Economics Textbook
Mankiw, Principles of Economics, 6th Edition

Rick Anderson and I have been kicking around the possibility of working on a project to get college textbooks into academic libraries. Rick posted about that here, and my post is intended to complement his. I hasten to add that this post is not “the other half”; this is a big topic and the two posts together constitute only a small part of the many perspectives that ultimately have to be brought to bear on this subject. But you have to start somewhere. Rick notes that libraries may have a significant role to play in providing textbooks to (mostly) undergraduates. My view is that this could be a very good business opportunity for publishers, who are largely resistant to having classroom materials get into the hands of librarians. Among the perspectives not fully addressed here are those of students, instructors, administrators, and maybe even of parents, who stare in horror at the cost of putting a kid through college. Perhaps this is the second of what may become a string of posts.

The first thing we have to come to grips with is that there is no precise and inarguable definition of a college textbook. For that matter, even the lines that define “college” are blurry: Do we mean elite, private schools like Harvard and Stanford; four-year state institutions; community colleges; the world of the for-profits; and can we overlook the MOOCs? The “college textbook,” in other words, is a slippery concept, and it is important to know exactly what someone means when uttering it.

For my part, a college textbook refers to any required course material that consists primarily of words (text) for an accredited two- or four-year institution of higher education. That puts community colleges and the for-profits in the mix alongside the likes of The University of Chicago and Pomona College, and it also opens the door for books that not everyone would classify as a textbook. For example, most people would agree that a $200 hardcover book on organic chemistry qualifies as a textbook, but how about a paperback edition of the novels of Jane Austen? Both may be required reading for college-level courses.

Thus the first thing we need (yes, this would be an initial research project) is a taxonomy of classroom materials. We would then want to map that taxonomy against the actual books used in classrooms. That information (which books are used for courses) is hard to come by, but Joe Karaganis and Michael McClure are working on this at the Open Syllabus Project. Kitchen readers may recall that I gave a shout-out to this project before. Once we have that mapping in place, we will reach a decision point: Which segment of college texts should we begin with?

Let’s assume for the purpose of discussion that we decide to focus on the big, expensive texts used in freshman and sophomore courses. Purchased new, these books cost students $100 and more. (I just looked up the list price of Gregory Mankiw’s Principles of Economics: $300.) Now, a big chunk of that market has gone over to used books — about a third, by most estimates. Piracy takes a toll, too, on publishers’ sales, and then there is the growing and troubling case of the students who attempt to make do with no book at all. For publishers, the reasons to disrupt their own marketplace are growing. For example, a publisher may get an adoption (that is, the instructor puts a book on a list of required reading) for a class of 30 students, but after subtracting sales lost to used books, piracy, simple borrowing, rentals, and doing without, the publisher may sell only 10-15 books for that course. A publisher may say, Well, then, if I am losing all those sales despite my best marketing efforts, maybe I should raise the price to cover the lost business. The problem with that strategy is that higher prices simply encourage all the things — used books, piracy, etc. — that publishers want to prevent.

Why would publishers want to explore Rick’s proposal?

  • Publishers sell books at a discount, so that $300 economics text probably nets the publisher something like $180. This is a guess on my part; I’m assuming publishers sell print with a 40% margin. Selling to libraries instead of bookstores has just saved students $140 without costing the publisher anything.
  • Moves the business from print, with all its expense for printing and shipping, to digital. That should drop around 20% or more to the bottom line (of the $180 received for Mankiw). That brings the cost down to $144–without affecting the bottom line.
  • Remember those students who buy used books or no books? Let’s put them at one-half of the market. This means that the availability of books from the library effectively doubles the publisher’s user base. So that $144 figure is now $72 — with no loss on the bottom line.

Of course, some costs do not go away and there will be some new costs. An example of the former is the need to put a sales force in the field to get instructors to adopt the titles (just putting a book in the library doesn’t mean anyone will use it; in business terms, a library is a back end, not a front end). An example of the latter is the technical infrastructure to have a fully digital offering; and we can add to that additional marketing costs to make presentations to libraries and library consortia. Despite these quibbles, though, publishers could make as much money selling textbooks to libraries as they do now.

