Skip to content

Presentation

The purpose of this paper is to determine whether or not an open-access policy actually improves the visibility of research, and to what extent. The study provides essential insights for decision-making regarding the dissemination of research findings and the observed impact of access restrictions on the public’s (both researchers and the general public) access to these findings. The key issues in the debate are the choice of the ‘moving barrier’, that is to say, the period after publication before the journal itself makes the article openly accessible, and the duration of the ‘embargo’, that is to say, the minimum period before the journal authorises researchers to self-archive their articles.

The study sought to quantify the impact of the duration of the rolling embargo on the readership of the journal and the research itself. The results indicate that the existence of a barrier to dissemination causes the journal to lose readership, even after a period of just one year. Given that the marginal costs of disseminating articles on digital platforms are very low, or even zero, this loss of readership represents what is known as a ‘deadweight’ loss. Our findings therefore support the implementation of a relatively short rolling embargo period (less than one year) compared with the periods discussed in the public debate for the social sciences and humanities (2 to 3 years).

Key Results

    • There are 936 social sciences and humanities journals in France, 32 per cent of which are available on the Cairn and Revues.org platforms. Nearly 400 journals had not published an issue in the year leading up to the survey (from 2013 to spring 2014).
    • All other things being equal, the longer the rolling embargo period, the lower the journal’s annual ‘views’ figure: journals with a short rolling embargo are therefore viewed more than others.
    • The longer the paywall duration, the greater the loss of views from one year to the next. The loss of readership associated with the paywall becomes apparent after just one year.
    • Whilst the number of ‘views’ is higher the shorter the embargo period, the rebound effect upon lifting the embargo is greater the longer the embargo period.
    • The journal’s total readership is lower the longer the paywall remains in place. The loss of readership becomes apparent after just one year of a rolling paywall.

Partners

Ministère de l’enseignement supérieur et de la recherche
Last modified: July 21, 2026