The Future of Academic Publishing

Obviously, as researchers are rather motivated by means of reputation and peer recognition than financial gains from selling their academic works, one might ask why copyright protection for academic works is at all needed. In an intriguing paper, Shavell (2010) investigates this question on whether an abolishment of copyright for academic works is reasonable. As discussed before, he argues as follows: (1) scientists seek reputation which is increasing in readership, (2) readership is higher under open access and hence scholarly esteem, (3) the publication costs due to a shift towards the “author-pays” principle under open access will be covered by most universities, and (4) there are several reasons why a shift towards an open access publishing model will not be smooth without legislative steps (Shavell 2010; Eger and Scheufen 2012b, pp. 54–55). In our analysis we have pointed to several countervailing effects of a shift towards an universal OA regime by reconsidering primarily argument (3) of the Shavell model. The question now is whether a removal of copyright protection for academic works—which necessarily forces an universal OA regime—is feasible and reasonable from a law and economics perspective. In this regard, a policy (law) abolishing copyright for academic works may already fail the feasibility benchmark. The primer reason for the infeasibility of an abolishment of copyright is of legal nature. Peukert (2013b) highlights that an abolishment of copyright is eventually incompatible with the Berne convention. Accordingly, Art. 2 (1) of the Berne convention expands copyright protection to all creations in literature, scientific research and art (Hansen 2005, p. 382). It is this lack in “legal feasibility” that may also explain why with the “Public Access to Science Act” a similar policy attempt failed in US congress in 2003 (Hansen 2005, p. 382; Peukert 2013b, p. 18).⁴ Shavell (2010) simply omits this fundamental insight by only mentioning in a footnote that “Paul Goldstein has suggested to [him] that elimination of copyright for academic works could lead to conflict with the obligations of the United States under the TRIPS Agreement” (Shavell 2010, p. 339).⁵ Beyond the feasibility requirement there are also several reasons why an abolishment of copyright is also not reasonable from an economics perspective. First and foremost, our analysis has shown that OA may not necessarily increase the effort incentives of researchers due to the rent-seeking motive in the publishing game. Furthermore, there are several other authors who have been pointing to countervailing effects of copyright removal (McCabe and Snyder 2004, 2005; Mueller-Langer and Watt 2010, 2012).⁶ Most importantly, copyright may have important implications for the reputation of journals and hence scholarly esteem in the first place. If we believe in this argument, a removal of copyright would necessarily harm the ability of researchers to receive credit for their writings. As a consequence, there are several reasons—also beyond its incompatibility with international law—why an abolishment of copyright (forced OA) is not an appropriate means to promote OA publishing. However, a reform limiting the scope of copyright (i.e. exceptions and limitations of copyright protection) may be feasible and reasonable in this regard.

One option that seeks to limit the scope of copyright protection to accommodate the needs of science is the introduction of a so called “inalienable right of secondary publication”.⁷ Such an “inalienable right of secondary publication” would give the author more bargaining power in the contractual relationships with publishers. As we have seen, publishers typically ask for a transfer of an exclusive commercial right from the author. Nowadays, most of the journal publishers’ copyright agreements do allow for some form of self-archiving by the author. Nevertheless, it is still to the publisher to decide whether an author may also deposit a copy (pre- or post-print) of her own work in an online repository. An “inalienable right of secondary publication” would give the author the freedom to deposit a pre- and/or post-print version of her journal publication. The fact that the German parliament (Deutscher Bundestag) has enacted a bill for an “inalienable right of secondary publication” on 27 June 2013 shows that a modification of copyright in this regard is generally a feasible option for strengthening the position of authors and indirectly for promoting OA publishing. Nevertheless, the feasibility of the implementation of such an “inalienable right of secondary publication” as a general exception of copyright will decisively depend on the ability of national states to achieve collective action. Most importantly, a unilateral step forward by a single nation carries the risk of weakening both authors’ and publishers’ position in the international competition of science. As discussed in Sect. 4.​2, a reform process that forces a deliberalisation of the international copyright framework as codified under Art. 13 of the TRIPS agreement and Art. 10 of the WIPO Copyright Treaty is required for harmonizing the rules for limiting the scope of copyright and setting equal conditions in the competition of authors in the international publishing game. With regard to the reasonability of an “inalienable right of secondary publication” our assessment will depend on the specific design of the copyright exception. First and foremost, an “inalienable right of secondary publication” would give the author the freedom for reuse of her own published articles without any requirement clause. Accordingly, there is no reason to believe that such a policy could interfere with the researcher’s “freedom to publish” and could hence restrict the researcher in her ability to receive credit.