To put such a program together is a lot of work. We need:

  • A survey of all the kinds of books used in classrooms.
  • A list by course title (suitably tagged and cross-indexed) of all books currently in use.
  • A pricing methodology that is tied to classroom enrollment.
  • A number of publishers to participate (though for the big, expensive texts, about 85% of the market is controlled by just 5 companies).
  • Expertise on the library level. This is easy to overlook, and it will add to library overhead.
  • Better technology for textbooks than we currently have.

Then there is the thorny matter of where the money is going to come from. This is not trivial. If the funding for textbooks were to come out of existing library budgets (something that elected officials are sure to advocate), the library as we know it today would cease to exist.

Let’s take a pencil to the back of an envelope. We begin with some broad assumptions. Through superior workflow and all-digital solutions, we are able to get the cost of materials per student per course down to $25, a figure that is intended to represent all students and all courses — so we have just mixed Harvard with Middlesex Community College and public domain literature (Dickens, Thackeray) with the Mankiw title referenced above. There are 17 million college students in the U.S. If they each take 4 courses each semester, with two semesters per year, the total comes to $200 per student per year or $3.4 billion. Where are we going to come up with the dough? Yes, yes, I know: If only we had not gone into Iraq; think of the money we would have saved! But face it, guys: There will always be an Iraq. Governments don’t operate like an old-fashioned Chinese menu, where you can take one item from column A and one item from column B. We have just seen this with Brexit, with the claims that the money saved from payments to the EU will now flow into the National Health Service. This will not happen. We should not look for a political solution for a market problem: the academy will have to fund this initiative itself. But keep your eyes on the prize: that $3.4 billion estimate is approximately one-third the size of the current market, and this program will put all students, rich and poor, on an equal footing when it comes to classroom materials.

I would like very much for people to provide hard figures for those I grossly estimated above.

If publishers won’t be losing by doing this, and assuming that the money can be found (solution: charge a student fee of $400 per year and then waive the fee for half the students), who would lose out?

  • Printers and trucking companies, surely, and also the people who manage warehouse space.
  • Used-book distributors. Note that the students who resell used books also lose out.
  • College bookstores.
  • Amazon (15% of the textbook market).
  • Manufacturers of large backpacks.
  • Orthopedic surgeons (an aspect of the large backpacks).

I don’t mean to be flip about this; there is real human cost in this disruption. After all, the trucker who is out of a job may be trying to put a kid through college; but the benefits are apparent, too, as the trucker’s kid, like all kids, will be paying less for school.

We can argue about how and where to get started. In my view, it is prudent to avoid the temptation of going after the big textbook companies (Pearson, McGraw-Hill, John Wiley, Cengage, and Macmillan) first. Better to learn the ropes with smaller segments- – say, all course adoption titles from university presses (an $80 million market) or the public domain literature used in English courses: impeccably edited by a new service and sold for $1 per title per student, about 10%-20% of what you would pay for these in inexpensive paperbacks. Once these niche services are brought on board, we can pursue the bigger fish.

And the winners?

  • Libraries, which become closer and more valuable to their community.
  • Students, who save money.
  • Publishers, who maintain profits even as they steer their own companies through disruptive waters.

There may even be a little in this for the consultants who put it together.

Joseph Esposito

Joseph Esposito

Joe Esposito is a management consultant for the publishing and digital services industries. Joe focuses on organizational strategy and new business development. He is active in both the for-profit and not-for-profit areas.


33 Thoughts on "Textbooks in Academic Libraries: The Publisher’s Case"

There are several complicating factors that Joe’s presentation overlooks. There have been lawsuits pertinent to the pricing and sale of textbooks. E.g., the National Association of College Stores has sued publishers over price discrimination, charging that publishers sell textbooks to them at a 20% discount (which is the normal “short” discount for textbook sales) but at greater discounts to other buyers: In March 2013 the Supreme Court decided in favor of a student who resold cheaper foreign editions of textbooks into the US market (Kirtsaeng vs. Wiley). These suits have focused on the kind of textbooks Joe uses as his primary example here.

But, as any university press knows, any book can be adopted as a text for a course, and there is no easy way to predict in advance which books will be adopted. That is the reason why presses are hard pressed to make decisions about what books to exclude from the e-book aggregations. Any book included in such an aggregation is automatically doomed to lose potential adoption sales on any campus whose library subscribes to a service that includes the title. (I have given examples in the past of monographs focusing on one third-world nation as a case study that have sold tens of thousands of copies for course adoption. Who could have predicted that kind of success?)

