Open access failed

Open Access Failed. But Not for the Reasons You Think

by

Table of Contents

Introduction: The Wrong Question

Has open access failed?

It is a tempting question, and increasingly, a popular one. Critics point to rising publishing costs, predatory journals, paper mills, and the growing frustration of researchers who feel that the system is no longer working in their favor. From a distance, the conclusion seems obvious. Open access promised a revolution in scholarly communication, and what we have instead looks like a messy, expensive, and uneven transition.

But this framing misses something important.

If success is measured purely by access, then open access has not failed at all. The volume of freely available research has reached unprecedented levels, and millions of articles that were once locked behind paywalls are now accessible to anyone with an internet connection . Governments have stepped in with mandates, institutions have built repositories, and publishers have rapidly expanded their open access portfolios. On paper, the mission has been accomplished.

And yet, something feels off.

The system is more accessible, but not more equitable. It is more productive, but not necessarily more trustworthy. Costs have not disappeared; they have simply moved. Power has not been redistributed; it has been reorganized.

So the problem is not that open access failed.

The problem is that it succeeded on the wrong terms.

Open access did not collapse under its own ideals. It was absorbed into the very system it was meant to disrupt, reshaped by economic incentives, institutional pressures, and a prestige-driven academic culture that proved far more resilient than expected.

To understand what went wrong, we need to stop asking whether open access failed, and start asking a better question:

What, exactly, did it become?

The Ideal That Started It All

The original vision of open access was not complicated. It was, in fact, remarkably clear.

In the early 2000s, a series of foundational declarations, most notably Budapest, Bethesda, and Berlin, articulated a simple but powerful idea. Knowledge, especially knowledge produced through publicly funded research, should be freely available to everyone. The internet, still relatively young at the time, was seen as the perfect vehicle to make this possible. It could eliminate distribution costs, bypass traditional gatekeepers, and create a truly global knowledge commons.

This was not just a technical proposal. It was a philosophical shift.

Open access was rooted in the belief that scientific knowledge is a public good, not a private commodity. It aimed to remove barriers not only for readers, but for societies, institutions, and entire regions that had historically been excluded from the global research ecosystem. In theory, it promised a more inclusive, more efficient, and more just system of knowledge dissemination.

And for a moment, it seemed achievable.

The convergence of digital infrastructure and scholarly norms created a rare opportunity. Researchers already produced and reviewed content largely without direct financial compensation. If distribution could be made nearly costless, then perhaps the entire publishing model could be reimagined.

But beneath this clarity of purpose was a quieter problem.

The movement lacked strategic focus.

Open access was expected to do too many things at once. It was supposed to reduce costs, challenge commercial publishers, increase global participation, accelerate innovation, and reform academic culture. These goals were not necessarily incompatible, but they required coordinated structural change across multiple layers of the system.

That coordination never fully materialized.

Instead, the transition unfolded in a fragmented way, shaped by funders, institutions, publishers, and researchers, each operating with their own priorities. This lack of alignment created an opening, and the most powerful actors in the system moved quickly to fill it.

Open access was not defeated.

It was reinterpreted.

The Economic Capture of Open Access

If there is a single moment where the trajectory of open access shifted, it is here.

The transition from subscription-based publishing to open access did not eliminate the underlying business model. It replaced it. Instead of charging readers or institutions for access, publishers began charging authors for publication, primarily through Article Processing Charges, or APCs.

At first glance, this seemed like a reasonable trade-off. Remove the paywall for readers, and shift the cost to those producing the research, often supported by grants or institutional funding. In theory, this would align incentives and broaden access.

In practice, it created a far more scalable system for revenue generation.

Under the subscription model, revenue was tied to access. Libraries paid for access to a fixed set of journals, and growth was constrained by budgets and subscription limits. Under the APC model, revenue is tied to output. The more articles a publisher accepts, the more it earns. This seemingly small shift has enormous implications.

It transforms publishing into a volume-driven business.

Large commercial publishers adapted quickly. Having already consolidated significant portions of the market over decades, they were uniquely positioned to scale open access at speed. Today, a handful of major players still control a substantial share of the global publishing landscape, with some maintaining profit margins that rival or exceed those of leading technology companies.

Open access did not disrupt their dominance.

It gave them a new growth model.

This is where the logic of the system begins to change. When revenue depends on the number of articles published rather than the number of subscriptions sold, the incentive structure shifts. Selectivity becomes a cost, not a benefit. Speed becomes a competitive advantage. Volume becomes the primary driver of success.

The result is what can only be described as a “volumes game,” where publishing more is not just desirable, but economically rational.

