to be sold, displayed, or performed in certain ways. The settlement would have essentially given Google permission to digitize and display books unless writers specifically objected. Marybeth Peters, who until the end of 2010 served as the United States register of copyrights, the country’s top copyright official, testified before Congress that “the proposed settlement would give Google a license to infringe first and ask questions later.”17 Even some lawyers who supported the idea of an online library worried that Google and the Authors Guild were essentially using a private agreement to get around copyright law.
Opposition to the Google Books settlement made for some of the strangest bedfellows in legal history. In August 2009, the U.S. Department of Justice opened an antitrust investigation into the proposed settlement. A month later, Microsoft, Yahoo!, Amazon, and some writers groups came together to fight the settlement as the Open Book Alliance, under the leadership of Gary Reback, the attorney who pushed the antitrust case against Microsoft in the 1990s. A dislike for the deal also brought together strong believers in copyright with legal scholars like Pamela Samuelson, a professor at the University of California, Berkeley, who favors more exceptions to copyright and worries that the settlement will give Google a monopoly on out-of-print books. (Samuelson would have preferred that Google didn’t settle and instead tried to set a precedent that scanning books qualifies as fair use.) And the Google Books settlement is certainly the only legal deal in history to draw objections from both DC Comics and the Federal Republic of Germany—the former due to concerns that it would cover material it might want to put back in print, and the latter because it would apply to works by German authors published in the United States.
In March 2011, after studying the amended settlement for more than a year, judge Denny Chin rejected it, partly on the grounds that it would “grant Google significant rights to exploit entire books, without permission of the copyright owners.”18 Although Google set out to display snippets of books, the settlement would have given it additional rights. As Chin pointed out, Google would have also gained a decisive advantage in the search business that Microsoft and other companies would never be able to match, since they wouldn’t be able to scan copyrighted books without risking stratospheric damages. If information is the oil of online search, books contain the largest untapped reserve, and the Google Books settlement would have locked up the drilling rights. “The establishment of a mechanism for exploiting unclaimed books,” Chin wrote, “is a matter more suited for Congress than this Court.”19
Google and authors and publishers could submit another amended settlement to the court, but Chin objected to the opt-out provision at the heart of the arrangement, which is important to Google. The suit could also proceed, with considerable conseqences for the losing side. More likely, Congress will again consider the issue of “orphan works”—those whose copyright holders can’t be identified—in a way that would allow any company to digitize books as long as it conducted a search for their owners.
It’s hard to know how Congress would act, and Google and other technology companies could easily take the opportunity to lobby for more sweeping changes to copyright. In the worst-case scenario, a law would allow technology companies to digitize books after a cursory search for rights holders. But legislation could also make the bookselling market more competitive. “A huge concern in the publishing industry is that there’s going to be only one online portal that will have any significance for the sale of books,” Aiken says. “A world in which Amazon controls all the growing parts of the industry is a dangerous one for authors.”
The publishing world seldom conducts business with the shark-eat-shark aggressiveness common in Hollywood. But it may now have as much or more control over its online future, with two major technology companies selling digital versions of its products, plus Barnes & Noble getting into the e-book business and Sony selling digital reading devices. Unfortunately, publishers also get the most resistance to the prices of their products online.
Like Henry Blodget, most readers assume that books are priced according to the cost of printing, binding, and shipping them to stores. Since hardcovers almost always run between $25 and $30 and trade paperbacks usually sell for about $15, there’s a tendency to think those prices have something to do with packaging. That’s why a $15 e-book sounds so expensive; there is no packaging.
So what do we talk about when we talk about books? If you think about it, $25.00 is a lot of money for a sheaf of paper held together by cardboard. But that’s not what most people pay for, of course; they buy the text itself. And that’s the largest source of cost as well as value. Publishers spend only about $3.50 to print and distribute a hardcover.20 (Hardcovers are sold on a returnable basis, so the costs of retailer returns of unsold stock adds about another dollar or so to the price of each book, depending on how they’re accounted for.) In many cases, the biggest expense in traditional book publishing is royalty payments to the writer, who can make more than the publisher (especially when sales don’t make up for a large, nonreturnable advance against royalties). But those expenses don’t change much in the digital world—Macmillan actually raised author royalties on e-books—nor does the need for editing or marketing.
As for the difference in price between hardcovers and paperbacks, a better binding doesn’t cost another $10 any more than the larger seats and better food in first class cost airlines an extra $1,000 per passenger. Like first-class travel, hardcovers represent a way to maximize revenue in order to meet a high fixed cost—in this case, the money required to acquire, edit, and market a book. In many cases, publishers wouldn’t be able to sell enough $15 paperbacks to make back their investment. Hardcovers let them do that. A book’s price doesn’t come from its packaging—its packaging comes from the price.
