Den Kreislauf der Zurückweisung durchbrechen
Das wissenschaftliche Publikationswesen stützt sich weiterhin auf veraltete Kennzahlen aus der Ära der Subskriptionsmodelle, obwohl sich die Branche in Richtung „Gold Open Access“ bewegt. Um dem zu begegnen, sollte das derzeit fragmentierte System durch einen einheitlichen Beschaffungsrahmen ersetzt werden, der auf einem zentralen Budget sowie klaren Evaluationsstandards basiert.

Essay
Rejecting the Rejection Cycle
Towards an Academic Publishing Market that Truly Supports Open Science: An Essay
Peter Roth
Aka Ltd. Independent Academic Publishing, Grosspeteranlage 29, 4052 Basel, Switerland;
peter.roth@aka.swiss
Abstract
Academic publishing is undergoing a fundamental transition, yet many of its institutional structures, evaluation criteria, and funding mechanisms still reflect the subscription era. This essay argues that the transition to Gold Open Access is not simply a change in business model but also a change in the rules of the publishing market. Using the analogy of moving from poker to chess, it contrasts a system based partly on restricted access and limited price transparency with one in which research findings, editorial processes, and publishing costs can become more visible. This essay examines several consequences of this transition: publish-or-perish incentives, repeated rejection and peer-review cycles, differing editorial concepts, and the often poorly defined notion of journal ‘quality’. It argues that highly selective journals have a legitimate role, but that the much larger group of mainstream journals requires different measures of value, including accessibility, transparency, publication speed, indexing, and usefulness to researchers. Finally, the essay proposes treating Gold Open Access and subscription publishing as one competitive market. Libraries, funders, and research institutions should evaluate both models within the same publishing budgets and according to transparent criteria, including price, indexing, peer-review practices, copyright retention, reuse rights, and editorial services. Such a framework could increase competition, improve transparency, and ultimately reduce the total cost of scholarly communication.
Keywords: scholarly publishing; gold open access; subscription publishing; peer review; research evaluation; publishing costs; academic libraries; scholarly communication
---------------------------------------
“There are simply too many academic journals these days... Too few submissions are rejected… Everything’s driven by the mantra, ‘Publish or perish’... Publication lists nowadays define not only individual academic careers but the standing of entire research institutions… Everyone’s talking about quality, but no-one seems able to define it… A handful of big players dominate key parts of the market... Policy-makers pay lip service to Open Access, but they don’t back up their words with actions…. The academic publishing sector is just an impenetrable mass of contradictory guidelines, classifications, and outmoded procedures… And it’s all too expensive – for everyone!”
What’s going on in the world of academic publishing? The market has never really recovered from the rise of Open Access (OA). This new publishing model constituted a hugely disruptive intervention in a well-established market, challenging a model around which research institutions, libraries, publishers and researchers had comfortably organized themselves for decades.
Inspired by the conviction that scientific research should be freely available for everyone to read, OA offered a completely new route for funding the publication of scientific papers in the form of Article Processing Charges (APCs) – charges paid by the author to the publisher for the service of reviewing, editing and publishing his or her paper in a universally accessible format. Like David defying Goliath, this approach contravened the traditional subscription-based publication model.
Thirty years after its first appearance, however, this David has grown to massive dimensions, while the Goliath of subscription-based academic publishing is still standing. Two very diverse routes to publication exist for individual scholars today, and thus two means of facilitating the publication of academic output exist for research institutions.
The central argument of this essay is not that Open Access is good and subscription publishing bad. It is, rather, that the economic and informational rules of scholarly publishing have changed, while institutional incentives, evaluation systems and budgets still partly operate according to the logic of the subscription era. If Open Access and subscription publishing were treated as parts of one competitive market, with comparable criteria and a common view of total costs, the system could become more transparent, more efficient and, potentially, less expensive. And that would be in everyone’s interests.
From Poker to Chess
The transition from subscription publishing to Open Access may be compared with a switch from the game of poker to the game of chess. The point is not that chess is morally superior to poker. The point is that the information structure of the game is different. In poker, important information (the hand held by each player) is hidden, and access to that information, by means of experienced guesswork, is a crucial part of the game. In chess, which centres by contrast on how the players can combine the potential of what is physically on the board in order to achieve victory, the pieces are visible on the board. Applying this metaphor to the world of academic publishing, if the playing field changes but the players continue to behave according to the rules of the previous game, misunderstandings are inevitable.
