We’ve all heard the argument that e-books should be cheaper because they don’t have printing, shipping, and warehousing costs. Blogger Deanna McFadden, who works in digital publishing, takes exception to writer Michael Chabon’s recent statement that it’s unfair publishers should limit authors to the same royalty rate as paper books as for “an e-book that costs them nothing to produce.”

McFadden attributes this sort of statement to ignorance of all the costs and hard work that go into readying a book for digital publication, and grumbles that “Chabon essentially thinks my role, that of e-book person, is essentially worthless”.

Let’s take again the idea that for those two earlier books that essentially cost a publisher “nothing” to produce and actually look at what’s involved. Because the books are older, and let’s assume they are a fair bit older, no digital files exist. That means a gross, labour-intensive few days of scanning the text into a file. Then there’s the time it takes to clean up that file, to strip out all the gunk that scanning creates and make sure the e-text is as accurate as it can be to the p-text, again, more time. And sure, there are companies all over the developing world that will do this terrible work for a low cost (and my thoughts on that are, well, far too long to include in this already too long post). In the end, you’ve got the work into a digital format and it’s ready to be converted to epub, more buttons are pushed, coders come in and clean everything up, and you’ve got an ebook that’s ready to go. It’s proofed and checked and proofed again. Metadata is built, ONIX is created, files are FTP’d to the vendors, Amazon’s mobis are created, various different formatting things are checked, and then the book is for sale — but, hey, apparently, this costs all of us nothing to produce, right? There’s no time or energy or effort or anything put into pushing a button and magically having an ebook show up around the world, naw, it’s nothing compared to boxes and gasoline and shipping and warehouses, right? And I’m even simplifying things here for the sake of making a point — not a week goes by where all of our books end up in the right place, where none of them have some sort of technical issue, where there’s not a problem with data that needs to be solved — it’s complex, time consuming, and, at times, really frustrating.

McFadden does admit she may not be putting Chabon’s words into the right context. And I think that’s right. She’s kind of comparing apples and oranges. Yes, e-books have fixed costs of production, of the sort she describes. So do paper books. And they share many of the same costs, and (as McFadden notes) since e-book publishing and print publishing are often treated as two separate things instead of one thing with two separate products, they often duplicate those costs.

But what Chabon is talking about when he says they “cost nothing to produce” is the marginal cost. It costs a certain amount of money to run off each paper book, to ship it, to store it, and possibly to ship it back and destroy it. But e-books have such a miniscule production cost per unit—a few watts of electricity, a few kilobytes of bandwidth—that they effectively cost nothing to run off each digital copy. So if we assume that fixed costs are equal, e-books will earn those costs out much faster while not adding any marginal ones.

That doesn’t mean that the work of someone preparing those e-books doesn’t have value. It just means that when you start selling a lot of them, there’s a lot more room to cut price and/or raise royalty payments on the electronic version. Or so the perception goes, at least. There’s still considerable argument over whether that is actually true.


  1. Just from a rhetorical standpoint, probably not a great idea to include “FTPing a file” on a list of things that require a lot of work. It makes me doubt the degree of difficulty of the others.

    “Metadata is bult.” Ooh, typing>. Author, title AND publisher? Where does one find the time?

  2. I can’t comment on McFadden’s company, but several of the books I have purchased for the Kindle don’t give any indication of having even been spell-checked in a word processor, much less proofread.

  3. A couple of people spending a week to scan, proof and prep an ebook sounds about right. If the ebook goes on to sell 5,000 copies that equates to somewhere around 50 cents a copy. If it sells 50,000 that equates to 5 cents per copy.

    Now if publishers had any understanding of how digital goods scale they could price their back catalog at $1.00 an ebook and they might sell 500,000 copies to avid readers who would love to build up a big collection cheaply. At that point the cost of creating the ebook would drop to around half a cent per book.

  4. And another thing: Screen book advocates disparage print as “dead trees”. Print books are organic and biodegradable, but screen book devices are “toxic landfill”. And another thing; our print books are products of US based manufacturing while screen devices are imports. We could also bridge to discussion of exploited labor.

  5. I think ebooks and print books are apples and oranges; with ebooks the cost is fixed per title and near zero for each copy. With print books, there is a fixed cost but each copy still bears a substantial cost for printing, shipping, and (most of all) returns. I can see publishers offering one royalty rate for the first x number of ebook copies and a higher rate after that. If they don’t then big name authors are going to want to self-publish their backlists instead of letting the publisher do it.

  6. I think Chabon’s comments are taken out of context. He is/was negotiating the digital rights to his backlist (and was unhappy with offered terms).

    For someone like Chabon, when the books have already been edited and proofed etc., with the sales he is likely to generate, the *per-unit* cost of releasing those e-books is essentially zero. He’s saying that, because of that, he should get a higher royalty rate.

    And he’s right.

  7. If it takes her “a gross, labour-intensive few days of scanning” to get a single book into tiff or jpg format, I have to wonder how badly she does everything else.

    With a *good* scanner setup–which is not cheap, and certainly not everyone has access to–scanning a 400-page paperback with the spine chopped off so the pages run through the feeder takes about eight minutes. With a DIY two-camera setup, maybe 2 hours… allowing for half an hour to get the settings tweaked just right. Oh, maybe double that if you’re new to it. With a cheap scanner (they’re slower) and a book being pressed flat to scan every page, it can take several hours. 45 seconds to a minute per page adds up.