While I agree that this idea is worth exploring, it runs into huge problems if it potentially encompasses every single book any academic publishers puts out.

The legal issues concern not buyers but resellers. Libraries don’t resell books; therefore there is no legal claim. So this part of Sandy’s comment is irrelevant. The point that it is difficult to know what books will be adopted is addressed by noting that pricing must be on an enrollment basis; and the point that attempting to encompass all books will be challenging is addressed toward the end of the post where the recommendation is that the project start small in a tightly focused manner. It pays to read these posts to the end.

And it pays to read what i said, Joe. I did NOT claim that there was a legal issue. I said that there have been lawsuits that create complicating factors for the marketplace, and the issues of discounting and resales certainly do affect the marketplace for textbooks. And you should know that the typical discount for textbooks is 20%, not 40%, a big difference for your calculations.

The 20% discount is peculiar to the U. press world and a small handful of other academic publishers. It is not true of the “big 5” publishers, who in any event use net pricing and allow retailers to add margin–typically 40% or more. I will not be commenting further on this sub-thread.

What you propose, a version thereof, was the basis for the Courseload business model, which I think is now defunct … Instead of trying to adapt the one student/one purchase model to the library, why not adapt the institutional model: one price unlimited users? For a book like Mankiw, the library might have to agree to a three-year, multi-user that is a multiple into the thousands, but the per student user cost could plummet. Then you figure out how to assess the costs, e.g. student technology fee …

I may be late to the party but I have not heard anything about how the authors who write most of the textbooks and get their 20% of every sale are going to be protected. The authors are often faculty that spend enormous amount of time writing, working with publishers, and actually involved in the marketing of their book are at risk here. Take away the royalty from the author (faculty) and there will be few new textbooks. Textbooks in history and the social sciences are not that expensive but authors depend on that revenue. Faculty members that I personally know make $30,000 or more a year from their textbook sales. I don’t think many faculty are going to write for the fun of it. So add authors to those who are impacted.

The big issue for publishers is that the traditional market for textbooks is flat lining at best and on severe downward decline at worst. Students are getting their books from so many different places these days (both legal and illegal) plus putting up much more resistance to buying them new that this trend will only increase. Moreover textbooks are just one source of information and there are so many other options open to them including a trend to OER by some academics. This means I think that publishers have to look at the whole totally afresh working more at an institutional level rather than with individual academics on a piecemeal basis. Libraries want to meet their students needs more effectively in this area and publishers need to look to the future so it looks there are deals to be done on a new model!
Dominic Broadhurst
University of Manchester

The college book business is actually doing much better than this comment suggests.

On Fri, Jul 8, 2016 at 10:30 AM, The Scholarly Kitchen wrote:


this might be a geographical difference, but if you speak to publishers in private they have a different view to public utterances plus you only have a) look at companies in difficulties such as Pearson and b) ask yourself why so many are trying to diversify into what they lovingly call themselves as Learning Solutions companies. Yes some books still sell well, but as an overall the market is changing.

Joe once again you present another side to a discussion. I would like to add another.

One of the problems with the library e distributing books is that textbooks are rather complex in layout. Computers are rather inept in presenting technically laid-out materials. In the Mankiw example there are tables, graphs, formulas, boxed examples and other apparatus which are embedded in the text. The problems regarding using the text as a study platform is just not too conducive to computer presentation. In other disciplines there is the problem of maps, intricate biological, chemical and physical presentations both photographically and through line art, etc.

So, we eliminate those disciplines which use complex presentation of ideas. Thus, what we have is basically black and white books. These books probably comprise about 35% of the market and are or at least were the least expensive.

Recently, I had a discussion with a VP in a major STM publisher who also publishes text books. The company wants to go entirely e and get rid of paper. The problem is with the presentation of materials and cost of maintaining computers – new equipment, programmers, tech staff, etc. He commented that it was just as expensive to print and warehouse as it is to not.

I am not too sure that the library providing the platform for textbooks will draw students back into the library. After all, they can download or access from their dorm, apt or home without ever entering a library.

I have seen and used a P/H program which allows semester long access to the adopted text via their platform. The cost was covered in tuition.

I am still not too sure why the library wants to become a retailer and a production shop for books. Both areas where they have little or no expertise.

Harvey, just to respond to the library-related parts of your comment:

I am not too sure that the library providing the platform for textbooks will draw students back into the library. After all, they can download or access from their dorm, apt or home without ever entering a library.

