Peter Suber has an interesting item in his Open Access News, referring to a report from the Research Information Network in the UK on the costs of OA publishing vs. the current "pay to read" model.
As Peter points out, a glaring false assumption of the report is that the only alternative to "pay to read" is "pay to publish" or author charging. This gives point to my insistence that to lump all OA publishing under the one label of "Gold OA" leads to this kind of assumption, since it is in the interests of publishers to pretend that author charging IS the only alternative. Let us talk of "Platinum OA" when we mean "free to publish and free to read" - the model of Information Research and many more OA journals. As Peter again points out 67% of journals in the DOAJ make no author charges.
One example of the impact of the false assumption is the calculation that the savings to libraries of a move to e-publication only would be offset by the need to pay 17.5% VAT on the subscription. This amounting, according to the report, to £5million. But "Platinum" journals incur no VAT, since there is no subscription - and, given the low costs of self-publication by universities or consortia of universities of e-journals, a true analysis would ask, "How many free journals could the UK University system publish with £5million?" I think it would be more than one or two!
Until universities break free of the false assumptions under which this report has been written, they will be locked into commercial relationships with publishers that will inevitably limit access to scholarship and research.