    A single book could only take “a few days” if the scans are (1) high-res color (say, 400-600dpi) and (2) larger-than-average page sizes. A 300-page trade comic book compilation scanned in archive-quality color on a low-speed flatbed scanner might take a couple of working days. Somehow, I doubt those are the kinds of books she’s talking about.

    I get really tired of publisher apologetics that claim “but we have to charge more! Our rents are so high and our work processes are so inefficient that we’d go broke charging less!”

    (FTPing to vendors… um. Yes, various venders have different file-exchange systems, and someone has to learn how to navigate them, and some are amazingly slow & awkward… but we are talking “maybe one hour per title” added to the production time. For all the vendors total. If you’ve got so much of this going on that it’s costing you administrative-staff time, hire a damn intern already.)

  8. @elfwreck; Agreed on the pro-grade scanners.
    More, even if you don’t destroy the book being scanned, you can still scan a 400 page paperback by hand in less than an hour with a decent $200 *consumer* scanner. OCR’ing the scanned images takes less than an hour on a vintage Pentium 4 so they’d have to have really bad systems to need more time. Both are things a AA or intern can do in between their real duties before passing them on to the proofreaders.
    Anybody with a *lot* of scanning to do can afford better gear and a full staff of minimum wagers to get the job done right. *IF* they care about doing it right.
    Pretending otherwise simply proves they *don’t* care.

  9. I regret to have to say that this is a really disingenuous article who’s motives must be questioned.

    Firstly the original statement by Chabon was clearly, when read fairly, intended to refer to the production costs of an eBook and not the preparation/editing. A paper book requires paper, ink, printing, distribution, warehousing and stocking in retailers. An eBook has no such costs at all except uploading to the eRetailer. In that he is absolutely correct.

    Secondly this article then proceeds to chose the most expensive form of producing an eBook, when there is no digital file. This of course does NOT apply to the millions of titles published in recent years – where all of those ‘production costs’ she goes into at length do not apply.

    Thirdly even when those costs do apply – when the paper book has to be scanned. Modern scanner software is incredibly accurate. Even when it makes mistakes, a reasonably educated student can pick up 99.9% of the errors in a few hours, and a scan by a more seasoned pro can complete the rest in short shrift. That is the fact of the matter. And when the costs of a few hours of such work is added to the hour or so of input time for meta dat and format conversions, the cost, spread over … let’s say a 1,000 copies … is trivial.

    This pompous attempt to hyper inflate these costs is being pumped up by publishers all over, but it isn’t making any impact on people because there are too many people, like me, who know about costs and know the truth.

  10. Binko wrote: “A couple of people spending a week to scan, proof and prep an ebook sounds about right.”
    Having seen my son of 19 recently be involved in a writing project in his school in cooperation with a local publisher, my belief that your estimate is grossly excessive (no offence!) has been confirmed.
    A scan using a modern tech scanner takes less than an hour. The latest OCR software is incredibly accurate, though in special format cases it does make bigger mistakes I agree, but those are the exceptions.
    A fair quality read through, with notes or errors would take about 4 hours. Another couple of hours to make those changes. Next thing is a final slow read through, which will take about 6 hours if we are generous. An hour to make those corrections. Then an outsourced read through by an amateur panel of readers who get free reads in exchange produces a 100% perfect product. The job all done in about 14 paid hours max. All done by any reasonably educated student with reasonable english for about 15 dollars an hour ? 15×14= $210 spread over an initial issue of 1,000 copies = 21c a copy. Duh.

  11. If it’s seriously taking them that long and costing so much to create ebooks, they should just hire the people at a company like 52 Novels. They claim to be able to get a manuscript ebook-ready for as little as $250 for a straight manuscript; by my calculations, they could get a 300 page old-school hand-scanned novel done for $1250 at the *most*, as their pricing is quite transparent–there’s no mystery involved. Compared to the $30,000 investment needed to print a 10,000 copy treebook run (according to Curt Matthews at IPG)? Yes, it does–or can, at least–cost practically nothing to put an ebook out there–even one that needs scanning in.

  12. @Howard; you’re awfully generous. 😉
    $15 an hour for interns to proofread texts will swamp you with applicants.
    But your point stands: it is hardly very expensive to convert from print to digital.

  13. To me, this is like arguing that music CDs cost $x.xx to produce (more like $x,xxx,xxx.xx) but since mp3 files are electronic and don’t have to be shipped to music stores, they should be free.

    Both arguments are severely flawed.

  14. Alan, try discussing the actual topic under consideration rather than ranting against something that nobody has said.

    Nobody, literally nobody, is arguing that ebooks or music or anything else should be free unless they are public domain works or the creator wishes to make them freely available.

    What people ARE arguing is that items that have a lower production cost, a lower distribution cost yet more restrictions on use should indeed cost less. If you disagree with this premise please feel free to present some intelligible arguments.

  15. I do agree that they should cost less; as a returning college student I have been appalled to discover that while, in some cases, the textbooks required for certain classes are available for the Kindle, more than a few of them cost the same as the physical textbook — to the tune of ~$140.

The TeleRead community values your civil and thoughtful comments. We use a cache, so expect a delay. Problems? E-mail newteleread@gmail.com.