When I started studying economics in the ‘sixties, there was no Xerox. Journals were printed, and then mailed. Because printing (type-setting by hand, no computer at that time) was expensive, only selected articles were distributed through journals, and journal editors had to select carefully. Researchers and even students subscribed to journals in order to have articles of interest available; otherwise they had to copy them by hand, or excerpt them, or go to the library to have a look. Distribution by print was the cheapest and most economic way of distributing research.
Hence the journals had a dual function: 1.) They selected research articles and 2.) they distributed them. The first function (selection) was necessary because printing–especially printing of mathematical formulae–was quite expensive. So the bundling of selection and distribution had an economic reason.
This reason has vanished. It is possible to distribute practically for free (through MPRA for instance). So the question is: Do we need journals, simply for the purpose of selecting articles, as the function of distributing articles is redundant nowadays. Let me share some thoughts on the issue. I concentrate on research journals, whether open access or not. Survey journals like the Journal of Economic Perspectives or commentary journals like Economists’ Voice are another matter.
Do We Need Quality Stamps?
Some people argue that journals provide a “quality stamp” for scientific contributions, just like rating agencies assess firms or assets. We know that rating agencies may induce unwarranted herding effects, yet the point that journals perform a rating function is true. But is it needed? And if needed, can’t it be provided more cheaply?
As to whether a quality stamp is needed: This may be different for different groups of users. So look at different groups that may benefit from a quality stamp.
Researchers
In my fields of research, I certainly do not use journal names for selecting articles. I search the Web and have my subscription to NEP. Most articles (99%) in top-ranking journals are of no concern to me because they discuss issues I don’t work on and seem too specialized, technical and boring as to make it worthwhile to read more than the abstract. But I get the abstract much earlier through NEP and other services. And further, I obtain the articles I am really interested in much earlier (one or two years earlier) over the net than through the journals. (Note that the articles in good journals are typically available on the Net at the time of publication.) If I find an article on the net that I like, typically a pre-print, and see it later published in a good journal, I feel a kind of satisfaction about the journal, but this does not seem to justify the existence of journals.
Further, I am not interested in seeing only the good papers that some referees approve of. As I know my field, I do not think that referees know better. Actually many papers in top journals are not so good, and mediocre journals publish excellent papers. Further, many rejected papers are rejected for reasons such as being badly written, ill organized, or employing faulty reasoning, but they often do contain useful references and interesting ideas, and therefore they interest me as much as a superbly crafted paper elaborating on rather sterile detail.
Yet there may be the benefit coming with having an article revised during the refereeing process. The probability that the mathematics are correct is slightly increased. Typically, the exposition is improved, too. Further, the references are enlarged by adding some quite relevant stuff, but also by adding things suggested by the referees for sundry reasons that hurt overall consistency. But this does not hurt much.
The benefits going with having an article refereed carry side-effects, however: Sometimes the editors’ and referees’ demands make papers worse. In the same vein, have a look at Bruno Frey’s amusing paper, and especially at what he reports about Robert Frank.
Regarding the publishing of my own research I see that publishing in a journal does not affect citations, but making a paper available on the net does so. Hence journal publication is of very limited value to me (but I don’t have to care about the journals I am publishing in because I am close to retirement).
So, overall, I think that researchers do not benefit significantly from journals that publish research papers.
Hiring Committees
A benefit from having quality stamps is that this helps hiring committees to select candidates under conditions of ignorance. This may be true, but I would consider this a dysfunction: In the first place, hiring committees should comprise knowledgeable members; otherwise you would not need hiring committees and leave the decision to bureaucrats; and second, citation numbers are much better indicators for the impact of an author’s work than the journals the author has published in. So ignorant hiring committees may better resort to RePEc citation scores, rather than being enthused by journal titles. (But then they will end up with hiring candidates who work in fields many people work in. So they end up with conventional candidates, rather than creative ones. But this will be the case whenever you have incompetent hiring committees.) In any case, hiring committees won’t need journals, as RePEc citation scores are independent of journal names and do not rely on the existence of journals.
However, the reliance of hiring committees on journal rankings may entail strictly negative consequences. I read, for instance, that Notre Dame University intends to dissolve the department of economic history because the economic historians do not publish in mainstream journals.
It seems to me that hiring committees do not benefit from the existence of journals either.
