« Steve Randy Waldman on the new Fed plan | Main | Hotel Theory »
Why does NBER want your $5?
John Eckstein, a loyal MR reader, writes:
NBER is still charging $5 to download a working paper [for non-academics]. Don't economists want to get their work out to the broadest audience possible? Given the number of downloads for most economic working papers it seems like the fee is just enough to discourage the dissemination of information, but not enough to raise significant funds. The budget is not on the NBER website so I can't check it myself.
I get these papers free, but that is because GMU pays a fee to NBER. The real source of money, Alex suggests, is from selling NBER membership to universities (inelastic demanders, just ask Kluwer), and that requires a nominal fee to stop professors from downloading it themselves. Still, so many other institutions offer free access, and now with the advent of blogging many non-academics want to read NBER papers too. Cannot the current system be reformed?
Posted by Tyler Cowen on December 13, 2007 at 07:35 AM in Economics | Permalink
Comments
The thing that really puzzles me is that, almost every time you link to an NBER paper, you link to a non-gated version. So who's paying the $5? What can it possibly be deterring? And why are universities paying at all? Is it worth the convenience to their professors of one-stop shopping? Do they just not understand Google?
Posted by: Andromeda at Dec 13, 2007 8:35:28 AM
Two thoughts:
- Non-academics are not likely interested in reading more than the abstract, or perhaps the NBER reporter summary, both of which are free.
- As Tyler said, the $5 is really just to enforce Library subscriptions, I wouldn't be surprised if most papers got zero paid downloads.
Posted by: Jack at Dec 13, 2007 8:53:30 AM
Actually the entire field of academic publishing is due for a complete reboot, not just economics.
Posted by: Matthew at Dec 13, 2007 8:57:34 AM
Jack - I'd expect that most non-academics who want to read the abstract would also want to at least skim the paper itself.
Posted by: eddie at Dec 13, 2007 9:13:12 AM
I think the fee is fine, but there are some caveats.
I've bought a couple NBER papers for the five bucks. I received plenty of consumer surplus from the transaction.
I'm not an academic, but I am a researcher and the papers have been valuable. Buying all of NBER's papers would cost a pretty penny, so I have to do some presearch. I have to find out what I expect to take out of the paper -- and it's not always in the abstract.
I haven't found a lemon paper from NBER yet, but I have in other journals (and they usually charge $20-$40 per paper). It's usually not that the paper is bad, it just didn't contain what I expected. I'd love to get a refund, but I can see how there could be problems doing so.
Posted by: JH at Dec 13, 2007 9:31:08 AM
The real mystery is why other journals charge so much for online access. (One example, which could be multiplied many times over: A two-page "in memoriam" article from the Economics of Education Review that costs $30 for a PDF.)
$30 is so far above marginal cost that it's ridiculous. What's going on?
Posted by: Stuart Buck at Dec 13, 2007 9:56:10 AM
NBER will let you have papers for free if you have an email address ending in .ro, for example. Get one, for free, in minutes, and you're done.
Posted by: Gabriel at Dec 13, 2007 10:00:26 AM
Thank goodness I kept my University log-in from Grad School. It is the only way I can do proper research for work.
Posted by: nvm at Dec 13, 2007 10:01:55 AM
I can see that NBER charges the fee to keep its library subscriptions paid up, but it makes no sense for authors to gate their work that way. Authors should *want* their work widely-disseminated. Perhaps they prefer the NBER cachet (so-and-so, of Harvard, NBER and God Squad, etc...) to staying on the outside/putting their stuff on SSRN. OTOH, they may not be interested in communicating with non-academics :\
Posted by: David Zetland at Dec 13, 2007 10:22:36 AM
Even more embarrassing, computer science is the same way. The ACM Digital Library is gated! I don't know what the ACM is using it's fees for.
It's the same in CS as Andomeda suggests, though. Most papers are available online for free, just not at the main digital archives.
Honestly, I don't think that the ACM and IEEE archives should be gated. Electronic copies of papers should be given out for free. If a fee must be charged, charge it to the authors -- they will gladly pay it to get their paper more accessible.
Posted by: Lex Spoon at Dec 13, 2007 11:05:21 AM
I've never understood why so many journals are locked behind pay-to-read walls. Why couldn't the federal government subsidize journals in order to put every single article ever produced online for free? Yes, there would be some problems here (how would you treat the dozens of new journals that might pop up to take advantage of the subsidy, etc?), but I imagine they could be worked out.
Posted by: Dave at Dec 13, 2007 11:19:07 AM
I do not even want to try to calculate the amount of time I have spent trying to find articles that could have been easily accessed on NBER (had they removed their $5 fee). But between EconLit, JSTOR and Google, let's just say I am able to spend the $5 I save on something more worthwhile: cafeteria food.