⁸ However, a unilateral solution where, for instance, only German authors enjoy the freedom for secondary publication could easily generate the opposite effect. That is, publishers could indirectly circumvent such rules in the selection of authors. Obviously, a disputatious argument, but one that should be considered in a truly global science community.⁹ Second, as the “inalienable right of secondary publication” does not impose any further conditions or requirements for authors to provide OA to their works, it may be limited in achieving the actual goal of OA—immediate and unrestricted access to scientific knowledge. In this regard, a combination of different policies may be necessary. A possible instrument may be to combine an “inalienable right of secondary publication” with a contractual requirement imposed by funding agencies.

A complementary approach to a reform of copyright is to reach contractual commitment among researchers to deposit a copy of their un-/published paper in an institutional or subject based repository. We have already discussed the various opportunities that the government and funding agencies have to contractually regulate OA to publicly funded research. We have seen that in fact a growing amount of countries and several large funding agencies have introduced some form of OA policy that requires OA to publicly-funded research.¹⁰ This suggests that OA mandates—i.e. a contractual commitment of the author to provide OA to her publicly-funded research—may be a feasible policy option to foster an unrestricted access to scientific knowledge. A frequently raised concern against such contractual requirements is the argument of academic freedom or particularly the constitutionally guaranteed freedom of publication. In this regard, a group of researchers among the literary scholar Roland Reuß—that became known as the “Heidelberger Appell”—have raised such concerns.¹¹ Taking account of these concerns, an assessment of OA mandates from both feasibility and reasonability perspectives will decisively depend on the design of such mandates. In this context, both roads of OA—i.e. the gold road (OA journals) and the green road (self-archiving or repositories)—provide options for such mandates. Obviously, a policy mandating authors to publish in OA journals only would necessarily undermine the freedom to publish and would likely forward more protest from both researchers and publishers. Not only that such a “gold mandate” would be against the law,¹² there would also be strong arguments against it from an economics perspective. We have seen that a policy should allow for a balancing of both sides, ensuring that researchers receive adequate credit for their writings and maximize knowledge diffusion at minimal costs. Accordingly, a commitment that would bind the author to publish in OA journals would—given the prevailing competitive advantage of established CA journals in terms of impact—force OA publications at the costs of lower impact/credit for authors who received such funding. Authors without such contractual constraints—for example due to private funding or other means that offer her freedom to publish—would necessarily have an advantage. That is, there are many arguments why a “gold mandate” would neither fulfil the feasibility nor the reasonability benchmark. However, our conclusions may be different when looking at the latter form of OA mandate—the “green mandate”, i.e. a request or requirement to provide OA to publicly funded research by means of repositories. In this context, the “academic freedom” or “freedom to publish” argument immediately looses weight. The reason is easy to grasp: the “green mandate” leaves it to the author where to publish her works, but asks for a self-archiving of her works after a certain period of time (embargo period). The reasonability of a “green mandate”, however, will decisively depend upon the design and definition of an OA policy for funding agencies.¹³ Obviously, an OA request or encouragement to self-archive a pre- or post-print version of a paper seems little promising for reaching a sufficient level of OA. In fact, with the “alliance of scientific organizations” in Germany many attempts were made to encourage researchers to deposit a copy of their work in both subject-based and institutionally based repositories. In Sect. 3.​2 we highlighted the “Max Planck Digital Library” (MPDL) which provides an online platform for the exchange of publications by affiliated authors. The wording of the MPG OA policy that “calls upon its academic staff to observe the principle of public availability of basic research” (MPG OA policy)¹⁴ already reveals its non-binding character. Not surprisingly, Eger et al. (2013) find that less than 20 % of the 2,151 respondents of their survey show experiences with self-archiving—with large differences between the disciplines.