Does anybody know what the ratio between subscription fees and the real article processing charges (so actual administrative costs) is? So in other words, how much money is made out of the subscription fees.
This will vary wildly by journal, and may be extremely difficult to even meaningfully define, given that many journals are "bundled" together in confidential and institution-specific contracts. What is clear, however, is that:
- Large for-profit publishers are very concerned with keeping their de facto subscription costs difficult to determine, and have quite high profit margins, and
- The potential profit from open access subscription fees has drawn large numbers of predatory publishers to the field, and
- Society publishers are often able to sustain journal publications with quite minimal costs, especially online-only journals.
You may draw your own conclusions from this: my conclusion is that a lot of money is made from subscription fees.