Arnold Kling  

Access to Academic Research Papers

PRINT
Real Real GDP... Business Experience...

Megan McArdle writes,


Kevin Drum takes a courageous and rare stand on the internet, arguing that yes, downloading millions of files from JSTOR while evading attempts by both JSTOR, and the owner of the network you're using, to stop you, is, well, pretty close to stealing

[stepping onto soapbox]
Forget the analogy with stealing. The online pricing model at JSTOR and the major academic journals is totally inappropriate. The basic business model is to charge academic institutions a ton of money to give students and faculty "free" access to content. In order to protect that business model, those of us who are not academics get charged ridiculous amounts--I've seen $35 to download one journal article, for crying out loud.

In my opinion, there should be two sources of revenue for repositories of online research (by a repository, I mean ideally a database of just about every journal article and working paper on every subject; you do not have to achieve that level of coverage, but I want to make it clear that I am talking about more than just a single publisher's journals) :

1. Annual user fees from frequent users of, say, $20 per year. You are a frequent user if you download 10 or more papers a year. Up to that point, your downloads are free.

2. Donations. If the donors who give to higher education would divert just a small fraction of their funds to support an online repository, that would be plenty.

I am not wedded to this solution. What I want for online research pricing is "Don't be evil." JSTOR, the NBER, and the journals are evil. The way they treat intellectual property gives the slogan "property is theft" some credibility.
[stepping down from soapbox]

Joshua Gans has another proposal.


Comments and Sharing





COMMENTS (13 to date)
PrometheeFeu writes:

That's because intellectual property IS theft.

The journals have found the most amazing business model ever. Stop me if I'm wrong. But the people who write the articles pay to get in the journal. The peer-review is done for free. And then people pay to have access to the journal. So they provide bandwidth and hosting. I'm pretty sure that Google Scholar would be willing to provide bandwidth and hosting for free. As for the peer review process, social network technologies could be leveraged to make that much more open, transparent and decentralized.

Bruce Bartlett writes:

I think you are a bit too hard on the NBER. It only charges $5 per paper and makes all of those more than two or three years old available for free. Moreover, it makes all of its out of print books available for free.

As to JSTOR, I agree that their pricing model is stupid. But the real stupidity lies with the universities and professional groups that sponsor academic journals. They are the ones that create the content that publishers exploit. I believe it is fundamentally contrary to their mission to make these journals available much above cost. And all online articles more than two or three years old should be freely available. The National Tax Association does this.

I would add that think tanks are also guilty of stupidity in making their work available. Cato is pretty good about this, but others do a poor job. I don't know why Brookings and AEI haven't posted every book they have ever published online.

AC writes:

Universities get direct and indirect subsidies, don't they? They should at least have to contribute to open access to knowledge. Why are the top schools in particular helping to maintain the journals' monopoly position?

Chris Koresko writes:

I think we must be missing something here. Article authors are willing to pay page charges to the journals, referees are willing to give their valuable time for free, and researchers willingly pay substantial amounts for access to the product. The question in my mind is, why?

Do the journals subsist on rent-seeking somehow? Is there some mechanism which denies credibility or prestige to an article whose copyright has not been transferred to a journal? Is there some reason an expert would be unwilling to provide free reviews of articles which are sent to him by the author, or by an interested reader?

Suppose there were an outfit like Wikipedia dedicated to academic research. Anyone could write and publish articles there, anyone could write reviews and submit them, anonymously or not, for the author's edification. And anyone could download any article, either in its 'raw' form or (using a filter) once it's passed peer review. Articles could be restricted by content (size, number of figures, etc).

One missing element of that model is the mechanism for selecting reviewers. Many reviewers value their anonymity, and there are social reasons for that (and for denying rivals the ability to review each others' work).

Another might be content filtering. A paper journal can only be so big before it's unwieldy. And if it's going to stay relevant, it needs to keep quality up. That gives it a strong incentive to be picky about what it publishes. And since readers generally don't have unlimited time to keep up with current research, having quality work picked out for them is valuable.

Cheap, fast electronic communication has largely obsoleted one of the main functions of traditional journals, namely the formatting, printing, and mailing of text. That doesn't prove there are no other important functions that have yet to find adequate substitutes.

Peter Gerdes writes:

Why do we even need that level of funding.