This, I suggest, is one reason why the academic publishing sector finds change so difficult. Open Access not only introduced a new payment mechanism; it transformed deeply-held expectations about access, price visibility, reuse, speed, transparency, and editorial concepts. Yet many of the institutions that judge journals, negotiate budgets and advise policymakers still act on assumptions that were formed in the older environment.
The Questionable Incentives Behind “Publish or Perish”
It is frequently observed that academics today are slaves to the mantra, “publish or perish”. What is the source of this pressure? It is easy to point the finger at publishers, but in fact the responsibility lies largely with the academic system itself. Publication lists, journal metrics, Impact Factor, CiteScore, Q1 and Q2 rankings and related indicators have become a form of academic currency. They influence careers, promotions, funding decisions, and the standing of entire research institutions.
As research systems expand globally, these incentives affect more and more researchers, especially those in emerging countries who are working in research institutions that all mimic the same “publish or perish” concept. And because – even with more research institutions in existence – it is not possible to regularly generate and publish more and more groundbreaking news, authors are forced to find new ways of meeting their performance goals. For instance, a research project that would once have generated a single, substantial article might be divided into two or three publishable units, because a longer publication list may be more useful to its authors within an institutional evaluation period. This approach is sometimes described as “salami slicing”. In this context, it should be noted that to be included in the Times Higher Education World University Rankings, universities are required to publish more than 1,000 relevant publications over the previous five years, and more than 100 relevant publications in any single year. The number of publications is critical here, but not their impact factor.
This well-known cascading process – initiated by highly ambitious journal editors and founded on the adherence to an undefined standard of quality – triggers an enormous amount of work for many additional journals, very often subscription-based ones. It creates additional stress for authors, who cannot escape the requirements of their institutions, and it also places an increasingly unnecessary burden on more and more reviewers.
These incentives also explain why authors increasingly expect rapid editorial handling. If a researcher is obliged to demonstrate output within a defined period, a publication lead time of twelve months is not simply inconvenient: it can negatively affect a career, a grant application, or the continuation of a research programme.
And yet there is a paradox. Journals reject enormous numbers of manuscripts, but a very large proportion of research eventually appears anyway. A manuscript rejected by one journal may be accepted by another, sometimes after several complete peer-review cycles, and sometimes even within the same publishing group.
Peer Review Should also Switch from Poker to Chess
Peer review is essential, but it should not be treated as a ritual whose value increases automatically with the number of revisions requested. Reviewers quite properly act in accordance with the respective journal’s aims and scope, and they may identify weaknesses that require revision. But a point can be reached whereby a reviewer is no longer evaluating the submited study but effectively requesting that its authors should conduct a different study.
In a fast-moving research environment, it is worth asking whether this is always the most useful response. If a paper cannot support a scientifically sound conclusion, the answer may be to reject that conclusion, not necessarily to demand months of additional experiments. In some cases, a preliminary observation or a limited finding may still deserve to enter the scientific record, provided that its results are correct and completely transparent. Different journals can legitimately arrive at different editorial decisions on such maters.
If an author either wishes or is obliged to publish an average article primarily to achieve an institutional performance target, his or her manuscript might enter the above-mentioned lengthy ‘cascading process’ from one journal to the next until it is eventually accepted.
The Legitimacy of Multiple Editorial Concepts
Academic publishing often acts as if journals had only one task: to disseminate the very best science. That makes sense in the case of a small group of highly selective journals. A journal that defines its role through exclusivity can legitimately ask whether a manuscript is exceptionally novel, timely or important for its field. Science benefits from journals that perform this function.
It makes little sense, however, for the entire journal market to imitate that model. It is axiomatic that not every journal can publish only the most novel and most important work within its given field. The very large mid-range section of the journal market needs a different editorial concept founded on the efficient and transparent publication of scientifically sound, relevant and useful research. This requirement is valid even if the research in question is not destined to redefine its domain.
Consider, for instance, the case of a technically sound paper addressing a question that has already been studied elsewhere. Its author, working in an under-funded research environment, may have had only limited access to the subscription literature and therefore an incomplete view of the totality of work published earlier on the subject. Paywalls not only restrict readership in purely quantitative terms; they also restrict what researchers on the outside are able to ascertain before they conduct their own research. Our new research paper may nevertheless include some important new aspect, such as a different study population, setting, method or observation. A highly selective journal might with reason reject it for its lack of genuine novelty. A journal with a soundness-based editorial concept might, by contrast, have fully justifiable grounds for publishing it.