Libraries
It is sometimes said that journals permit journal rankings, and this is a help for librarians for deciding which journal to subscribe to. This is, of course, not an argument for supporting journals. Without journals, there would be no problem of selecting journals, and the librarians could concentrate on selecting books.
So I conclude that libraries would perform better if we had no journals.
Economics Without Journals
Imagine we had no economics journals. What would happen? Presumably people would write more books. I would consider this an advantage, as knowledge is much too fragmented at the moment. Further, institutions would be in demand to channel the flow of information better than possible through journals, such as blogs specializing on some topic or another, and meta-blogs like Econ Academics. I could imagine that collections of papers on certain topics would emerge. The Special Issues feature of the economics E-journal provides an example.
A Suggestion for a Next Step
My impression is that the existence of journals is a feature of the past. Journals will die, and this will be an improvement for academic economics. The process will be sped up if new ways of channeling information are devised. So here is just one idea:
I could think, for RePEc, to devise a feature that lists related papers to any given paper. Google Scholar has a feature like that, but I think that could be improved tremendously for our specific purposes. An easy way would be to look at the citations of any given paper and give all papers with similar citations. This could, theoretically, be achieved by building on the citation data created by the CitEc project. If someone with programming expertise could adopt such a project, this would be a great help for economists world-wide. (As a side effect, such a feature would put pressure on Elsevier to release its citation data.)
There are certainly many more suggestions. I am looking forward to see them, perhaps in comments to this blog. And certainly my general point must be controversial. I must have overlooked some important aspects. The world can not be as inefficient as I portray it. Otherwise we would have no journals right now.
Maybe we can have an exchange of ideas.
The RePEc Blog 
[…] I’m not alone in thinking that the end for journals is near. Both the Economic Logician and Ekkehart Schlicht of the RePEc blog have recently posted about this idea…..in an admittedly more convincing fashion than I would […]
Very good arguments and I agree with most of it.
But as an librarian I have one counter-argument: there is still no acceptable solution for the long-time preservation of electronic documents. As long as there is no solution the printed Paper or Book is still the best long-term preservation medium we have.
Most people forgot that RePEc or Google Scholar are only portals and not archives or libraries. O.K. they are fantastic and great information intermediators but if the publisher of the Paper deletes the document it is lost for RePEc and for future generations.
Sometimes Economics Scientist tell me, that a Paper which is not published in a ranked journal is not worth to archive it in an library. So as a RePEc-Fan You can say, a E-Paper which is not cited in Citec (or not found in RepEc) ist not worth to archive it.
But maybe in the future someone founds out (if it has been long-term preserved) that the ideas in this forgotten paper are very good and relevant for future problems.
I remember of Gregor Mendels inheritance laws (foundation of genetics) which were forgotten at his time and 16 years after his dead rediscovered (because printed in a journal and archived in libraries). Imagine if a nowadays “Sociologist Mendel” post a Paper with relevant content for Economists on a website which is not ranked very high in Google and he is not aware of MPRA, because he is not interested in Economics???
I agree with the point, but there is a slight misunderstanding. I have not taken “Journals” as meaning “printed journals.” There are many journals that are not printed, but they are ranked, and sometimes highly esteemed, like Berkeley journals. Yet the long-term availability is an issue. But why not preserve everything? Why not have a copyright library for everything? By the way, Thomas Krichel works on establishing a long-term archive for RePEc.
Regarding citation scores, I do not agree with those economists. They seem to have internalized this nonsense, and they must, because their career depends on their rankings.
I have commented on quality rankings in a comment on <a href="http://www.andrewdyck.com/the-end-for-journals-is-closer-than-we-think/" here. As explained my blog, I don’t think we need quality rankings, but rather a mechanism to find out relevant contributions, independent of “quality”. My view is that we should care about the progress of economics, not about the problems of hiring committees or, pardon, librarians. These people need and should be given other resources.
And last but not least I would not be so sure that Mendel’s publication would have survived a refereeing process of the nature encountered in economics to-day (but not in other disciplines to that degree).
While most seasoned members of a given field have the ability to discern between quality articles and more poorly written articles. As someone still towards the beginning of my academic career, I’m not sure what I would have done when I was first starting out without those journals to fall back on. While I definitely do see a use for more aggregation and chronicling sites that contain a wider range of articles arranged topically that can be accessed on-demand, I do think that the academic “stamp of approval” can be very useful for newbies. Is it possible that journals may still have a place as such but perhaps ought to be given less import? Thoughts?