And once the system starts optimizing for volume, everything downstream begins to adjust.

Editorial models evolve. Peer review becomes more strained. New journal formats emerge, including mega-journals that prioritize technical soundness over perceived importance. None of these developments are inherently problematic, but together they signal a fundamental shift in what publishing is designed to do.

Open access did not remove the commercial logic of publishing.

It refined it.

The Production Paywall

One of the most persistent myths about open access is that it removed barriers.

In reality, it relocated them.

The traditional paywall restricted access to readers. If you were not affiliated with a well-funded institution, your ability to read the latest research was limited. Open access addressed this problem effectively. Today, a growing proportion of the scholarly literature is freely available to anyone with internet access.

But the cost did not disappear.

It moved upstream.

The APC model introduced a new kind of barrier, one that operates not at the point of consumption, but at the point of participation. Instead of paying to read, researchers now pay to publish. And these costs are not trivial. Article processing charges can range from a few thousand dollars to well over ten thousand for high-prestige journals.

This creates what can be described as a production paywall.

On the surface, the system appears more open. Anyone can read. But not everyone can contribute. The ability to publish, especially in high-impact venues, increasingly depends on access to funding. Researchers in well-resourced institutions can navigate this system with relative ease. Those in underfunded environments cannot.

The implications are profound.

Open access has shifted inequality from access to participation. It has made knowledge more visible, but not necessarily more inclusive. In some cases, it has even deepened existing disparities, particularly for researchers in middle-income countries who fall outside formal waiver systems but still lack sufficient funding to cover APCs.

This is not just an economic issue.

It is a structural one.

When the ability to publish is tied to financial capacity, the system begins to filter not just what is read, but what is written. Entire perspectives risk being underrepresented, not because they lack merit, but because they lack funding.

Open access solved the problem of who can read the literature.

It created a new problem around who gets to produce it.

The Prestige Trap That Open Access Could Not Escape

If the economics explain how open access was reshaped, the prestige system explains why it stayed that way.

Academic publishing is not just a system for disseminating knowledge. It is a system for evaluating careers. Researchers do not publish simply to share findings. They publish to secure jobs, promotions, funding, and recognition. And for decades, those outcomes have been tied to a relatively narrow set of signals, most notably journal prestige and impact metrics.

Open access never replaced this system.

It tried to work around it.

This is where the tension becomes unavoidable. On one hand, open access encourages researchers to prioritize accessibility and openness. On the other, the academic reward structure continues to prioritize where you publish over how accessible your work is. High-impact journals, many of which were historically subscription-based, still dominate hiring and promotion decisions. Even as these journals adopt open access options, often through hybrid models, their prestige remains the primary draw.

This creates a predictable outcome.

Researchers optimize for prestige, not openness.

Hybrid journals thrive in this environment because they offer the best of both worlds. Authors can publish in high-prestige venues while paying to make their individual articles open. From the publisher’s perspective, this is an ideal arrangement. It preserves subscription revenue while adding APC income, a practice often criticized as “double-dipping.” From the researcher’s perspective, it is a rational compromise. From the system’s perspective, it is a deadlock.

Open access did not fail to gain adoption.

It failed to change behavior.

As long as career advancement is tied to prestige metrics, researchers will continue to make decisions that reinforce the existing hierarchy. Open access becomes a layer on top of the system, not a force that reshapes it.

The result is a kind of structural inertia.

The system moves, but it does not change direction.

The Global Inequality Problem Open Access Made Worse

One of the most compelling arguments for open access was its promise to level the global playing field. By removing paywalls, it would allow researchers, students, and institutions in lower-income regions to access the same body of knowledge as their counterparts in wealthier countries.

In terms of access, this promise has largely been fulfilled.

In terms of participation, it has not.

The introduction of APCs has created a new layer of inequality that disproportionately affects researchers in the Global South. For many, the cost of publishing in reputable open access journals can exceed several months of salary or even an entire research budget. While waiver programs exist, they are often inconsistent, limited to specific countries, or administratively burdensome. Researchers in middle-income nations are particularly disadvantaged, as they may not qualify for waivers but still lack sufficient funding.

This creates a structural imbalance.

Researchers in well-funded systems can publish widely and visibly. Researchers in underfunded systems face barriers not only to publishing, but to publishing in venues that carry international recognition. Over time, this affects more than just output. It shapes whose work is cited, whose ideas circulate, and whose voices are considered authoritative.

This is where the concept of “testimonial injustice” becomes relevant. It refers to a situation where certain groups are systematically undervalued as sources of knowledge. In the context of open access, this is not simply about access to journals. It is about access to legitimacy.