Open Access often strengthens the value of the second model because the resulting published literature can be read globally, without subscription barriers. Research published by OA also becomes easier to use for systematic reviews, meta-analyses, text mining and, increasingly, AI-supported research tools. To return to the analogy with chess, all the information is on the board.
Different Business Models and Biased Comparisons
The Gold Open Access publishing model – whereby the peer-reviewed version of a research paper is made free for anyone to read online via the publisher’s website – frequently attracts the criticism that because the publisher earns revenue (via Article Processing Charges) on accepting the paper, it has an economic incentive to publish more articles. From a purely economic perspective, that statement is absolutely correct. The incentives at work within other academic publishing models should be viewed with the same degree of scepticism, however. Subscription publishers have incentives to protect subscription revenue, to bundle journals, to negotiate large agreements, and to preserve products for which institutions are willing to pay, owing to the absence of direct competition. Transformative agreements combine several of these mechanisms. Let us consider the costs involved.
In August 2026, Mathias et al. posted a preprint estimating global APC expenditure for 14 publishers between 2019 and 2025. Their estimate is USD 15.08 billion. Importantly, this is an estimate, not an audited global figure, and the paper itself is currently a preprint rather than a peer-reviewed publication. I cite the number without endorsing it as a definitive measurement.
This example is also useful for another reason. Open Access costs of this order of magnitude easily become the subject of public discussion because APC list prices are visible and can be multiplied by publication counts. It is much more difficult to make the corresponding calculation for subscription publishing. Institutional prices, bundles, discounts, transformative agreements and confidential or non-standard contract structures make it difficult to arrive at a like-for-like global estimate. If one nevertheless constructs a rough list-price scenario for comparable large publishers, the resulting amount can easily outstrip the USD 15 billion estimate relating to total APCs. A figure in excess of USD 50 billion might in fact be realistic.
This distinction is important. The purpose of the comparison is not to prove that one model is cheap and the other expensive: it is to ask why the costs associated with one side of the market are routinely discussed in terms of highly visible unit prices while the costs associated with the other side are much more difficult to reconstruct from publicly available information.
This is the poker-and-chess problem once more. A visible APC can be criticised precisely because it is visible. A negotiated subscription bundle may be much harder for an outsider to analyse. The transparency of the OA model ought not to be a competitive disadvantage – as it is at present.
Open Access: A Scientist-Led Movement
There is also a historical point to bear in mind here. The Open Access movement has often been spearheaded by scientists rather than publishers. Several of today’s leading Open Access organisations were initiated by scientists. Understanding the origins of OA publishing as a scientist-led movement conceived to bring about greater openness in scientific communication militates against the widespread perception that Gold Open Access was invented primarily as a commercial mechanism for collecting APCs.
Diamond Open Access – whereby research papers are published without any charges to authors or readers – provides another useful reality check. Removing charges for authors does not obviate the cost of publishing. Essential factors such as editorial management, technology, metadata, preservation, compliance, production, ethics, discoverability and long-term infrastructure still have to be financed. Diamond models do not mean that professional publishing becomes cost-free merely because the invoice is moved away from the author. Diamond OA is funded variously by institutional subsidies, library memberships, academic crowdfunding, grants, or even volunteer labour. It should be remembered, however, that this system goes in the direction of poker, and the difficulty of obtaining funding might bias editorial decisions.
The Case for a New Conception of Quality
“Quality” is one of the most frequently used and least precisely defined words in academic publishing. In general usage, quality can mean meeting measurable requirements, satisfying expectations, performing well relative to alternatives, or being fit for a particular purpose. These definitions apply equally to the sphere of scientific journals. Methodological soundness, editorial process, peer-review quality, transparency, reproducibility, accessibility, publication speed, indexing, citation impact, author service, and cost efficiency are all quality factors. Significantly, they are all very different from one another.
This matters because the word is often used as a form of mental shorthand: “This journal is high-quality”, “That publisher produces low quality”, or “A high rejection rate means high quality.” Such statements sound scientific, but they frequently fail to specify the criterion on which the judgement is based.
Let me therefore offer several deliberately unconventional examples of what quality might actually represent.
Imagine an indexed Gold Open Access journal that publishes 1,000 articles a year and has an Impact Factor of 3.0. This journal creates a countable new item on their publication list for 1,000 author groups. For many researchers around the world, that can help sustain a career, demonstrate research output, and support the continuation of a research program. This usefulness is also a dimension of quality.