¹⁵ Consequently, a simple request that asks authors for self-archiving is little promising. The second OA policy option for funding agencies are OA mandates that explicitly require green OA as a contractual condition for the receipt of funding. In this regard, we already discussed different types of mandates and showed that only rights-retention mandates may guarantee that academic works do not remain non-OA.¹⁶ In this context, the funding agencies require to retain a non-exclusive right to authorize OA throughout self-archiving platforms (Suber 2012, p. 80). In general, this form of OA mandate seems promising as it may provide with an effective tool to enhance free online availability of publicly funded research without interfering with the researcher’s “freedom to publish”. Nevertheless, the particular design and means by which such a rights-retention mandate is enforceable may be subject to debate. This debate may involve two general questions. On the one hand, a right retention may be reached by different means. Suber (2012) highlights the Harvard approach which pioneered this approach for universities. Here, the “faculty members vote to give the university a standing nonexclusive right […] to make their future work OA through the institutional repository” (Suber 2012, p. 80) but with the flexibility “to opt out of the grant of permission to the university, but not out of the deposit requirement” (Suber 2012, p. 80). Another means may consider the option to require OA by copyright legislation. We have already discussed the option of an “inalienable right of secondary publication” that gives the author more bargaining power in the contractual relationship with journal publishers. In this regard, both policies (a right-retention mandate by funding agencies and an inalienable right of secondary publication by copyright legislation) may complement one another. On the other hand, the specific design of such a mandate involves several dimensions. The first dimension may be the length of the embargo period which trades off the need to offer enough incentives for publishers to provide the necessary publishing services and the costs of banning an OA for a certain period after publication. Whether the embargo period should be 6 or 12 months after publication may then depend on many aspects and may hence differ between disciplines and countries.¹⁷ Second, it will be debatable how to ensure that authors follow the requirement to provide OA to their publicly funded works. Here, Stodden (2009) proposes the Reproducible Research Standard (RRS) as a promising solution for both to ensure attribution and facilitate the sharing of academic works. Other approaches revert to the option to require a deposit of an electronic version at the national library which would then provide OA to all the works.¹⁸ Third, the type or form of the work that is deposited. No doubt, an effective OA mandate would have to assure that a final version of the paper (including all changes from revisions) is made available. Other than that would leave us with an insufficient substitute for the original. Obviously, the shape of such a policy leaves us with enough caveat for further research to not only ensure the feasibility of an OA mandate for publicly funded research, but also to guarantee an effective policy framework.

Furthermore, several authors have advocated that scientific societies and universities should launch new OA journals and should hence act as a counterbalance to the dominance of commercial publishers. A closer look at the history of the academic journal publishing market actually shows that both academic societies and universities can look back on a great tradition in publishing academic journals. We have seen that the first journal titles were launched throughout the sole initiative of academic association. Commercial publishers started to enter the market for academic publishing after the second world war. The dominance of commercial publishers—with all the consequences we are currently facing (serial crisis)—is a rather new phenomenon. The vision in this process is easy to grasp: If academic societies and universities would become more active as publishers of academic journals they could create a balance weight against the prevailing dominance of commercial publishers. Two arguments are broad forward in the literature to believe that this approach could be a feasible solution. First and foremost, several universities and academic societies have experiences in publishing journals.¹⁹ A famous example is the Oxford University Press as the world’s largest university press. The Oxford University Press currently offers a portfolio of 250 journal titles.²⁰ Second, this particular group has a strong incentive to self-enforce a change in paradigm as the universities, in general, and the university libraries, in particular, are directly affected by the constraints imposed by the serial crisis. There is no reason to believe why universities and/or academic societies are not able to launch new journal titles. However, just like OA journals in general, also these journals will necessarily lack in reputation. As a matter of fact, especially young researchers would be reluctant to publish in these journals as they would provide with a lower credit or impact as compared to well-established commercial publishers. This leaves us with the previously outlined dilemma of OA journals and the question whether the reward structure in academia may be seen as the biggest hurdle in an evolutionary process towards an OA publishing mode.

