I want to tell you a story about a book that doesn’t exist anymore. It was a novel, originally published in Spanish by a small press in Buenos Aires in 2003. The author was a retired schoolteacher. The print run was 500 copies. The book got a few positive reviews in Argentine literary magazines, sold most of its print run over about three years, and then quietly went out of print. The author died in 2011. By then, you couldn’t find a copy for less than forty dollars online. Today, you’d be lucky to find one at all.

I know about this book because a translator I work with brought it to my attention. She’d picked up a copy at a used bookstore in Montevideo and thought it was extraordinary. She translated three chapters as a sample and sent them to me. They were extraordinary. The prose was precise and strange, the characters fully alive, the structure inventive without being gimmicky. I wanted to publish it. But the rights were tangled. The author’s estate was managed by a nephew who lived in Seville and didn’t respond to emails. The original publisher had gone out of business. The legal work required to secure translation rights would have cost more than we’d earn from the book in its first two years.

We published it anyway, eventually, after eighteen months of negotiation. It sold about 2,000 copies. And I tell this story because it illustrates something that the “free ebooks” conversation consistently ignores. The real cost of free ebooks isn’t piracy. It isn’t Amazon’s market power. It isn’t the devaluation of individual titles. The real cost is all the books that never get made because the economics don’t support them.

Let me back up. When people talk about free ebooks, they usually mean one of three things. First: piracy, the unauthorized distribution of copyrighted ebooks through file-sharing sites, Telegram channels, and dedicated piracy platforms. Second: the perception that ebooks should be free or nearly free, driven partly by Amazon’s early $9.99 pricing, partly by the existence of free books in the public domain, and partly by the general internet-era expectation that digital content should cost nothing. Third: the actual free ebooks that publishers and authors give away as marketing tools, through services like BookBub, permafree first-in-series strategies, and Kindle Unlimited’s subscription model, which isn’t technically free but feels free to the reader.

I have thoughts about all three, but I want to focus on the second, the perception problem, because I think it does the most damage.

The perception that ebooks should be cheap or free is based on a misunderstanding of what makes a book expensive to produce. People look at a physical book and see paper, printing, binding, shipping. They look at an ebook and see… nothing. No physical materials. No manufacturing. No shipping. Just a file. And since the file costs almost nothing to distribute, they reason, it should cost almost nothing to buy.

This reasoning misses about 80% of the cost of producing a book. The physical manufacturing of a trade paperback, including printing, binding, and the cost of the paper itself, typically accounts for about 10 to 15% of the retail price. For a $16 paperback, that’s maybe $1.50 to $2.50 in manufacturing costs. The rest of the retail price covers the author’s advance and royalties, editorial work (developmental editing, copyediting, proofreading), cover design and interior layout, marketing and publicity, overhead (rent, salaries, insurance, all the things that keep a publishing company functioning), and distribution and retail margins. None of those costs go away when you produce an ebook instead of a physical book. The author still needs to be paid. The editor still needs to edit. The designer still needs to design. The marketing still needs to happen. The company still needs to exist.

There are some savings, yes. You don’t pay for paper, printing, or warehousing. You don’t pay for returns (the physical book industry’s bizarre practice of allowing bookstores to return unsold books for full credit, which costs publishers billions every year). But those savings are partially offset by ebook-specific costs: digital conversion, DRM implementation, platform fees (Amazon takes 30% of ebook revenue for books priced between $2.99 and $9.99, and 65% for books priced outside that window), and ongoing digital distribution management. When you add it all up, the savings from eliminating the physical product amount to maybe 20 to 25% of the total cost. That’s meaningful, and it’s why ebooks are typically priced lower than hardcovers. But it doesn’t justify the expectation that ebooks should be $2 or free.