I’m not sure where you’ve gotten the idea that the goal of this proposal is to bring students physically into the library. That’s completely irrelevant to what Joe and I are talking about. The point is to get textbooks to students, not to get students physically into the building.

I am still not too sure why the library wants to become a retailer and a production shop for books. Both areas where they have little or no expertise.

And I’m not too sure where you get the idea that the library wants to become a retailer and a production shop for books. What we’re talking about is the library being a broker of access to books — an area in which the library has deep and centuries-old expertise.

Textbook publishing provides important revenue that supports activities at the American Society for Microbiology, nonetheless ASM Press has experimented with its textbook distribution. Our ebooks collections allow unlimited consumption of the content and contain titles that could be used in whole or part for instruction. The authored textbooks, written specifically to teach virology, molecular genetics, bacterial pathogenesis, food microbiology, etc., are not included in these collections, as they are adopted by instructors. Many universities have asked to have the textbooks included in the collections, but as Joe points out, the loss of revenue would be substantial. Instead, we have sold etextbook files requested by faculty to universities to host locally for unlimited use by all authenticated patrons. Pricing is by number of students expected to take the course per year multiplied by the average number of years between editions, then a large discount (40-50%) is applied. This means the university commits to paying for their students’ access to the textbook, which signals a change. In addition, we are copublishing an OpenStax College OA textbook, microbiology for allied health students, later this year.

This is an issue that will not resolve quickly, in part because it is tied to the changing nature of education in general and the HEI’s in particular. An example comes from the world of the “for profits” and the rise of MOOC’s where there are teams of professionals responsible for the creation of materials for students to access as they pursue competence which, in the long term is not mastery of the materials as measure but the ability to demonstrate the absorption and applications.

This holds particularly well where the knowledge base is not as dynamic and evolutionary but rather static. It fits well with the rise of non-tenure track faculty at the traditionally defined undergraduate level where much of the didactic materials are scripted by others as noted above. It is also the market that was referred to by Dan when he indicated that faculty reaped large rewards, particularly if they could write a widely adopted freshman/soph text.

Most of the discussions above, joe’s and rick’s are assuming the traditional academic, and thus, text book market- one that hold and must be the frame for addressment.

Next, until one gets to graduate seminars shaped by narrow disciplined faculty, as we have seen from examples from MOOC’s, internationally, mastery of concepts for competencies may not be dependent on “text books” but access to the very literature which is the subject of STEM/STM publishers here in the Kitchen. Much of that can, and probably will depend on the work published by scholars but not necessarily needing to be formatted into a “text”.

For example, we need to seriously consider the growing number of free and readily accessible web summarizers, simple versions, currently, of Watson and off spring. These are able to drill through literature from title to text to footnotes and even to interpret graphic materials. Thus students seeking competency qualifications are able to use the libraries to which they have access to create their own tailored text books. Some AI, like that of Narrative Sciences are even able to write articles using large databases. They are not yet able to do this with narrative documents, YET.

In other words, the ideas of Rick and Joe are based, not on where knowledge acquisition is going but where it has been and is today, but gradually changing. Both Rick and Joe admit the issue is complex and will take a long time to solve. I agree, but if it is done as suggested here, when they get to the new “here”, it will have moved on.

The change is “here”, not well developed but sufficient to access this manually with minimal ICT expertise. At present text books are static. But with AI and programs like Amazon’s recent announcement of “Inspire” providing free and open access to P-12, and evenutally HEI faculty, the future is here just not fully visible.

And there is a potential legal challenge here that could be devastating to nontextbook, regular academic publishers whose monographs are used for teaching. The major library organizations including ALA and ARL have bought into the idea that “transformative use” means that monographs and other such works (including novels) originally written for other audiences can be re-purposed and copied for use in courses for teaching undergraduates without any permission from the publishers or authors under the notion that this kind of use is “transformative.” This is a main argument of the Guide to Best Practices in Fair Use for Academic Liibraries and reflects the views of a chief legal advisor to libraries, Jonathan Band.

Thought this was addressed when universities were providing xeroxed books of readings for course usage.

Not at all. The “transformative use” argument has only come into its own more recently. Traditional coursepack copying–or ereserve copying–is not considered “transformative,” at least by the judge in the Georgia State case.

