An article in the NY Times making the rounds on Twitter (via Colleen Lindsay) makes the case for a high price point for e-books.  I responded to it on Twitter and Facebook and thought I’d repost some of that response here.

My basic critique boils down to this:  I am uncertain (because I’ve seen no data) that e-books are substitute goods for print books in the market as it exists today.  If they are substitute goods — and sales of e-books more or less perfectly substitute for sales of print books, i.e. the one cannibilizes the other — then I find the logic of the article persuasive.  If they aren’t substitutes, then I think the article errs in justifying the spread of certain fixed costs onto the cost to produce e-books (as opposed to using incremental cost as the starting point for e-book pricing).

An example just to illustrate the point:  Publisher A has 1,000 print titles and none are available as e-books.  Publisher has built its business model on this reality, such that certain costs associated with those print novels (overhead, art, marketing, warehousing, etc.) are built into the per unit price of the print novels.  Tomorrow, Publisher A decides to make all of those novels available as e-books.  The incremental cost of that decision is quite low (per unit royalties to the author, and whatever conversion to the new format may cost).  At that point, the books aren’t remarketed, recopyedited, or re-covered with new art.  Given that, is it justified to factor in, again (because remember, overhead, art, marketing, etc. have all been factored into the price of the print version already), all of the costs associated with the book when it was only in print format?

Now let’s consider an ex ante example.  Publisher A has just bought a book (but has not yet published it) and is making price and format decisions based on sales expectations.

Let’s assume that e-book sales are additive to sales overall (a big assumption, I realize, but, as I mentioned above, I haven’t seen any data one way or another; my own very limited experience suggests that e-books are additive to sales, though).  Why might this be so?  Well, because many people may buy a book in both print and e-book formats, and some people may buy an e-book who might never have bought the print version (due to availability in international markets, say).

In that situation, it seems to me that either (a) the incremental cost of producing the e-book is the right starting point for price, in which case the e-book will be priced well below its print analog; or (b) the decreased costs associated with e-book production should drive down the cost of producing the novel overall, resulting in a somewhat decreased price in both formats (but the price of both formats will be roughly equal).

Point (a) is intuitively obvious.  Point (b) may be less so.

Look at this way.  If a publisher has fixed costs (marketing, editing, art, etc.) of $1,000 and expects to sell 1,000 print units (with no e-books) then the fixed cost per unit is $1. If fixed costs stay the same when e-books are added to the mix (and I think they would, largely), and adding e-books increases sales to 1,500 units, then the fixed cost per unit is $0.66 per unit.  Theoretically, the publisher could then lower prices on the books in all formats to account for this different cost structure.

Of course, as I mentioned above, if e-books simply cannibalize print sales, such that sales in our example above remain at 1,000 units even with e-books, then fixed costs remain at $1.00 per unit and no change in price for either format is warranted (i.e., the prices should be roughly equal and about where they are today).  But I think the jury is still out on that question, though publishers typically assume cannibalization in the various analyses that I see (including the article above).

Thoughts?

After I posted some of this on Twitter, friend and colleague Matt Forbeck observed that the publishers may be looking ahead to an era in which e-books will be the norm rather than an add-on for early adopters (i.e., a time when e-books will not be additive to sales overall).  I think he may be right, and, if so, that’s entirely reasonable.  But if so, I’d prefer that the various analyses be more forthcoming with their assumptions.

It’s also possible that I’m missing something obvious, since I’m far, far removed from my limited education in accounting and economics.  🙂