Publishers were “windowing” new releases before Hollywood. Most readers buy hardcovers because they’re available earlier, not because of the binding, and publishers want to keep that strategy as they move online.21 They could make titles available initially for $15, lower the price to $8 or $10 over the course of a year, then bring them down further when they wish. Airlines use a more complicated version of the same strategy: they try to get passengers to pay as much as they can without losing potential customers. Planes fill up, while books don’t, but the idea is the same: defray a high fixed cost by pricing in order to maximize revenue.
Traditionally, this strategy also served the interests of retailers. But like Walmart, which sometimes sells certain titles below cost to draw shoppers into stores, Amazon doesn’t really need to make money on book sales. Until June 2010, when Amazon reduced the price of the Kindle in the face of competition from Apple, the company was making between $80 and $100 on each $259 device, according to analysts.22 So it made financial sense to sell popular e-books at a loss in order to drive demand for the reader, much as Apple forgoes significant profit on music sales to push the iPod and iPhone.
Therein lay a problem for publishers. If Amazon had kept using its Kindle profits to compensate for lower margins on e-books, booksellers without an outside source of profit might not be able to compete. That would limit the number of e-book retailers in a way that would ultimately hurt publishers, just as iTunes dominated the digital music market because it was too hard to make money selling music at ninety-nine cents a track without also having a device to promote. “That $9.99 price may have been reinforcing one player’s business model, so it wasn’t a spur to competition in the market,” Sarnoff says. “Other players that wanted to be good sellers didn’t want to have the loss-leader approach, and this was the genesis of the so-called agency model.” The Authors Guild has said that, by changing the dynamic of the e-book market, Apple made it possible for Barnes & Noble to compete.
Some technology executives say publishers need to cut e-book prices to discourage illegal downloading, which has become a problem in the book business now that digital reading devices have become more popular. But the experience of record labels suggests that some consumers will pirate books no matter what their price, others will always buy them, and the behavior of the rest depends more on convenience than anything else.
Amazon tells publishers that by cutting prices, they’ll sell more books, and it promotes the fact that Kindle owners buy three times as many books as they did before purchasing the device. But although several publishing executives do believe Kindle owners buy more books overall, and Amazon says it drives sales of older books from popular authors, this statistic is deceptive, since it compares digital and physical sales only from Amazon itself. Imagine that a regular reader of physical books does about a third of his shopping from Amazon.com and the rest at local stores. If he buys a Kindle, he’ll probably change his habits, since e-books are cheaper and more convenient to buy—and he can buy them only from Amazon. So the online retailer might sell three times as many books to that reader, but that doesn’t mean more titles are being sold—just that Amazon gets a larger share of them.
Taking a larger share of book sales is a key part of Amazon’s Kindle strategy. In the physical book business, Amazon has to compete for every sale. In the digital world, once Amazon sells a Kindle, it knows customers will buy almost all their books from its site. So it wants to speed the move to e-books as fast as it can. Publishing executives would prefer the transition to proceed more slowly. Bookstore displays are an important part of their promotion strategies, as are retailer recommendations, especially for new fiction. And as sales of physical books decline, the production expenses for each physical book will rise, since costs for warehousing and shipping will be divided among fewer units. The adjustment will be far easier if it happens more slowly.
“If people keep moving to e-books because of extremely low prices, then physical bookstores will be in trouble,” Napack says. “Throughout history publishers have counted on retailers wanting to keep prices up in order to make money. If the reading public is willing to pay $28 for what’s between the covers of a book, it troubles me that someone came into the market and, for their own strategic reasons, consistently undervalued it.”
Amazon seems to see publishers the way technology executives see record labels: as companies that manufacture and distribute physical goods no one needs anymore. Naturally, it wants to deal directly with writers and cut out what it sees as the middleman. So, starting in June 2010, the company began offering 70 percent royalties to authors and publishers who agreed to sell their books on Amazon’s Kindle platform for between $2.99 and $9.99, and at least 20 percent below the print version. In December 2009, the business advice writer Stephen R. Covey signed a deal to make two of his books available as digital editions exclusively on Amazon.com for a year.23 (Print versions of the same titles will still come out through Simon & Schuster.) And in December 2010 the marketing author Seth Godin announced that he would write and package a series of “Idea Manifestos” using Amazon’s distribution for both digital and traditional print editions.