Missing in this discussion is anything about what students prefer, and what format is best for their learning experience. My (admittedly random) survey of undergraduates continues to show a preference for print among a solid majority. What I am told most frequently is: “I only buy ebooks when it saves money.”

I’ve always been kind of bemused by these surveys, because (from what I’ve seen, anyway) they generally ask questions along the lines of “What would you rather have, a print textbook or an electronic textbook?” The answer to this question is usually “Print.” But a more relevant question would be, for example: “Which would you prefer: a print textbook that costs $200, or a flat-file .pdf of that same textbook for $50?” I promise you, the answers to that question would be very different. This reality is hinted at by the comment you quoted, Lynne — “I only buy ebooks when it saves money.” Well, right. That’s kind of the point.

I’m sorry to disagree, Rick, but you seem to be ignoring the point that, given equal pricing, the majority of students prefer print. (As an aside, it is a mistake to assume that there is always—or even usually—a big disparity between the price of the print and the e editions. For many of us publishing material used in significant numbers in university teaching, the two editions have the same price; and that price, by the way, is nowhere near the $200, or even the $50, that you mention.)

Sorry, Lynne, I seem to have clumsily communicated the exact opposite of my point. My point was that if pricing is equal, then yes, the surveys suggest that students will generally prefer print. However (to my mind anyway) a major purpose of e-textbooks is to bring the price down. If the price is the same, then I don’t see much reason even to produce e-textbooks.

I’m not assuming that there’s always a big price disparity between formats — I’m pointing out that a significant disparity in price would almost certainly shift students’ expressed preference for print. (The dollar figures I used in my comment above weren’t mean to be realistic, exactly, but merely illustrative of that point.)

But in fairness it should be pointed out that e-textbooks can include dynamic elements, like videos of scientific experiments and such, that print textbooks simply cannot supply.

Replying to several comments here.

My sources: two former CEOs of Big 5 college publishers.

The Amazon numbers (the variance) is probably explained by (a) the difference between dollars and units and (b) whether or not used books are included. Amazon is the clear leader in used books, which boosts its market share. The 15% figure I quoted was for new books (publishers’ receipts).

As for discount, besides the comments from the former CEOs, all one needs to do is to look at pricing at Amazon, where the gap between the list price and the price to the consumer is far more than 20%. Amazon may occasionally sell books at a loss, but not $50 per unit.

Concerning how things look to a library tech books specialist, that’s irrelevant. Those jobs are going away–sorry! The point is not how things are done but how to accommodate changing market conditions. Illustration: a student can enroll in one of two comparable institutions, but one costs $1,000 or more because its library has not signed up for Rick’s program. Over time the heartlessness of the marketplace always wins.

Finally, while it is true that all things being equal, most students would (today) opt for print, the point is that not all things are equal. Lower pricing is an aspect of digital books. It may be that the bad drives out the good, but, as I said above, the market is heartless.

Joe Esposito

On Fri, Jul 8, 2016 at 7:02 PM, The Scholarly Kitchen wrote:

> Sandy Thatcher commented: “But in fairness it should be pointed out that > e-textbooks can include dynamic elements, like videos of scientific > experiments and such, that print textbooks simply cannot supply.” >

Concerning how things look to a library tech books specialist, that’s irrelevant. Those jobs are going away–sorry!

I’m puzzled by your dismissiveness. Why is the library perspective irrelevant to a conversation about the benefits of having libraries buy e-textbooks? We’re the proposed buyers; “market conditions” don’t exist without buyer participation.

Workflow follows strategy. Libraries, to the extent that they focus on collections, are continuing to participate in the affordances of print, where members of an institution physically went to the library. This model was moved (with enormous success) to digital publishing. But that is breaking down. With the focus on OA, libraries now seek to serve not their institutions but the global community; this is reflected by recent ARL comments about linking libraries together (all good ideas, in my opinion). Libraries have at the same time abdicated any role in user data. Into the breach move publishers, which are reconfiguring workflow to bypass libraries and collect end-user data. So today’s successful workflow is tomorrow’s rust-belt economy. This will manifest itself more quickly in books than journals because the current experience of books is unsatisfactory (hard for discovery, awkward display technology, etc.). The question is not how libraries work but how they will be made to work in the future.

At Cornell University, we are not unfamiliar with the movement towards open access (see ArXiv, ), linked data (see LD4L, ), or library partnerships (see 2CUL, ). These aren’t things we do instead of licensing e-book access for our own patrons; they are in addition to.