Or again, a journal that can move a manuscript from submission to publication in five or six weeks, while maintaining a transparent and rigorous review process, may provide real value to early-career researchers working to demanding institutional deadlines. Speed here is a dimension of quality – the quality of service offered by the publisher. Speed is not a proof of the quality of the published paper, of course, but one could hardly argue that a very slow editorial process automatically generates better published output. The real question is, what is being reviewed, and to what purpose? It makes little sense for reviewers to apply the most stringent evaluation standards to this average piece of work. It would be much more effective to involve more statisticians in order to determine the accuracy of the data.
Conversely, a journal with a 90% rejection rate may be highly prestigious, but the rejection rate itself should not be considered an indicator of quality. If a sound paper passes through three or four complete review processes before eventually being published – perhaps even by another journal from the same publishing house – the system has consumed valuable reviewer time without necessarily improving the final scientific contribution. High selectivity may be justified in the case of an exclusive journal. It should not, however, become the editorial aspiration of the entire market.
I therefore see three broad categories of scientific journal. First, there is a small group of highly selective and exclusive journals whose legitimate role is to identify particularly novel, important or field-shaping work. Second, there is the enormous mid-range of the journal market, accounting for no less than 80% of the total. In my view, these journals should stop trying to behave as if they all belonged to the first group and should instead develop editorial concepts appropriate to sound, useful and efficiently communicated research. Third, there are those journals whose practices are so weak or misleading that a transparent preprint may be preferable to the appearance of formal journal certification.
The Value of Third-Party Indexing
This brings me to a point on which I take a deliberately firm position. Academic publishing has external organisations that evaluate journals for inclusion in major indexes. Their criteria are not perfect, and whether the indexing system itself is fair is a separate discussion. But the system exists, institutions use it, researchers organise careers around it, and journals invest considerable effort in satisfying its requirements.
If a journal has passed the selection process of a major recognised index, this is not automatic proof that every article it carries is excellent. However, it does establish the fact that the journal has crossed an externally defined threshold. An observer should therefore be careful about declaring a journal “low-quality” simply because it is fast and publishes many papers, or because it offers an editorial model different from a traditional selective subscription journal. Indexed journals are comparable and maintain a certain standard: this fact should be accepted without regard to the individual perspective or the publishing model.
This is exactly where the poker-and-chess distinction maters. Many large Gold Open Access publishers expose their workflows, timelines, fees and policies in ways that make them unusually easy to inspect and criticise. Greater visibility should lead to better scrutiny: it should not lead to a presumption of lower quality.
Barriers to Fair Competition
The most important economic argument of this essay is also the simplest. In the past, public and institutional budgets used to fund libraries to purchase journal subscriptions. By this means, selected journals were made available to the scholars affiliated with the libraries that purchased the relevant subscriptions. Subscription publishers therefore became natural and long-standing suppliers to libraries and academic consortia. As Gold Open Access expanded, APC expenditure naturally tended to develop outside those budgets, with researchers finding money in departments, faculties, grants, or separate institutional funds.
This means that two models serving the same basic function – communicating scholarly research – are often not allowed to compete for the same pool of money. Subscription and transformative agreements are negotiated within the parameters of established library structures, while pure Gold Open Access publishers may remain partly outside them, subject to marketplace exclusion. Indeed, it is not going too far to describe the benefits accorded to subscription publishers as a species of regulatory capture whereby institutions designed to serve the public interest in fact serve the needs of commercial players.
This state of affairs is patently inequitable, and it may well also prevent the sharing and evaluation of much important new research. Imagine, by way of a comparison, the buyer of a major supermarket chain being instructed that the company’s wine budget must be used exclusively for the purchase of French red wine because French wine has traditionally been considered the elite reference category. Excellent wines from Italy, Spain, Switerland or elsewhere may exist, but if they are to be bought in, their purchase will have to be financed from another pot, or else they must be excluded from consideration. In such a situation, the buyer cannot claim to have objectively ascertained the market price for quality wine, for the purchasing procedure has protected one category – French wine – from genuine competition.
Academic publishing should be treated in the same way. If subscription publishing, transformative agreements, pure Gold Open Access and other serious publishing models provide scholarly communication services, these should all be compared using a single and transparent procurement framework. Only then can their respective merits and demerits be determined; only then can meaningful competition exist between them; and only then can their respective claims on the public purse be objectively evaluated.
A Proposal for Libraries, Funders and National Decision-Makers
National funders and major institutional decision-makers should make eligibility for public publishing budgets conditional on a common comparison framework. Pure Gold Open Access agreements should be eligible for the same national or institutional publishing budgets as subscription and transformative agreements. Institutions and libraries would then receive a defined set of criteria by which competing offers should be assessed.
