« GMU and Prizes | Main | IMF fact of the day »
Why are textbook prices so high?
...one of the major causes of higher priced new textbooks is the used textbook market. For example, if the fixed cost of producing a textbook is $500,000 and 5,000 units of the book are sold each year for 4 years then each textbook would bear $25 of the fixed cost.
However, if, due to the used textbook market, only the first 5,000 units are sold and, in each of the remaining three years these same 5,000 units are sold as used textbooks, then the publisher still has the $500,000 in fixed costs spread out over only 5,000 books. Thus each new textbook bears $100 of fixed costs, resulting in higher retail prices for all textbooks. This example demonstrates what has been happening in the textbook market over the past several years: As the used textbook market has expanded so have the market prices of new and used textbooks.
Here is more, but is that correct? To the extent this is a superstars market, where the leader becomes a focal choice and earns rents, the downward price pressure won't induce a proportionate supply reduction. (There would be, however, less ex ante competition to obtain this spot, which may involve supply reductions.) For less successful books, which inhabit a more competitive sector of the market where costs more likely bind, this analysis is more likely correct.
Paraphrasing Alex, I might note: "We know that textbook innovation saves lives and has a very high benefit to cost ratio. Thus, price controls or other restrictions that reduce prices are almost certainly a bad idea."
Posted by Tyler Cowen on February 3, 2007 at 07:51 AM in Books | Permalink
Comments
It's interesting, because I work the other way around: when a textbook dosn't cost too much, I'll buy it new and keep it.
If it's too expensive, I'll buy it used and sell it when I'm done with the course.
Posted by: Juliette at Feb 3, 2007 8:01:03 AM
How much faster than the rate of inflation have textbook prices gone up in the last decade? Twice as fast? Three times as fast?
I took a calculus course at GMU in 2005 or so and the book cost about $140. Is calculus such a fast changing field that a publisher needs to release a new edition every year to stay current? Whatever the reason for the price levels I can say that few things induced more disgust as a student than a trip to the GMU bookstore.
Posted by: MjrMjr at Feb 3, 2007 9:03:03 AM
I agree with Juliette. Why does used textbooks market exists? Because the textbook prices are so ridiculously high. Look at Greene's Econometric analysis, one of the most sold books in econometrics. 129 USD after 18% discount at Amazon? Is it the fixed cost? Oh come on!
I don't blame them much for collecting the rents on the books. OK, I complain just a bit. But saying the used textbook market is responsible for high textbook prices is cheap. The only factor responsible for high textbook prices is the publishers who have chosen to set the prices so high.
It's like the chicken-egg dilemma but with a clear answer now. What was here first? The textbook, or the used textbook?
Posted by: pinus at Feb 3, 2007 9:31:17 AM
Tyler,
The fixed cost of textbooks has next to nothing to do with their high prices. A high fixed cost may prevent a given textbook from ever being published, but once published, it is a sunk and/or an indirect cost, and only marginal replacement costs and demand elasticity determine the profit-maximizing price.
Textbook prices are high because they effectively have only one use and that one use is as a factor in obtaining a college degree. The price of a textbook is imputed from the perceived value of the college degree, and it has relatively little value as a stand-alone good. Also, of course, it has the third-party buyer effect, where professors determine how to spend the money of students, often from scholarships or parents.
Regards, Don
Posted by: Don Lloyd at Feb 3, 2007 9:37:21 AM
Isn't this just the old "things still suck for a durable-goods monopoly" problem? The current textbook publisher is competing with itself last year, itself the year before that, and so forth. But then why don't we see textbook publishers using all the standard durable-goods monopoly tricks, like guaranteed buyback pre-commitments to starve the used market?
The answer, I think, is that textbook publishers have already solved the used-books problem through techniques like revision-cycling. A small revision -- just enough to renumber the pages -- is enough to substantially inconvenience used purchasers. As this trick is widely used already (how many texts, even in seldom-changing fields, are on their sixth or seventh editions?), it seems unlikely that the used book market is a big unsolved problem. More likely it's just an excuse.