The author’s share is the part that concerns me most. The standard ebook royalty for a traditionally published author is 25% of net revenue. “Net revenue” means the amount the publisher receives after the retailer takes its cut. So for a $12.99 ebook sold through Amazon, the math works like this: Amazon takes 30%, leaving $9.09 for the publisher. The author gets 25% of that $9.09, which is $2.27. Out of that $2.27, the author’s agent takes 15%, leaving the author with $1.93. For a book that took two years to write.

Now drop the price to $4.99. Amazon still takes 30%, leaving $3.49. The author’s 25% is $0.87. After the agent’s cut, the author takes home $0.74. Per copy. For a book that took two years to write. To earn the median American household income from ebooks at that price, an author would need to sell roughly 75,000 copies per year. The vast majority of books, even reasonably successful ones, sell nowhere near that many copies in any format. The median traditionally published book sells about 3,000 copies across all formats in its first year.

This is why most authors can’t support themselves through writing. And this is where the cost of cheap ebooks becomes tangible. When ebook prices drop, author incomes drop. When author incomes drop, fewer people can afford to write full-time. When fewer people write full-time, the books that get written are increasingly produced by people who have other sources of income: trust funds, working spouses, academic positions, or day jobs that leave them with limited time and energy for writing. The diversity of voices in published literature narrows. The people who can afford to write tend to be people with financial privilege, and their work tends to reflect the experiences and perspectives that come with that privilege.

I’ve watched this happen in real time over the past decade. The mid-list, the category of books that sell respectably but not spectacularly, has been hollowed out. Publishers concentrate their resources on potential blockbusters and let mid-list titles fend for themselves. Authors who would have built sustainable careers publishing one book every two or three years now find that each book earns less than the last, until eventually the economics force them to stop writing or to shift to a more commercially viable genre. I’ve lost authors this way. Good authors. Authors whose books mattered to their readers, even if those readers numbered in the thousands rather than the millions. The books those authors would have written will never exist. That’s a cost, even if it doesn’t show up on any balance sheet.

Piracy compounds the problem, but I don’t think it’s the main driver. The publishing industry’s own estimates of revenue lost to piracy vary wildly and are probably inflated. Most people who download pirated books wouldn’t have bought them at full price. Some piracy probably even drives sales, as readers discover authors through pirated copies and then buy subsequent books legitimately. I don’t condone piracy, and I do think it hurts individual authors in specific ways (particularly authors of niche nonfiction, where the entire addressable market might be 5,000 people, and losing even a few hundred sales to piracy represents a significant percentage of total revenue). But piracy is a symptom, not the disease. The disease is the broad cultural devaluation of written work.

This devaluation didn’t happen by accident. Amazon deliberately conditioned readers to expect low ebook prices. The company’s $9.99 pricing for new releases in 2007 and 2008 was a loss leader strategy, selling ebooks below cost to drive Kindle adoption. It worked. Kindle became dominant, and readers internalized $9.99 as the “right” price for a new ebook. When publishers pushed back and secured agency pricing (which allowed them to set their own ebook prices, typically $12.99 to $14.99 for new releases), readers were furious. Amazon encouraged the fury by posting notices on affected book pages that said, essentially, “this publisher is charging you more than we’d like.” The damage was done. A generation of readers had learned to think of $12.99 as overpriced for a digital file.

The Kindle Unlimited subscription model takes the devaluation further. For $9.99 per month, readers get unlimited access to a library of over two million ebooks. That’s a spectacular deal for readers. It’s a much harder deal for authors, who are paid from a shared pool based on pages read. The per-page rate fluctuates but typically works out to about half a cent per page. For a 300-page novel, that’s about $1.50 per read-through, if the reader finishes the book. Many don’t. The system incentivizes long books (more pages, more revenue) and frequent releases (stay visible in the algorithm). It’s well-suited to high-volume genre fiction. It’s terrible for the kind of literary fiction and serious nonfiction that takes years to research and write.