Once we move the whole business of academic papers online there really isn't any point for the whole formal journal model at all. Academics spend huge amounts of time reading, evaluating (for themselves) and digesting academic articles. Currently virtually all of this effort is wasted because only the 2-3 official reviewers' judgements have effects easily visible by the community at large. Given nice reputation based rating and classification software with the ability to comment and discuss we could reduce both the time spent reading irrelevant articles, improve the accuracy of peer review by harnessing more eyes and increase the overall efficiency of academic work by using comments to avoid the duplicated work by everyone trying to understand a hard paper.

Moreover, we could separate the function of evaluating work for trustworthiness/reliability/non-crackpotness from the function of ranking articles by their significance and classifying them by subject. Currently journals collapse these all together as if everyone was interested in the same thing and had the same notion of importance. An online ranking system could easily reflect the full multi-dimensional nature of both importance and subject matter so rather than reading the same articles in nature as everyone else you could have a feed listing those articles tagged with your sub-specialty or mentioning techniques you are interested in even if few others care.

Academics would contribute to such a system to improve their online reputations and discuss/comment on the papers with their colleagues (obviously they would need to be able to screen those without sufficient credibility in the community). There would be a single hyperlinked repository of knowledge in each field, auxillary explanations of unclear concepts, and minimal duplication of effort.

Already academics do peer review for free and they would have to do much less of it in such an online system and the rating/recommendation software would handle merging the various ratings and provide feedback to the submitter for non-final paper submissions.

So the total cost to run such a system once created shouldn't be any greater than that of slashdot or other online content provider. It could be easily ad supported while providing free (or almost free) submission for those with sufficient reputation (crackpots and outsiders perhaps should be required to post a bond to discourage spamming).

Andy Hallman writes:

With the advent of the Internet, what function do journals serve at this point? Surely it's not facilitating communication between academics. Is it purely signalling?

wm13 writes:

Where did you go to college and grad school? Yale has free JSTOR access for alumni; maybe your almae matres do also. If not, they should.

James Reade writes:

Well, stop moaning and let the market do its work - as you constantly berate others than comment and moan about market outcomes to do.

Or - wait - are you suggesting the market outcome might not be the best solution here?

John Palmer writes:

I can imagine that in the future, it will be feasible for research journals to pair with Amazon or Google-Books, selling access to articles for low prices but with sufficient protection that it is difficult/costly to bootleg them.

rpl writes:

James Reade:


Well, stop moaning and let the market do its work - as you constantly berate others than comment and moan about market outcomes to do.
Or - wait - are you suggesting the market outcome might not be the best solution here?

The "market" for academic journals is built almost entirely on spending other people's money. Authors pay page charges from grants. Libraries spend money from state appropriations and endowments, and so on. The principal users of the service, readers of the journals at subscribing institutions, have virtually no skin in the game at all. If journal readers and authors were paying for this stuff themselves, then there's no way the current price structure would continue. That markets function badly when you have widespread third-party payments is not exactly a surprising revelation.

Andy Hallman:


With the advent of the Internet, what function do journals serve at this point? Surely it's not facilitating communication between academics. Is it purely signalling?

It's how academics keep score. It's possible that some of that function could be replicated in open journals using community ratings, but nobody's sure how much effort it would take to get that to work. Look at how hard Google has to work to keep people from gaming Page rank with link spam, and look at the lengths sites with community rating schemes have to go to to prevent cliques from having out-sized influence. The academic community just isn't ready to take that plunge yet.

ohwilleke writes:

Often I agree with the reasoning, but not conclusion; here, the reverse applies. The argument for the result is basically ipse dixit, it's right because I say so.

There are good reasons to treat academic journals differently.

1. The authors are on salary and are rewarded for their publishing by their institutions, not by the journals themselves.

2. Paper journals are insanely expensive relative to electronic access formats, because the printing cost of documents with very small markets defeat the economies of scale that make ordinary publishing possible. The new technology shouldn't be subsidizing costs distinctly arising from the old technology.

3. Apart from printing and shipping, academic journals are incredibly cheap to operate because they are stylistically dull, delegate all the time consuming writing and peer review work to people who work without charging the journal anything, and don't have to have a sales staff beating the doors down selling advertising or lots of new subscriptions. Most academic journals can be managed by less than 1 FTE of paid employees, apart from the printing part. One small team of specialist editors and IT people can manage the publication of lots of academic journals.