It’s an exaggeration to say that digital publishing is “breaking down”. Rather, some sales models are breaking down. Take, for example, short-term loans, which initially enjoyed some popularity as a money-saving patron-driven acquisitions method. But the model at its original pricing was unsustainable for publishers, so they increased prices; this was unsustainable for library budgets, so they stopped buying. It’s not “breaking down” for no reason – it’s breaking down because supply and demand couldn’t come to a mutually acceptable compromise, and the money stopped flowing.

Perpetual access sales models, on the other hand, are healthy; publishers iike Springer, Wiley, Elsevier, and ProQuest are making a lot of money from us on these. Libraries are very interested in purchasing unlimited-user, perpetual access e-books for class use. See the Charlotte Initiative for Permanent Acquisitions of E-books; some info at , and if you registered for ER&L 2016 you can watch the recording of their talk at .

We aren’t afraid to try new sales models, but they have to be ones that solve problems for us, instead of creating them. Evidence-based acquisitions is an example of this; we pay a lump sum up front for a year’s access to a whole pile of e-books; at the end of the year, the platform gives us the usage statistics, and we can use these to determine which of the e-books we would like to retain perpetual access to. The up-front payment pays for these, up to either a predetermined number or the amount of the prepayment. The problem this solves for us is that when we firm order e-books or buy them on frontlists, there’s uncertainty over whether any individual title will ever be used by our patrons. We’d like to know that we’ve spent our money on what they needed, instead of overlooking what they needed and buying something irrelevant to them. So EBA helps with that. I’m given to understand that it also helps publishers by providing them with a known quantity of income up front? I know less about that end, though. It remains to be seen whether this one will be sustainable for everybody or not.

Regarding the collection of end-user data – whether this is a feature or a bug depends on your perspective. We deliberately refrain from collecting user-specific data to protect patron privacy. Our nuclear physicists should be able to do their research without fearing that the FBI can raid our records and arrest them on the assumption that they’re building a nuclear weapon; our Arabic studies researchers should be able to work without fear that the NSA can get the paper trail of what they’ve been reading and accuse them of potential terrorism. We don’t consider patron privacy and academic freedom bugs. We kindly request that the host platforms who provide access to library patrons do the same.

This isn’t to say that all data collection is bad; we frequently collect aggregated usage statistics from e-resource platforms to determine which databases/journals/e-book subscriptions are in demand by our patrons and which aren’t, or aggregated data on user interaction with our public catalog, to determine how to make improvements. There’s a lot of data that can be useful without betraying the user’s identifying information.

I don’t disagree that there are problems that need solving in the e-book landscape. Many user interfaces are problematic; chapter-level downloads/printing are too restrictive on aggregator platforms; DRM poses an extreme frustration problem for legitimate patrons just trying to read the content we’ve purchased; discovery is always something we’re working to improve; smaller publishers may have difficulty backing up their content to Portico/LOCKSS, etc.

But none of these are problems with the acquisitions workflows, nor a result of the pricing structures agreed upon between seller and buyer. My original point stands: please pick a stable price that does not vary by class size.

As an example, here’s a link to Wiley’s pricing structure for their pricier e-book content, the “Major Reference Works”: . We also use their less expensive, regular “o-books” quite often; they sell those through GOBI/OASIS as well as their own admin interface. This list I’ve linked is for content that’s priced for $1000+. You can see that they’ve listed very specific prices, made the spreadsheet freely available for buyers to download and review without debate, and chosen three different pricing levels based on university size, using FTE. I’ll note that I’ve heard differing opinions from selectors on the pros/cons of the “updating MRW” model, so I’m not necessarily endorsing that bit, but I am endorsing the ease with which I can determine how much a reference work is going to cost us.

Thanks for this very informative post, Heather. I’m curious about how “use” os measured for e-book collections? is it just the recording of a mere access “hit” that is measured, or is there some way of finding out what sort of use was made of the material accessed in more detail? It strikes me that mere hits alone don’t tell us much about the value of a book to users.

Heather, I think what Cornell is doing is terrific; whoever said otherwise? The question is whether or not it is relevant. Once again, strategy leads, workflow follows. It’s simply backward to entertain a new idea by saying how it would fit into a preexistent workflow. The question is, What new workflows will have to be designed in order to create a situation where all undergrads have less expensive access to classroom materials. As Rick noted, OER has not lived up to its promise. The idea being floated here is another way to approach the situation. It is not dependent on libraries’ participation, but it would be in libraries’ interest were they to pursue it–which is the core of Rick’s argument.