I want to be fair to Amazon here. They didn’t create the expectation that digital content should be free. That expectation existed long before the Kindle. The music industry went through the same thing with Napster, iTunes, and Spotify. The news industry went through it with free online articles. The pattern is familiar: a new distribution technology makes copying trivially easy, prices collapse, creators struggle to earn a living, and the industry eventually settles into a new equilibrium that works better for consumers and platforms than for the people who make the things being consumed. Publishing is following the same trajectory, and Amazon is facilitating it, but the underlying forces are structural.

At ScrollWorks, we’ve made some deliberate choices in response to all this. We price our ebooks at $11.99 to $13.99, which is at or near the hardcover discount price. We’ve gotten pushback from readers who think that’s too high for a digital file. I understand their perspective, but I’m not going to apologize for it. Our ebook prices reflect the real cost of producing the book: the author’s years of work, the editor’s months of engagement, the designer’s skill, the marketing team’s effort. If we priced our ebooks at $4.99, we’d either have to cut author advances (hurting writers), cut editorial investment (hurting quality), or accept losses that would eventually put us out of business (hurting everyone). None of those options serve readers in the long run.

We also don’t participate in Kindle Unlimited. The per-page payment model doesn’t work for our books, which tend to be densely written and moderately length. And we’re uncomfortable with the exclusivity requirements. Amazon’s KDP Select program, which is the gateway to Kindle Unlimited, requires that the ebook be exclusive to Amazon for 90-day periods. We believe our books should be available on every platform: Apple Books, Kobo, Barnes and Noble’s Nook, Google Play, and independent bookstore platforms like Libro.fm’s ebook partner. Exclusivity with any single retailer concentrates too much power and limits readers’ choices.

I know this stance costs us some sales. Books in Kindle Unlimited get visibility boosts in Amazon’s algorithms that non-KU books don’t. Some readers only read through their KU subscription and won’t buy individual ebooks at any price. We’re leaving money on the table. But I think the long-term health of the book industry depends on publishers maintaining diverse distribution channels and resisting the gravitational pull of any single platform, however dominant it may be today.

The free ebook conversation also needs to account for libraries, which are the original “free books” institution and which are under increasing pressure in the digital era. Libraries license ebooks from publishers, typically at prices significantly higher than consumer prices (often two to three times the retail price) and with restrictions on how many times a single license can be lent (often 26 loans before the license must be repurchased). Publishers justify these terms by arguing that ebook lending, unlike physical book lending, involves no degradation. A physical library book wears out. An ebook doesn’t. Without loan limits, a single ebook purchase could serve an infinite number of patrons.

There’s logic to that argument, but I’m not entirely comfortable with it. Libraries are public goods. They provide access to books for people who can’t afford to buy them. Making ebook access difficult or expensive for libraries hurts the people who most need free access to books. I don’t have a solution that’s fair to everyone. Authors need to be paid. Publishers need to cover their costs. Libraries need affordable access. Readers need options. The current system is a series of imperfect compromises, and I suspect it will remain that way.

What I keep coming back to is that book I described at the beginning. The Argentine novel that almost didn’t get translated because the economics were too fragile. Every link in the chain that brought that book to English-language readers depended on someone being willing to invest without a guaranteed return. The original publisher took a chance on an unknown author. The translator spent weeks on a sample she wasn’t being paid for. I committed to a project that I knew would lose money in its first year. We did it because we believed the book mattered. But “believing the book matters” is only sustainable if the broader economics of publishing allow for risk-taking. When ebook prices collapse, when author incomes shrink, when publishers tighten their budgets, the books that get cut first are exactly these books: the ones that matter but don’t have obvious commercial potential.

Free ebooks aren’t free. Someone always pays. And when readers demand cheaper and cheaper digital books, the people who pay are the authors who can’t afford to keep writing, the editors who get laid off when margins shrink, the translators whose rates stagnate, the small publishers who close their doors, and ultimately the readers themselves, who end up with fewer and less interesting books to choose from. The cover price of an ebook isn’t just what you pay for a file. It’s what you pay to keep the ecosystem functioning that produces the books you want to read. I think that’s worth more than $4.99.