“The average research grant in South Africa, excluding strategic and infrastructure investments, is approximately 146,000 rand (€8,500). In 2021, the average charge for publishing an open-access research paper was nearly €1,600. High-impact journals charge far more: €9,500 at Nature, for example.

Here, in a nutshell, is the inequity of the financial model of open-access publishing. As currently constituted, publishing charges are stifling research capability and progress, as well as the career progression of researchers in low- and middle-income countries, and preventing a full transition to open research.

We need to move towards a globally agreed payment system for academic publishing services that is fair, equitable, transparent, and does not require the author to pay. In this article, we sketch the outline of such an alternative payment system….

It is unclear why APCs and transformative agreements are not priced as a function of what local markets can bear. The consequence, however, is stark: for the most part, researchers and institutions based in lower- and middle-income countries simply cannot afford either of these pay-per-article models. While some of these countries have negotiated cost-neutral transformative agreements, it is not clear whether these are equitable in terms of local purchasing power.

In much of the world, the money is not there to pay APCs geared to the richest nations—especially as APCs have consistently risen faster than inflation. Countries in the Organisation for Economic Co-operation and Development spend an average of 2.2 per cent of gross domestic product on R&D. For the United States, the figure is 3.5 per cent. In Latin America and the Caribbean, in contrast, the average is 0.7 per cent, while South Africa’s figure of 0.75 per cent is well above the continent’s average of just 0.4 per cent.

Admittedly, some researchers may apply for publishing fees to be waived, but there is no globally agreed way for publishers to handle waivers, and researchers working in middle-income countries tend not to be eligible for such support. Moreover, waivers are often perceived as patronising and neocolonial. They are an in-or-out mechanism unilaterally controlled by the publisher, denying any agency to recipients.

Asking for a waiver imposes significant effort on authors. Waivers are also a financial risk to publishers who are understandably reluctant to award them. A submission to a 2020 consultation by the UK foreign ministry calculated that 57 per cent of hybrid journals from major publishers, which offer both subscriptions and open access, and 70 per cent of open-access journals published by small independent or university presses, did not offer fee waivers or discounts. The most visible global initiative delivering waivers, Research4Life, reports that, while effective, usage of its resources remains limited and has declined.

The system for meeting the costs of academic publishing is globally inequitable. This was underscored by the landmark United Nations Educational, Scientific and Cultural Organization (Unesco) Recommendation on Open Science of November 2021, which insisted that scholarly communication adopt “the principles of open, transparent and equitable access”. The recent memorandum from the White House Office of Science and Technology Policy, setting out a plan for open access to federally funded research from 2026, adds ‘equitable’ as a third condition to the more familiar requirements for ‘free’ and ‘immediate’ access. Equitable open access has therefore assumed a particular urgency….”