Confronting the Crisis in Scientific Publishing: Latency, Licensing and Access

The serials crisis in scientific publishing can be traced to the long duration of copyright protection and the assignment of copyright by researchers to publishers. Over-protection of scientific literature has enabled commercial publishers to increase subscription rates to a point at which access to scientific information has been curtailed with negative social welfare consequences. The uniformity costs imposed by such over-protection can be addressed by tailoring intellectual property rights, either through legal change or private ordering.

Current open access channels of distribution offer alternative approaches to scientific publishing, but neither the Green OA self-archiving nor the Gold OA author-pays models has yet achieved widespread acceptance. Moreover, recent proposals to abolish copyright protection for academic works, while theoretically attractive, may be difficult to implement in view of current legislative and judicial dispositions.

Likewise, funder open access mandates such as the NIH OA Policy, which are already responsible for the public release of millions of scientific articles, are susceptible to various risks and political uncertainty.

In this article, I propose an alternative private ordering solution based on latency values observed in open access stakeholder negotiation settings. Under this proposal, research institutions would collectively develop and adopt publication agreements that do not transfer copyright ownership to publishers, but instead grant publishers a one-year exclusive period in which to publish a work.

This limited period of exclusivity should enable the publisher to recoup its costs and a reasonable profit through subscription revenues, while restoring control of the article copyright to the author at the end of the exclusivity period. This balanced approach addresses the needs of both publishers and the scientific community, and would, I believe, avoid many of the challenges faced by existing open access models.