The comparison should go beyond the headline price. It could include many factors, all of them worthy of scrutiny: cost per published article, cost per researcher served, indexing status, cost per download, peer-review model, publication speed, waiver policy, transparency of editorial procedures, open peer review, licensing, machine readability, preservation, data and reproducibility policies, and the degree to which authors retain rights in their own work.
Importantly, these characteristics could be translated into explicit cost adjustments. A publisher that requires authors to transfer copyright could receive a cost surcharge in the procurement comparison. A publisher that requires additional permissions or fees for legitimate secondary uses could receive another surcharge. Conversely, author copyright retention, broad reuse rights, transparent peer review, recognised indexing and other features valued by the research system could improve the assessed value of an offer. The precise weighting should be decided transparently by funders, libraries and research institutions, not by publishers themselves.
This would not produce one mystical score called “quality”. It would produce something more useful: a transparent description of what the research system receives for its money.
The role of national decision-makers would therefore be to set the framework rather than to choose individual journals from certain publishers. Libraries and institutions would continue to make professional procurement decisions, but they would compare the same categories and would be required to include credible Gold Open Access offers in the market if they wished to receive funds from the national publishing budget.
The Pressures of Our Changing information Environment
The problem does not stem from a lack of imagination, professionalism or goodwill among individual actors within the academic publishing sector. Rather, it is the product of institutional inertia. Decision-making is distributed across ministries, funders, consortia, libraries, universities, faculties and individual researchers. Expertise is fragmented, responsibility is fragmented, and budgets are fragmented. Under these conditions, yesterday’s structures can survive long after the technology on which they were founded has changed.
Today’s academic publishing market deserves to be viewed through a different lens. The task is not to defend Open Access against subscriptions or subscriptions against Open Access. It is to ask whether the market structure still serves the research system that pays for it.
Conclusion: One Budget, One Market, Transparent Criteria
The debate about academic publishing is too often constructed in terms of oppositions: Open Access versus subscriptions, high rejection rates versus low rejection rates, prestige versus volume, quality versus quantity. These antitheses obscure the more practical question: what kind of publishing system best serves the broad research community in relation to the public and institutional money spent on it?
Three principles could provide a starting-point. First: one budget. Subscription, transformative and Gold Open Access expenditure should be visible within the same overall publishing budget. Second: one market. Serious publishing models should be allowed to compete for that budget rather than being protected by traditional funding structures. Third: transparent criteria. Price, access, rights, indexing, peer review, editorial services, speed, reuse, reproducibility and transparency should be compared explicitly.
Highly selective journals will continue to have a legitimate place in the united publishing market of the future. So will journals with a broader outlook that publish sound research at scale. The mistake is to judge both according to a single editorial concept while financing them through structures that preclude genuine competition. Including Gold Open Access publishers in the annual library budget does not automatically mean additional costs. Rather, it means incorporating efficient and fair Open Access publishers into the mix, as opposed to the few subscription publishers at the bottom of the library’s client list, which are often included more out of tradition than on merit.
The publishing world has moved from poker to chess. It is time for the institutions that finance and evaluate it to move the pieces accordingly.
Author Contributions
Conceptualization, P.R.; writing - original draft preparation, P.R.; writing - review and editing, P.R. The author has read and agreed to the submitted version of the manuscript.
Funding
This research received no external funding.
Data Availability Statement
No new data were created or analysed in this study. Data sharing is not applicable to this article.
Acknowledgments and Use of Generative AI
During the preparation of this manuscript, the author used OpenAI ChatGPT for language editing, restructuring, support in drafting revised passages, and assistance with reference checking. The tool was not used to determine the authorʹs arguments, opinions, or conclusions. All AI-assisted output was reviewed, revised, and verified by the author, who takes full responsibility for the content of this publication.
Conflicts of Interest
The author was formerly employed by MDPI AG as Head of Publishing. MDPI AG had no role in the conception or writing of this essay or in the decision to submit it for publication. The views and arguments expressed in this essay are those of the author and do not necessarily represent the views of MDPI AG.
References
1. Times Higher Education (THE) World University Rankings 2026 methodology. htps://www.timeshighereducation.com/sites/default/files/breaking_news_files/the_world_university_rankings_2026_methodology.pdf
2. Mathias L, Schares E, Alperin JP, Butler L-A, Kuang S, Schonfelder N, Haustein S. Estimating global article processing charges paid to 14 publishers for open access between 2019 and 2025. arXiv:2608.16322. 2026. Preprint.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