Thus, far from any exogenous market intervention a new paradigm of OA could be induced from within the system itself. In fact, a closer look at the “OA dilemma” reveals that it originates in the prevailing reward structure, i.e. the ways scholars receive credit for their performance.²¹ We have seen that the general performance measurement used for assessing the standing of a researcher within her peer group is the impact factor of a journal publication. Rankings are calculated which shall display the performance of individuals or institutions based on a weighted sum of all publications, where the impact factor reflects the weight or quality of a certain publication.²² Nowadays, the tenure procedure has evolved to unilaterally focus on (journal)²³ publications only and has induced an environment that is typically referred to as the “publish or perish” environment of scientific research.²⁴ Several others criticize that a “taste for rankings” paradigm has crowded out the traditional academic notion of a “taste for science”, which emphasizes the relevance of motivational factors like autonomy and peer recognition as opposed to monetary rewards (Osterloh 2013, p. 106; Roach and Sauermann 2010).²⁵ As a matter of fact, it is exactly this “reward system” that may induce a lock-in to the traditional publishing model and may prevent an evolutionary process towards an otherwise superior OA regime. As we have seen in Sect. 3.​1 there is clear empirical evidence for a reputation or impact factor advantage of established CA publishers over OA publishers. Consequently, researchers seem to be locked-in to the weak equilibrium, where especially young researcher will prefer to publish in established CA journals to receive credit for their writings.²⁶ Others have been pointing to this dilemma as the chicken-egg characteristic of a co-existence of both regimes (CA and OA), where publishing in a CA journal dominates any OA attempt and leaves the academic community locked-in to the weak Nash-equilibrium.²⁷ A solution to this “OA dilemma” may be to induce a self-enforcing process towards OA by a reconceptualization of the reward system in academia. This reconceptualization could take many different forms which could be enforced by the community of academics itself. First and foremost, performance measures and tenure procedures should not be unilateraly focused on rankings only. Consciousness about the various problems associated with rankings is important is this regard. Rankings often glorify realities. The position of a journal, institution or researcher in a ranking is easier to assess than the real quality of research. In economics, this is known as the “multiple tasking”-effect (Holmström and Migrom 1991; Ethiraj and Levinthal 2009). In complex systems, like scientific research, goals in terms of performance measurements induce a situation where easy targets (like the position in a ranking) are reached, while the often more important goals (like scientific progress and the diffusion of knowledge) are neglected (Osterloh 2013, p. 105). A solution would be the introduction of additional measurement that directly account for OA features of a researcher’s CV. With other words: The degree to which a researcher provides OA to her research, both in terms of publications in OA journals or self-archiving, should be incorporated when assessing the performance of a researcher. As a consequence, such a performance measure would also account for the positive spill-overs to academia and practice likewise. Second, well-known researchers (e.g. nobelprice laureates) should take a lead in the transition towards more OA publishing. Both OA publications but also editorial positions of renown researchers could serve as a signal of valuation in the academic publishing market.²⁸ Last but not least, the academic society should take advantage of new opportunities like “Open Assessment” for complementing the quality selection process in academic publishing. Obviously, quality selection is important to prevent from the Akerlof lemmons. However, several scholars have been criticizing the ability of the traditional “peer review” model to reach an objective selection process.²⁹ In this context, an Open Assessment similar to customer evaluations in online markets could be used as a means to complement the quality assessment. A broader discussion on the design of the future quality selection process seems necessary.