Wednesday, November 17, 2010

Pricing digital textbooks – and how versioning could increase profits

(Post by Jinho Suk, Jungmoo Park, Sangouk Kim, Allan Jaenicke, & Arkajit Dey)

The market for digital textbooks is expected to grow exponentially over the coming years, but we see the current approach to pricing as a significant barrier to rapid growth.

In particular, we have focused on the effect of the strong 2nd hand textbook market. A survey conducted at MIT Sloan shows that 60% of textbooks are purchased 2nd hand. Through mining pricing data, we conclude that the presence of the 2nd hand market has a significant impact on pricing. On the one hand, it has the effect of raising prices through indirect appropriability: Books can be resold at around 70% of the original price. We also see less price competition and less variation in pricing across new, used and digital formats then e.g. in the ‘fiction’ category, which has a weaker 2nd hand market.

Below is an example of our findings:


  1. I really liked your presentation.

    Have you by chance made any research on renting digital e-books? I know someone who does this in a college bookshop with normal books. He rents them for 30% of the list price for a Semester. According to his experience, he can rent the book usually 5-6 times and makes a higher profit for each book than with selling. So far, he has been pretty successful with his policy. At what price do you think should be a digital textbook that you rent e.g. for a Semester? I mean, the marginal costs for the producers are zero...

    I am so far really confused why the pricing e.g. at Barnes and Noble is so weird:

  2. Guys, how did you obtain Amazon data?

    More generally, do you think that with the increasing dispersion of ebooks we will see much more of an a la carte model where we will be buying individual chapters (think songs) vs entire textbooks (think albums)?

  3. Image isn't showing up for me

  4. I assume this was a class presentation? It sounds interesting. Was more empirical analysis detail presented?

    Also of interest - one fundamental underlying assumption we may make is that optimal pricing is being chosen by Amazon (stock is cleared at maximum surplus)... but my guess is Amazon is "pulling levers" to see what happens - all just experimentation at this stage with learning’s from some product genres or consumer segments possibly being more mature – which would make for some interesting potential for pricing algorithms. Stock differentials (over a series of fixed time intervals) might be interesting data to add to the analysis.

    Stepping back from this though - it must be awesome to be working in the backroom at Amazon :)

  5. Despite my last comments (bit of a tangent) this seems to be pretty compelling evidence for versioning as a means of shortening the life of books and clearing the way for new sales. The software industry has been doing this sort of stuff for some time; some versions adding value, some simply force a new/resale.

    This reminds me of the parable of the broken window (bastiat)… If new versions are created simple to create another sale of essentially the same product - adding just enough change to artificially age it... then this is eroding overall economic growth. I'm not sure Moby Dick would be considered as valuable a contribution to American literature if we were on v56, Ahab's Grandsons Ghost Revenge Part III (ok, probably not the best example).

    The essential thing would be how to get a person to pay for the value they derive. Some people want to read a book once, and then they sell it. Others find it rewarding to keep the book and put it up on a shelf for display - as some sort of symbol of accomplishment. For the former – keeping re-sales from happening via “artificially aging” through superficial updates would not be something that should be incented. Rather a system that could extract the value from the read, or one time use, and then leave nothing for resale thereafter would be the best solution – it should drop the price of the individual read, and increase the revenue for the book/content seller. So, I’d suggest that versioning is less appealing than renting.

  6. One problem that I see with using versioning to sell textbooks is that the material in the books doesn't change much year to year and even across different educational environments. The purpose of textbooks is to provide the same basic knowledge to all students regardless of what school they attend. The only versions I'm aware of now is international editions. But even so, students often opt for whichever book is cheaper because, again, not much difference in material. And, unfortunately, digital books may be too easy to copy in student communities to provide much profit growth for sellers.

    Digitizing books is a unique challenge. My group will be discussing Digital Libraries in class soon so I won't go into it now but it's quite interesting how the market is changing.

  7. Amanda, It would be interesting to see your group presentation - unfortunately I cannot.

    I would argue that there is a substantial amount of artificial/superficial changes being made to seemingly static content, simply for the sake of "versioning" and triggering new sales. The incentive is there, and as a result, we should expect some authors to respond.

    I will avoid pointing out specific authors - but there are a number of very influential thinkers in management/technology that have done this sort of thing, I'm sure some of you could name a few (to be clear, it's definitely not Brynjolfsson). It's waste of talent, and consumes resources unnecessarily.

    I believe we should see the old paradigm of a text book eventually breakdown. We're seeing this sort of thing happening now with the emergence of companion sites - some of which are so good, that the text becomes secondary. What about a text personalized to the individual - placing greater focus on what you don't know? To the professors syllabus? or to the school? Simply digitizing the text misses these opportunities, and just carries the old paradigm over from one medium to the other.

  8. This was a great presentation.

    I remember when I started at MIT Sloan, I explored digital textbooks and textbook rental. I determined that my best outcome was to buy the books on my own and then sell them after done with them. The problem with this is the chance the books will change next year. With renting, this unknown is mitigated.

    Regarding digital books, resale is probably out of the question. However, it would be innovative to get cheaper prices for the books. There are a lot of people that skip getting the books or pirate the books from India. Cheaper prices would help them compete against "non-use," as Clayton Christensen would say.

  9. Tt is an interesting idea about 2nd-hand textbook digital marketplace. To do it, either C2C or B2C model can be employed, and I wonder whether the team have any preference towards these two channels? I personally hold the hypothesis that B2C channel is better, like Amazon and COOP, a brick-and-mortar bookstore that purchases 2nd-hand textbooks. The buyers bear the risk of not selling the 2nd-hand textbooks out before it get outdated.

    And besides, I am curious the reactions from professors who may also share book royalties from sales. Will they take any action to block this 2nd-hand market and promote more sales?