Posted by: Grant Gould at Feb 3, 2007 9:38:53 AM
This argument also includes one of the sorts of bits that make economists look dumb, namely the assumption that all the books sold go back in to the used market. This can't even be called a reasonable approximation since a significant percentage of the books obviously don't go back in to the used market (people keep them, throw them away, the wear out, etc.) It looks much more like one of the too common "I need this assumption to make my model work" types of things. It's obviously false and quite stupid.
Posted by: Matt at Feb 3, 2007 9:45:24 AM
Is there a feedback loop of some kind with the production quality and perceived quality of the books? Are they finely made to justify a high price, or priced high because they are finely made?
'been a long time since I bought books, but I'd have been happy with a low grade paperback for a lot of my non-core classes.
Posted by: odograph at Feb 3, 2007 9:48:18 AM
Sorry to double ... but I have seen what I believe were Taiwan issue versions of computer science books ... smaller format paperbacks with exactly the same text per page as the larger and more polished US premium hardcovers. If I'd had a choice I'd much rather had the more convenient and inexpensive Taiwanese versions.
Maybe an economist can explain to me (the economic rube) the publishers' choices in Taiwan vs. US.
Posted by: odograph at Feb 3, 2007 9:52:12 AM
Matt, I don't know your background, but if I may help, any trained economist would read the assumption that used books go to the market as really meaning that if some books go back to the market then the fixed costs in each new book sold will be higher. Tyler gave an example rather than a model, anyway. And imagine if he said "suppose a proportion alpha of used books are sold back to other students".. THAT would be silly in this context, wouldn't it? Best regards.
Posted by: Avinash Goldfish at Feb 3, 2007 10:00:09 AM
I think one major factor is agency issues.
Most textbooks are *chosen* by the professor and bought (paid) by the student. Often any other book except the professor recommendation is of inferior quality *for that specific class* (even if it would be adequate in general) as exam questions are based on having read it.
Posted by: luispedro at Feb 3, 2007 10:38:40 AM
I'm a non-economist and I'm surprised it took 11 posts before luispedro pointed out the issue of who selects the textbook for the course. The student has a choice between buying a new or used textbook or going without (maybe by relying on friends or the library). Choosing a different title is not an option. The textbook salesforce does not aim its efforts at the students.
The professoriate as a class benefits from high prices. A few benefit directly from royalties and many benefit from the freebies from publishers. There is some similarity to physicians and drug prescriptions, but patients often have a say in the decision and 3rd-party payers have a lot of market clout. Neither applies to the textbook "market."
I'm tring to teach myself spanish. I went to the bookstore and looked at the text used in our local intro Spanish class: Claro Que Si, available on Amazon for a mere $94.36! Instead I did some browsing on Amazon and purchased Spanish for Gringos for $18.68 (which comes with two CDs). Of course it's not the same as the textbook, but I don't think the difference in price is due to the fact that the non-textbook is taking a less classy approach to teaching the subject. The textbook has lots of illustrations and gimmicks which are a cost for the publisher, but to the professor selecting the text those gimmicks come free. In general I think language instruction is a good model of the issue because there is instruction going on outside college classrooms and operating in an environment where the purchasing decision is quite different.
Posted by: John G at Feb 3, 2007 11:45:27 AM
_________
" Why are textbook prices so high? "
__________
...is this a trick-question on an economics blog ?
If the price is 'high', then demand is high relative to supply.
Actually, Don Lloyd (above)(& John G) correctly state the issue as one of '3rd-Party control' of the normal buyer/seller relationship.
Professors/Instructors control the textbook 'demand by mandating student use of specific textbooks, but bear no direct economic cost for such arbitrary (3rd-party) 'purchase' decisions. Thus, they have low economic-incentive to seek lower-priced textbooks or pursue normal consumer tradeoffs in the quality/quantity/substitutes of 'textbook' goods.
College professors are also routinely in the 'supply' side of things by authoring textbooks themselves, supporting those written by friends/colleagues/favored-publishing-companies --- and mandating their use to students. If students do poorly in class, that deficiency is never attributed to the quality/availability of the textbooks (.. and rarely blamed on any deficiency in the professor).