Like Radiohead and Nine Inch Nails, Covey and Godin are stars of a sort, so they don’t need the promotion and marketing that publishers provide. They can get higher royalties by dealing directly with Amazon. And since business advice authors generally make good money performing live, they can cut the price of their books to reach more potential speaking clients.
But what does Amazon offer less established writers? Just as Napster didn’t really replace labels, an online retailer can’t replace publishers—just their printing and distribution infrastructure. (Technology companies never say they want to disrupt the shipping and warehousing business; it’s not nearly as cool.) Like record labels, publishers do much more than manufacture and distribute; they advance money to creators, shape projects, and help them find an audience. Doing those things well is far more complicated than getting a book to a store on time.
Amazon does an efficient job fulfilling demand for books, which small publishers sometimes have trouble with. But a more important function of publishers, like labels, is generating demand, which Amazon doesn’t have much experience with. While the Internet makes it less expensive to reach prospective readers, by using tools like Twitter and Facebook, so many authors are taking advantage of this that breaking through the clutter still requires expertise, money, or some combination of the two. Independent writers can hire independent editors or marketers, of course, but that means Amazon’s higher royalties get offset by expenses that publishers once covered.
It’s hard to write a book about the book business without referring to my own experience, so I’ll say up front that there’s no way I could have written this without a publisher. Even if I had the money to hire an editor and a marketing expert, it would have meant gambling a substantial amount of time and money on a single project, without having any idea how much, if anything, I’d make on it. Publishers spread out their risk among scores of books in a way that writers can’t.
Although technology has revolutionized the process of distributing books, it hasn’t fundamentally changed the process of writing one. The Internet makes it easier to find information quickly, and computers have transformed the editing process. But reporting and writing are still painstaking work, and they can get expensive as well. Work on this book took me to Los Angeles, Washington, D.C., London, Brussels, Copenhagen, and Düsseldorf. I hired reporters who each gave me a few days of help with interviews and translations in German, French, and Swedish; paid a fact-checker and a lawyer to make sure the book was accurate; and worked with a part-time researcher to help me make my deadline (or at least miss it by less than I might have otherwise). All of this costs money, which authors need while they write, rather than later. That’s basically what Ken Auletta explained to Sergey Brin: it’s not that easy to give a book away, since it’s harder than most people realize to write one.
Just like in the music business, most of the ideas to change publishing run into some version of this problem. Writers could give away books and earn money on the speaking circuit, but that works best for authors like Godin, whose books are practically written in PowerPoint form already. Even if that works, what would authors live on for one or two years—or, in some cases, more—while they write? It’s far more practical to borrow money against the delivery of a copyrighted work than against the vague prospect of speaking fees.
Unlike Hollywood, book publishing hasn’t exactly been swimming in money for some time: annual sales grew 1.6 percent between 2002 and 2008,24 independent bookstores are closing, and Borders filed for bankruptcy in February 2011. But physical books may decline more slowly than other forms of traditional media. Unlike CDs and DVDs, books aren’t tied to a particular platform; they work just fine on their own. It’s hard to find anyone who misses the cheap, plastic packaging of CDs or DVDs, but books are appealing objects to own and display. (It could take years for people to figure out what else to put on bookshelves.) Priced right and marketed well, books could have a slow, steady decline instead of a steep, sudden drop. That would be a huge help to publishers, who need more time to prepare for a digital market.
One question is what kinds of books will sell in that market. In October 2010, Amazon announced it would launch Kindle Singles—written works that are longer than a magazine article but shorter than most books.25 The company suggested they would range from ten thousand to thirty thousand words, “the perfect, natural length to lay out a single killer idea, well researched, well argued and well illustrated—whether it’s a business lesson, a political point of view, a scientific argument, or a beautifully crafted essay on a current event.”26 Amazon sells Singles from publishers as well as direct from writers, but the business model seems better suited for the latter, and Amazon’s descriptions of Singles—“lesson,…argument,…essay”—don’t exactly make it sound like a format for reported work that would require significant time and money. In most cases, it seems unlikely that editing would be part of the package.
“Their interest is fundamentally different from yours as a publisher—they’re interested in driving traffic,” Napack says. While publishers commit to particular authors and books, Amazon doesn’t care what it sells, as long as it can be read on a Kindle. That’s a significant difference. “There was a time when the advantage of being a big publisher over little publishers or an individual is that we had the infrastructure to get a book printed, ship it to a warehouse, ship it to a bookstore, and take returns,” Napack says. “In the digital world, that becomes less important. We have to rely again on what we did well—what should have been a core advantage all along—which is editing and marketing books.”