A perspective from Library Tech Services, as an e-books acquisition specialist:

A pricing methodology that is tied to classroom enrollment.

This particular proposal isn’t going to work for libraries.

(a) There are a lot of people in the acquisitions chain in between the person who knows the enrollment numbers and the person who orders the e-book; this proposal introduces a new and large burden of investigation on the part of the library. We can provide university-wide FTE numbers pretty easily, but nothing so granular as class-specific enrollment.

(b) Books can be used for classes multiple years, with varying levels of classroom enrollment. We prefer to buy perpetual access to e-books that might be used for classes; trying to price them based on the first year’s enrollment is a more arbitrary measure of the book’s value than is appropriate. And if you want to sell these books on a semester-by-semester subscription basis, librarians aren’t going to appreciate the unpredictability of the hit to their yearly budgets, and faculty aren’t going to be happy if the e-book they were quietly using for classes suddenly disappears because we didn’t have perpetual access to it and the librarian chose to stop subscribing.

(c) Speed of acquisition can be critical when we’re ordering e-books for class use. Faculty may assign a reading from an e-book we held through a subscription, and only discover at the last minute that the publisher has withdrawn it from subscription plans, and we need to scramble to buy replacement access. E-book platforms that use a standardized price and allow us to order through a web admin interface that will grant us instantaneous access to the e-book are enormously helpful to us.

(d) Even in less time-sensitive cases, we still care a lot about the time our staff spend investigating and acquiring e-books. Variable pricing is going to involve a lot of back-and-forth conversations with faculty, with funding selectors, with the vendor – that is not appealing. We need it to be easy to buy from you on a regular basis. We do most of our title-by-title purchasing through GOBI/OASIS, where we can see a specific price, make decisions quickly, and have a consistent activation/invoicing workflow.

I can’t emphasize (c) and (d) enough – being able to acquire new e-books and fix broken e-books quickly, easily, and efficiently is INCREDIBLY important to maintaining faculty trust in using e-books at all. We can’t do that if we have to wrangle over pricing.

The above items are the things that are instant disincentives to course-enrollment-based pricing, but there are some other issues I’d like to put out there:

(e) A lot of faculty are savvy about the cost of textbooks to their students, and choose to assign e-books that they can already find in the library catalog. So you’d be losing sales from that behavior. You’d gain some of them back from a library’s “just in case” purchases – our selectors buy a lot of e-books because they judge that the content is important, and may be valuable to our patrons either now or in the future, and then they’d be in the pool of books these faculty choose from. But it’s really unlikely that we’ll buy unpredictably-priced e-books “just in case”, especially if the only way we can buy them is contingent upon having course enrollment numbers.

So, please don’t limit the sales of the e-books to course-use-specific models. If you’re selling an important book, we may want to buy it for our collection even if we don’t know of anybody who’s using it in classes.

(f) Stable pricing structures that can accommodate the usage expected from various sizes of classes already exist – 1-user access, 3-user access, unlimited-user access, and the “non-linear”/”concurrent access” model that allows unlimited simultaneous use, up to X uses/year. By default, my team tries to buy the best multi-user access we can for a title we know will be used for a class, regardless of class size. Sometimes a selector knows the class is small and their budget is tight, and they’ll request a 1-user. Sometimes a selector will know that the class is enormous, and if we can’t buy a true unlimited-user license, we’ll buy multiple 1-users or 3-users or non-linears or some combination of those, to try to approximate it.

So, please do make an unlimited-user version available; we accept that it’s going to cost more, and really appreciate being able to avoid patron frustration when we can. If you want to sell unlimited-user access for a pretty high price, please also make 1-/3-/non-linear options available, so we can buy what we think we need and upgrade if we discover it wasn’t enough.

Just a couple of quibbles about details. First, I used to have a bookstore and sold textbooks to college students for about seven years. I’d have to agree with Sandy that the majority of textbook publishers use a short discount on their textbooks. I’m not talking just UPs here or the Big 5, I talking about Prentice Hall, McGraw, Norton, Blackwell, Wiley, etc. They really don’t offer trade discounts on those books intended for the classroom, those glossy printed case intro. books.. Perhaps more units of trade books are sold, thanks to classics like To Kill a Mockingbird, but my experience consistently showed that the majority of the revenue came from the sales of new books that carried short discounts.