Academic journals in law go one step better and have all the work (except printing and shipping) done entirely by student slave labor that is paid in future employment prospects fronted by other people, instead of money fronted by the academic instiution.

4. Most university libraries make a subscription to access to their print collection available to members of the local community who are not academics for something on the order of $50-$300 per year, which provides access to all of the print journals in their collection. Thus, the going price for access to an entire year of print journals is on the order of $25 a month or less. A lower cost way of providing the same thing should cost less.

5. The real constituents of academic journals are the authors, not the people who own the publishing houses (which are often non-profits anyway). What the authors want more than anything else is maximal citation - citations are exchangable for cash at your local tenure and full professorship promotion committee meeting. Charging big bucks for online access to journal articles is contrary to the enlightened self-interest of the de facto owners and constituents of these organizations. A royalty driven pricing policy only makes sense in a world where somebody cares about royalties.

ohwilleke writes:

Continuing a list of reasons:

6. The experience of participants in the e-publishing market (Amazon, Barnes & Noble, etc.) suggests that the revenue maximizing price is much less than $35 for whole books, let alone articles. The sweet spot for e-books is somewhere in the single digit dollar each range, and excerpts from anthologies (which is what journal articles are) ought to have a revenue maximizing price that is a bit lower.

7. The risk of leakage into a per article purchase arena that dilutes the core revenue stream of the journals is pretty low when the marginal access price for 98% of the users (academics and students) through the existing core part of the system is zero.

8. Academic journals are basically intermediate goods, not final consumer goods. Mostly, academic journals are read with an intent to engage in further research or analysis, often in published work or towards societally beneficial ends. Hence, the per article access value of these works to the public domain may be much higher than in other kinds of works, so the balancing test between private gain to the journals and public benefit from access to the journals is quite different and since they are usually organized as non-profits with a stated purpose to disseminate ideas and education the public regarding their fields, the journals, in principle, should care.

Non-academics who are reading academic journal articles are a pretty exceptional group of people who are far more likely to create further value derived from access to the journal article than, for example, vast numbers of students who have free access to the journal, and almost all final users of published works intended merely for entertainment value. In the same vein, while there is free access by request to any member of the public to decisions of the Colorado Supreme Court and Colorado Court of Appeals from the court system itself, view few non-lawyers (who would otherwise lack access to these materials) actually do request these materials but the people who do are often mass media journalists or political opinion makers who generate lots of value with their use of these materials.

9. Academic journals have some of the same character as comic books economically. While there is money to be made in simply selling comic books at a profit, a good share of the total revenue stream of a comic book publisher comes from licensing movies, television shows, costumes and toys for a much wider audience than the relatively modest number of people who actually buy and read comic books. Comic books translate into these big money spin off products very naturally. But, comic book publishers need to test the waters to see which ideas work by debuting them at a very low production cost with the elite group of people who buy new comic book titles (e.g. with new comic book characters) that allow comic book publishers to prove to investors that big investments in spin off goods are a safe bet. Academic journal articles serve a similar purpose - the amount of reception they get from the narrow pool of people who routinely read them without being pointed to them by someone else, and elite layman awareness of their existence (often via press releases or alumni publications) provide a major means of marketing the research programs in them to potential future sources of funding at a small up front cost.

10. In many cases, the intellectual property protections available for academic journals is modest, because of exceptions from intellectual property protection for pure ideas (as opposed to expressions of those ideas), for scientific principles and laws of nature, and for federally funded research. Fair use rights also tend to be more expensive in the case of academic articles than in the case of commercial entertainment publications. So, the viability of using legal means to restrain a black market are limited, making it desirable to set prices low enough that convenience alone discourages resort to black market copies. Low article purchase prices designed to secure compliance (either on a subscription or per use basis) - this is basically the strategy behind Netflix and Congressionally mandated mandatory content licensing for Internet radio stations and has been quite successful in those contexts.

ohwilleke writes:

"Suppose there were an outfit like Wikipedia dedicated to academic research."

There are. In physics there are arxiv (endorsement from an academic required) and vixra (no endorsement required). In law and a few other fields, the Social Science Research Network does the same thing.

Comments for this entry have been closed
Return to top