Textbook publishing companies thus lack general competition to their products based on normal quality/price factors. Their incentive is to establish comfortable, long term niche markets via direct marketing to specific professors, related professional
groups/associations, and colleges --- rather than the ultimate consumers (students).
High (.. often outrageous) textbook prices therefore reflect the quite limited consumer choices available to typical students. A similar situation exists in the overall American college-education-purchase options available to student-consumers.
Posted by: Burgess at Feb 3, 2007 11:54:45 AM
The used-market argument is theoretically valid, but I don't believe it is empirically large. The agency problem dominates. Suppose that we (successfully and costless) banned a used-book market. Publishers would keep each edition longer, which would be generally a good thing, since more student would buy each edition, saving the industry inefficient fixed costs. But prices would not fall --- in fact, they might rise! --- becasue (1) MC would not fall, and (2) demand would become less elastic in the absence of substitutes, due to the fact that instuctors do not bear any of the costs.
Posted by: o at Feb 3, 2007 12:11:26 PM
I took a calculus course at GMU in 2005 or so and the book cost about $140. Is calculus such a fast changing field that a publisher needs to release a new edition every year to stay current?
I agree MjrMjr. Many subjects, like most branches of mathematics and introductory physics, have changed so little that you could learn the same thing from a 1970s textbook that you learn from today's versions. I have a physics textbook from the 1950s. It isn't nearly as pretty as modern textbooks but I would say it is still greater than 90% current with what is taught in an introductory physics class today. The only weak areas are updates in electronics (the basic formulas of electronics haven't changed, but we don't use vacuum tubes any longer) and no elementary coverage of relativity theory (not that the single chapter devoted to this subject in a modern textbook provides the student with much of a background anyway).
We could be getting by on the same mathematics and physics textbooks for half a decade without any problem. In fact, we should be able to use a textbook for that long in many other subjects as well. Has English, writing, or art appreciation changed so much that we really need a new textbook in each of these fields every year or two? Do foreign languages change that often?
So that's part of the question: what exactly motivates this rapid turnover in textbooks?
Posted by: mike at Feb 3, 2007 12:58:51 PM
But Avinash, is the assumption is "some books go back on the market" why is the assumption in the model above _all_ books go back? And if the model is for all books going back, would it not be a mistake to assume it applies when _some_ do? Economics is a very useful tool. But, it's often a pretty crude one, and massively counter-to-fact assumptions are reasons why. Given that the assumption here is obviously quite different from reality I'm pretty dubious that it will tell us very much about reality.
Posted by: Matt at Feb 3, 2007 1:23:23 PM
Some of the previous comments imply, but don't really say explicitly, that the publisher/professor relation sometimes becomes outright collusive, with all sorts of kickbacks (cash or goodies) flowing from the former to the latter. I don't think anyone knows how widespread it is (except the sales reps). But teh Chronicle of Higher Ed had an article a couple of years ago that presented examples of some pretty seamy deals. According to the article, professors in some disciplines appear to be more susceptible to these arrangements than those in others....
Posted by: eweininger at Feb 3, 2007 1:26:31 PM
I think luispedro and John G have excellent points here. It often happens that when students complain to teachers that the books are too expensive, we realise the teachers really don't care. The price is absolutely not taken into account when choosing the textbook. When they do realise the price, they'll tell you "It's only $80, you can afford that for my amazing class" without realising that we getat least five or six teachers telling us this at the same time. It might not look like much when you have a salary, but on a student budget, it weighs A LOT.
Hence the used textbooks market.
I currently study in Italy, and there is a market much worse going on: photocopies. It's an entire industry, despite the many laws against it. Teachers here rely more on textbooks than in any other country I've been to, and even relying on used books becomes too expensive.
It's annoying, especially when you arrive at the exam and realise you didn't really need the textbook in the first place, despite the teacher telling you the world would come to an end if you didn't read it.
Posted by: Juliette at Feb 3, 2007 1:33:00 PM
Addendum: would we not have an interesting natural quasi-experiment if capable teachers in various fields were--for reasons altruistic or otherwise--to produce "no frills" intro texbooks and make them available for free or for nominal cost? Say, something like this: http://www.introecon.com/
Posted by: eweininger at Feb 3, 2007 1:39:46 PM
I'm in the camp of those - Don Lloyd, luispedro, JohnG - who consider this an agency issue. The profesor makes the selection and the student pays. I don't think it's necessary to ascribe nefarious motives to the professors, just a lack of price sensitivity.