Posted by: Samir Nurmohamed at Dec 13, 2007 11:23:50 AM
John Eckstein,
Try finding the author's homepage (often linked by NBER), (I'm approximating) 80% of the time you can find a non-gated download in his vita or usually there is a publications/papers section.
Posted by: Mat at Dec 13, 2007 11:31:14 AM
NBER is $5, but that's trivial (especially given the likely availability of non-gated versions). What's not trivial are the $30 fees asked for by other publishers.
The speculation that these are really only designed to protect library subscriptions is reinforced by typical behavior -- I'll pay $5 for a paper that might be useful, but try to get a favor from somebody with an academic appointment rather than pay $30 for a paper that might be useful.
The whole system has not really adjusted at all to the realities of an internet world in which the dead-tree version of the journal has little value relative to a searchable electronic version that doesn't take up file space.
Posted by: ZBicyclist at Dec 13, 2007 11:42:32 AM
I have access for free to Nber because I live in the Third World.
By the way Uk University sites charge you as many as 30 Pounds for 48 hours
Posted by: Juan at Dec 13, 2007 12:46:28 PM
And with the exchange rate control i can not pay for it
Posted by: Juan at Dec 13, 2007 12:48:13 PM
If you want free papers, just sign up for an email with Yahoo India--then you have a email address that appears to be from India (I simply use India because the site is in English), and use that when asking for access.
Posted by: v at Dec 13, 2007 1:27:13 PM
Authors of academic journal articles don't get paid for them in money. What they want is citations. And open access articles tend to receive more citations than articles behind a "paywall".
Ed Felten observes the phenomenon of journal editorial boards defecting to open access, and asks, "It’s surprising that this sort of thing doesn’t happen more often. The value of a journal comes from the quality of articles in it; and this quality derives mostly from the reputations of the editorial board members and the work they do in choosing and editing articles. If a journal’s management takes a direction that the scientists on the editorial board don’t like, there is something they can do about it!"
So here's the economics question: Since open access benefits both authors and readers, why are the middlemen still in business?
Posted by: Don Marti at Dec 13, 2007 1:53:51 PM
Techie Solution: Block academic IP addresses from free downloads unless the University has a subscription. Free otherwise.
Posted by: Raul at Dec 13, 2007 3:05:08 PM
Does anyone think free puplications lead to a proliferation of low-quality junk? The incentives to the journal and the writers are lost if nobody attaches a cost to the quality of product they make. Is this a curse of the e-age?
Posted by: Rahul at Dec 13, 2007 3:07:49 PM
I've worked at universities that subscribe to NBER and ones that don't.
I think the fee is a way for NBER to solve a prisoners' dilemma between dean's and libraries.
Faculty just want the papers.
But, libraries don't want to pay for working papers and then again indirectly when the paper is published. They also are concerned about buying items that don't have a sustained checkout lifetime.
Deans don't want to pay either. They view this as a nasty way for the library to shift costs (which has some basis).
The prisoners' dilemma is that they are both steering towards an equilibrium where the NBER series is unavailable on campus. The small fee encourages faculty to push a bit away from that position.
Posted by: David Tufte at Dec 13, 2007 3:13:07 PM
I come from a discipline -- physics -- that managed to get this right. Researchers submit their papers sumtaneously to a journal and to an on-line pre-print archive (http://xxx.lanl.gov/) that is accessible to anyone. The archive was set up by a physicist as a public service, independent of any journal or professional society.
One factor that did make this easier for us is that this was already up and running by the time any journal or professional society had noticed the web, and had become so integral to research that they couldn't stop it. When the funding agency of the guy that set it up accused him of misusing one of his research postdocs to run it, he shut it down for a couple days. After the resulting uproar among research physicists, he got funding for a few more postdocs to help out.
Posted by: David Wright at Dec 13, 2007 4:01:02 PM
Rahul, if free meant "low-quality junk" wouldn't the OA journals tend to receive fewer citations from other authors, not more? Academics can use their footnotes to vote on the quality of others' work, just as webmasters can use their outgoing links. (Citation statistics is where the Google founders got the idea of PageRank, by the way.)
Posted by: Don Marti at Dec 13, 2007 4:44:02 PM
Don, I don't know really; I was just speculating. But in any case do you think citation # is a good indicator of either quality or even usefulness? I've nothing against OA ; I like it actually. I just think the barrier in general is pretty low OA or not.
Posted by: Rahul at Dec 13, 2007 7:49:29 PM
I worked for a newspaper, not a university, so an institutional subscription to NBER was not a possibility. I did once or twice pay for an individual article, but usually the reason I wanted to read an article was to determine whether it was something I could write about, and given the success rate -- about one in 10 -- $5 for a peek was far from trivial. Mostly, if I couldn't find a non-gated version, or the author wouldn't send me one, I just never wrote about the research at all.
Posted by: Linda Seebach at Dec 14, 2007 3:29:06 AM