The problem is compounded by structural factors beyond APCs. Editorial boards of major journals remain heavily concentrated in the Global North. Indexing systems favor journals with established reputations, which are often based in wealthier regions. Language barriers further limit visibility for research published outside the dominant English-language ecosystem.

Open access expanded the audience for research.

It did not equally expand the platform.

The result is a system that appears globally inclusive on the surface, but remains uneven in terms of who gets to shape the conversation.

The Integrity Crisis Beneath the Surface

While debates about cost and access dominate discussions of open access, a more serious issue has been quietly intensifying.

The integrity of the scholarly record is under strain.

The shift to an APC-driven model introduced a new set of incentives into the system. When revenue is tied to the number of articles published, there is a built-in pressure to increase throughput. Most reputable publishers maintain rigorous editorial standards, but the broader ecosystem has become more vulnerable to exploitation.

Predatory journals are the most visible manifestation of this problem. These outlets present themselves as legitimate academic journals but provide little or no peer review, charging authors significant fees for rapid publication. They exploit the pressure on researchers to publish, particularly those who lack the resources or institutional support to navigate more selective venues.

More concerning, however, is the rise of industrialized fraud.

Paper mills, organizations that produce fabricated or manipulated research for a fee, have become increasingly sophisticated. They offer authorship slots, generate data, and even coordinate submissions to journals. Estimates suggest that hundreds of thousands of fraudulent papers may have entered the scholarly record over the past two decades . Retractions have surged, with some publishers forced to withdraw thousands of articles after uncovering systematic manipulation.

This is not a marginal issue.

It is systemic.

The combination of publish-or-perish culture, financial incentives tied to publication volume, and the availability of AI tools has created an environment where low-quality or fraudulent research can proliferate faster than it can be detected. Even legitimate journals are not immune. As submission volumes increase, peer review systems become strained, and the ability to thoroughly vet each manuscript diminishes.

Open access did not create scientific fraud.

But it changed the conditions under which fraud operates.

It made the system faster, more scalable, and, in some cases, more porous.

The consequence is a growing tension between accessibility and trust. Making research widely available is valuable, but only if the research itself is reliable. As concerns about reproducibility and data integrity continue to mount, the credibility of the entire system comes into question.

The challenge is no longer just about who can access knowledge.

It is about whether that knowledge can be trusted.

Policy Successes That Didn’t Fix the Core Problem

In response to many of these challenges, policymakers have stepped in with increasing force. Governments and funding agencies have introduced mandates requiring that publicly funded research be made openly available, often with strict timelines and technical requirements.

One of the most significant examples is the push for zero embargo policies, which require immediate access to research outputs upon publication. These policies also emphasize machine readability, metadata standards, and transparency in authorship and funding. In many ways, they represent a maturation of the open access movement, shifting from voluntary adoption to enforced compliance.

These are meaningful developments.

They have accelerated access, standardized practices, and signaled a clear commitment to openness at the highest levels of governance.

But they have not addressed the core structural issues.

Policies can mandate that research be open. They cannot determine how the system funds itself, how prestige is assigned, or how incentives are distributed. In practice, many of these mandates rely on the existing publishing infrastructure to deliver compliance. This means that the same large publishers, operating under the same economic logic, remain central to the system.

The result is a paradox.

Open access is increasingly mandatory, but the conditions under which it operates remain largely unchanged.

Researchers comply with mandates. Institutions negotiate agreements. Publishers adapt their models. The system evolves, but its underlying dynamics persist.

Policy has been effective at expanding access.

It has been far less effective at redistributing power.

Transformative Agreements and the Illusion of Progress

If policy represents the push toward open access, transformative agreements represent the mechanism through which that push is being operationalized.

At least, that is the idea.

These agreements, often framed as “read and publish” or “publish and read” deals, are designed to shift institutional spending away from subscriptions and toward open access publishing. In theory, they are a bridge from the old system to a new one. Libraries repurpose their budgets, publishers transition their models, and researchers gain the ability to publish openly without direct financial burden.

It sounds like progress.

In practice, it is something else.

Transformative agreements do not dismantle the existing system. They renegotiate it. Instead of paying for access to journals, institutions pay for the right to publish within them. The financial relationship remains intact, and often, becomes even more complex. Libraries must now manage voucher systems, track author eligibility, monitor publishing quotas, and navigate a patchwork of non-standard agreements.

Costs, meanwhile, remain stubbornly high.

