To be honest, the more information I get on this topic, the more confused I am. There seems to be too many factors at play, and it is difficult to judge their relative importance. I believe what I wrote is true; problem is, the important parts may be still missing. Here are the little details I learned yesterday:
The system encourages winners and sabotages new players. Yet, there is more than one winner per field. Depending on the field, there are like 15 journals that can give you points for publishing. Is this competition not enough to bring down prices? Prices can be as high as reader's $30 for reading a single article, or author's $1000 for removing a paywall from their article. I still have problem to believe these numbers. For comparison, $60 gets you a big paper book from the same publisher.
There are some "open journals" that give away articles freely. Before you say "open source FTW", let me remind you that these journals also ask like $1000 from the author. Hard to believe it's justified by their expenses.
Some publishers publish many journals, some of them being winners, some of them non-winners. Perhaps the money gained from winners is used to support non-winners; with the ultimate goal of winning in every scientific field?
These publishers also publish scientific books. Could there be some relation between book costs and article costs? For example, could the articles be very expensive to make the books cheaper? Just a speculation.
Economically speaking, from reader's point of view, each journal is a monopoly. (It does not matter that other journal is cheaper, you cannot buy the same article from another journal.) On the other hand, from publisher's point of view, most articles are commodities. Is perhaps this enough to explain the prices?
Also, in the article market you have very strong third degree price discrimination. For a layman, the cost may be $30 for an article; for a student in poor country it may be like $2 for yearly access to the whole library; for a scientist the cost is virtually zero, because the employer pays the costs. Of course there is a 'black market' solution: make friends at university and let them download the articles for you at lower price. Also if you are a university student, it is good to make friends in other universities, because each university may have access to different databases.
Let's not forget that this whole thing is a signalling market. By publishing in prestigious journals you signal that you are a good scientist. Also, if you are a good scientist, your employer pays the costs. So you don't care about the prices; you may even like them, because they help you signal better.
tl;dr -- Scientific article market is very unusual. Our intuitions about typical markets may be completely wrong.
Prices can be as high as reader's $30 for reading a single article, or author's $1000 for removing a paywall from their article. I still have problem to believe these numbers.
Those numbers felt a bit high to me too, so I did a very quick search to find something suggesting why an article might cost a thousand-odd dollars. I found an old white paper from the Public Library of Science — it lists some costs that might help to explain why open access publication costs run to four figures.
PLoS reckoned that it might cost PLoS Biology $20 to manage each submi...
link
Razib Khan found this paragraph rather striking (who is reminded of this episode of South Park) and I would tend to agree that its a rather convincing argument.
Are publishers really so successful as rent seekers or is there something the original article is missing here? Also what useful strategies would LWers recommend to help minimize costs for someone trying to practice the virtue of scholarship? The obvious suggestions (implied in the article) seem to be emailing authors (and perhaps those suscribed) asking for the papers and acquiring and paying for membership in some libraries.
Another obvious option is using ... liberated databases of such academic papers.
Edit: Just wondering, has this been discussed before on Lesswrong?