Second, you state that Amazon is currently controlling 15% of the textbook market. Cite, please? The graph on the bottom of this 2013 article shows Amazon controlling 32% of the textbook market back then. I understand it’s continued to grow. Why do you think it’s half of what Bowker thought it was in 2013?

For what it’s worth, used books are incredibly lucrative for textbook stores. The margin is amazing. And, a significant proportion of the books were sold back in perfect condition, so purchasing a “used” book back from a student for $5 provides the bookseller with the option of storing it until the next semester it’s going to be used, or to return it to the publisher for a credit worth the full price minus the short discount. You bet your bippy that happens. And if the publisher refuses the return, you can always return it to a wholesaler. They’ll take back damn near anything.

I once heard from a textbook sales rep that the secret to the Riggios’ early success was figuring out they could buy back mint condition “used” textbooks for pennies and then return them to the publisher for full credit. He’d said that back in the 70s, his company consistently had Riggios’ stores returning far more books than they purchased, year after year. He even went on to speculate that it was that flipping used textbooks that helped finance the Riggios purchase of Barnes and Noble. I have no idea how true any of that was, but I found that a fascinating anecdote.

A few things I have seen that were not mentioned yet. College students rent most of the books they don’t purchase new or used. They may save $3-$100. Probably averaging $30-60 depending on the book. In addition to renting from the college store they buy new or used from Amazon, also renting from Amazon, and buying used from Chegg too. Someone in the business once remarked that they could not understand how low Cegg could go on price. A race to the bottom is not good for any industry. Another subject not mentioned is Amazon ‘ s new reader that allows holding a half dozen or so page images that basically substitute for the need to flip back and forth between pages. A major reason students don’t like ebooks is the inability to flip around. I was surprised this received so little attention. It’s just in time for families to consider for kids going off to school. If it was meant primarily for Holiday sales they could have waited until September. If I think of something else, I’ll add it another day. Personally I think when I was in school several years ago, the proportion of cost of books to tuition and living expenses wasn’t so different. I understand the cost is not affordable for many students, but I’m not sure why affluent students seem so anti-book. It’s also interesting w rentals. Students just don’t want the books once they are done, even novels. Occasionally they will transfer a rental to a purchase if it has important content for their major. Books are a major out of pocket expense, but the cost of an education is the real problem.

Just today I noticed one of the college bookstore chains based at a selective college, was looking for an employee who would take rental info sign ups while students are in line rather than filling in all of the time consuming info for first time renters when they arrive at the register. Stores make a lot on rentals. New and used books can be rented. New rent at a hi gbh er price, that one time. They get a good price for rentals and can rerent for at least a few semesters in most cases.


I’m not seeing a “Reply to this comment” beneath your post – maybe we’ve threaded down far enough in the chain that further replies are difficult to display? So I’ll start a new one.

You’re welcome! It actually turns out that usage reports for e-books are terribly inconsistent, and what each number represents depends on how that particular platform serves up the content. Most e-book platforms at least provide COUNTER reports, which mandate that the numbers come in units of “section requests”, which could mean anything. A platform like Wiley/Cambridge/ScienceDirect, which serves up information by allowing users to download chapter PDFs, would measure in units of “chapter downloads”. If the platform also allows full-book downloads, they may or may not choose to add those to the COUNTER count. Worse, aggregator platforms like ebrary who have multiple methods of accessing/downloading the content actually throw all the different counts together and add them up in COUNTER. I consider aggregators’ COUNTER counts to approximate page counts, since the page units tend to dominate.

However, ebrary is also one of the platforms that recognizes the units problem in COUNTER, and they also provide a platform-specific “Title Report” that gives much more detailed numbers – how many page turns, how many hits, how many pages downloaded, how many full-text downloads, how many turnaways, etc. – that are much more useful. These are the ones I download when I actually want to see usage trends.

Usually an e-book platform will provide an admin interface you can log into, to download various kinds of usage reports. Sometimes you can customize them to show extra data fields that you might be interested in.

It’s kind of ironic that the COUNTER format was meant to provide a consistent format by which you could compare usage on one platform to usage on another, but actually ended up producing a result (for e-books) that makes it impossible to do so.

Comments are closed.