For research-intensive institutions, the total expenditure can even increase as publication output grows. Many agreements include caps on the number of articles covered. Once those caps are reached, authors are once again exposed to APCs, often without clear guidance on what happens next. The result is confusion, administrative burden, and financial unpredictability.

More importantly, these agreements reinforce the position of the largest publishers.

By tying open access publishing to existing journal portfolios, transformative agreements make it harder for alternative models to gain traction. Institutions become locked into multi-year deals with established players, limiting their ability to experiment with new platforms or support community-led initiatives.

The transition, in other words, is happening within the boundaries of the old system.

Transformative agreements are not a pathway out.

They are a more expensive way of staying in.

The Alternatives That Might Actually Work

For all the criticism directed at open access, it would be a mistake to assume that viable alternatives do not exist.

They do.

They are simply not the ones dominating the current landscape.

Diamond open access offers perhaps the clearest contrast to the APC-driven model. In this system, neither authors nor readers pay. Publishing is supported through institutional funding, consortia, scholarly societies, or public infrastructure. The goal is not to monetize individual transactions, but to treat publishing as a shared academic service.

In principle, this aligns far more closely with the original vision of open access.

In practice, it faces serious constraints.

Many diamond journals operate on extremely limited budgets, often relying on volunteer labor and minimal technical infrastructure. While there are thousands of such journals globally, their visibility remains low. They are underrepresented in major indexing systems, and their lack of prestige makes them less attractive to researchers navigating career pressures.

Overlay journals push the model even further.

Rather than hosting content themselves, they sit on top of repositories and preprint servers, providing peer review and curation without the overhead of traditional publishing infrastructure. This dramatically reduces costs and increases transparency. It also shifts the role of the journal from gatekeeper to evaluator.

Conceptually, this is a powerful shift.

But again, adoption is limited.

The problem is not that these models are flawed.

The problem is that the system does not reward their use.

Researchers are unlikely to prioritize platforms that do not advance their careers. Institutions are hesitant to invest in models that lack clear metrics of impact. Funders continue to operate within frameworks that emphasize compliance over systemic redesign.

The alternatives exist.

They are simply outcompeted by the current incentive structure.

The Real Failure: Misaligned Incentives

At this point, the pattern becomes clear.

Open access did not fail because it lacked vision.

It did not fail because it lacked technology.

It did not fail because researchers resisted it.

It failed because it tried to change outcomes without changing incentives.

Every major stakeholder in the system operates according to a different set of priorities:

  • Researchers are driven by career advancement, which depends on prestige, impact factors, and institutional recognition
  • Publishers are driven by revenue, which in the open access era is increasingly tied to publication volume
  • Institutions are driven by rankings, output metrics, and budget constraints
  • Funders are driven by compliance, accountability, and policy objectives

These incentives are not aligned.

They intersect, overlap, and occasionally reinforce each other, but they do not point in the same direction. Open access was introduced into this environment as a solution to a specific problem, access to research, but it did not fundamentally alter the logic governing how the system operates.

So the system adapted.

Publishers incorporated open access into their business models. Researchers complied with mandates while continuing to prioritize prestige. Institutions negotiated agreements that preserved access while managing costs. Funders expanded requirements without restructuring evaluation criteria.

Everyone moved.

Nothing really changed.

This is the core failure.

Not of open access as an idea, but of its implementation within a system that rewards entirely different behaviors.

You cannot build an equitable publishing ecosystem on top of incentives that reward exclusivity, volume, and prestige.

Until those incentives change, every reform will be absorbed, adjusted, and ultimately neutralized.

Conclusion: A Failure of Implementation, Not Vision

So, has open access failed?

Only if you misunderstand what it set out to do.

If the goal was to make research more accessible, then open access has succeeded, decisively and irreversibly. The paywalls that once defined scholarly communication have weakened. Knowledge flows more freely than at any point in history. Policies continue to push in that direction, and there is no realistic path back to a fully closed system.

But if the goal was to create a more equitable, sustainable, and trustworthy publishing ecosystem, the results are far more mixed.

The system is more open, but not more balanced.

It is more productive, but not necessarily more reliable.

It is more accessible, but still deeply unequal.

Open access did not dismantle the structures of academic publishing.

It adapted to them.

The next phase of this transformation will not be defined by more mandates, more APCs, or more agreements. It will be defined by whether the academic community is willing to confront the deeper issue at the heart of the system:

What do we actually reward?

Until that question is answered differently, the future of open access will look a lot like its present. Expanding, evolving, and yet constrained by the same forces that shaped it from the beginning.

Open access did not fail.

It became something else.

And whether that “something else” can still be reshaped is the question that matters now.

Leave a comment