Notice, in this regard, that textbooks, unlike almost all other books, usually do not carry cover prices. Even a teacher who wanted to take price into account would need to do a bit of work to get the information. This certainly suggests that publishers are aware of the agency issue.
And of course there will be a fair amount of inertia built into the selection process. The teacher's lecture notes and assignments may be organized around a given text. Changing involves evaluating new books and possibly reorganizing the class. That also works against price sensitivity.
Posted by: Bernard Yomtov at Feb 3, 2007 1:42:27 PM
Matt, you shouldn't use the word "stupid." No books will go to the used book market if the publisher changes the edition number(and usually nothing else.) The college bookstores are complicit is this--not satisfied with repurchasing perfectly good used books at 50%, reshelving and selling at 75%.) Since that is a 50% markup and the bookstore prefers(against student complaint) to sell the new textbook, what does that tell you about the markup on the next textbook.
Posted by: lee at Feb 3, 2007 2:06:24 PM
eweininger,
would we not have an interesting natural quasi-experiment if capable teachers in various fields were--for reasons altruistic or otherwise--to produce "no frills" intro texbooks and make them available for free or for nominal cost?
Yes, I've been thinking along the same lines for some time now. That would be a good idea in fields are not frequently updated. In fact, a downloadable textbook distributed under GPL or similar terms would be far easier to update, expand, and customize as needed. Unfortunately, it would probably be easier to find professors willing to write the textbooks than professors willing to adopt them.
Posted by: mike at Feb 3, 2007 2:33:20 PM
am I wrong or in the uk reselling is forbiden?Some books carry a notice saying but for the USA this book can not be resale....Professional books , (law books at least) and law classics that do not create royalties obligations are even costlier.European books printed in The Hague or the UK are 3 or 4 times more expensives than any american book.
Books by Duncan Black or Gordon Tullock are over 100 $, a paper( in the way of book) by Buchannan can cost more the 30 $
Posted by: jcm at Feb 3, 2007 2:43:34 PM
Matt, I insist that well applied economic reasoning helps understanding more than it hinders it. The point of the original quote is, merely, that the textbooks market is special in that there is a permanent supply of lower-quality, lower-priced imperfect substitutes to new books, and that some people are willing to buy those substitutes makes the fixed costs borne by each new book sold higher. That is, in principle, correct. Your point, that not all used textbooks are resold, to which my perpetually-about-to-fall pile of textbooks is the liveliest proof, would affect only the intensity of this effect. The price of new textbooks should go up with the proportion of used books resold.
I myself agree that the agency issue should be the most important factor there. First, if the publishers
did not believe they would be able to coerce enough students to buy the new textbooks they would never pay
the high signing fees we are seeing today, as Don Lloyd said above.
Posted by: Avinash Goldfish at Feb 3, 2007 2:48:01 PM
I took a calculus course at GMU in 2005 or so and the book cost about $140.
A lot of folks complain about the cost of intro Calc (and Chem and Bio) textbooks.
Yes the purchase cost is high.
But at my first university, the Calculus textbook was used for four quarters of calculus. At the time, it was $120. Your cost per quarter for the book went down the more Calculus courses you took. If you completed the series, the book only cost $30 a quarter ... not bad for a textbook if you ask me.
Same deal with intro to Chem and intro to Bio textbooks. You use the same book for multiple terms, and the cost isn't so outrageous.
Posted by: WWU_econ_grad at Feb 3, 2007 2:59:38 PM
While the 3rd party or agency effect is probably the dominant effect, as a student you have to ask yourself what factor of increase in textbook price would cause you to not take a given course.
Then, ask yourself what factor of increase would cause you to change universities.
Finally, what factor of increase would cause you to not pursue a college degree?
The value of textbooks is ultimately limited by the degree to which textbooks, or substitutes, are necessary to realize the value of a college degree.
Regards, Don
Posted by: Don Lloyd at Feb 3, 2007 3:12:34 PM
