In early December Knowledge Exchange (https://web.archive.org/web/20120611025454/http://www.knowledge-exchange.info/?), a partnership of JISC (United Kingdom), SURF (Netherlands), DEFF (Denmark) and DFG (Germany) released a report (https://web.archive.org/web/20120611025454/http://www.knowledge-exchange.info/Default.aspx?ID=413?) on submission fees that they had commissioned to Mark Ware Consulting (https://web.archive.org/web/20120611025454/http://mrkwr.wordpress.com/mark-ware-consulting/?). The report was also discussed by Robert Kiley on the UK PubMed Central blog (https://web.archive.org/web/20120611025454/http://ukpmc.blogspot.com/2010/12/submission-fees-viable-business-model.html?) and by Phil Davis on the Scholarly Kitchen blog (https://web.archive.org/web/20120611025454/http://scholarlykitchen.sspnet.org/2010/12/09/open-access-submission-fees/?).
Submission fees are more common than I thought, particularly in economics and the life sciences. The American Physiological Society journals, Cancer Research, FASEB Journal, Journal of Clinical Investigation, Journal of Immunology and Journal of Neuroscience all charge submission fees between $50 and $100, the Journal of Biological Chemistry dropped the $60 submission fee in 2010. Most journals that charge submission fees are society journals.
The report finds that submission fees would be particularly interesting for Open Access journals with high rejection rates (70% and more), as this would greatly reduce the article-procesing charges for accepted manuscripts. A journal that accepts 10% of submitted papers could use a $150 submission fee to reduce the fee for accepted manuscripts from $2500 to $1150. There wouldn’t really be a price reduction for journals that accept 50% of submitted papers.
Mark Ware reports that the Open Access publishers he talked to weren’t really interested in starting submission fees for their journals, mainly because of the risks involved in changing their business model. I personally believe that submission fees may be the only option for journals with high rejection rates to become Open Access journals (unless you want to cross-subsidize those journals). I like submission fees because they help to cover the actual costs involved, instead of the costs of handling manuscripts that are ultimately rejected being paid either by journal subscribers or the authors of accepted manuscripts.
The full 13-page report can be downloaded here (https://web.archive.org/web/20120611025454/http://www.knowledge-exchange.info/Files/Filer/downloads/Open%20Access/KE_Submission_fees_Short_Report_2010-11-25.pdf?).