Go back

Why hybrid journals do not lead to full and immediate Open Access

29/04/2021

In this brief note, we formulate 6 arguments that articulate why cOAlition S Organisations will not financially support the hybrid model of publishing. We define a hybrid Open Access journal as a subscription journal in which some of the original research papers are Open Access while others are only accessible via payment or subscription. A journal that publishes both original research papers and other professionally curated content (i.e. a magazine/news part) is not considered a hybrid journal under this definition.

Argument 1: Hybrid has not facilitated a transition to Open Access (OA)

Hybrid journals were initially conceived to allow publishers to offer an OA option without giving up their subscription business model (Prosser 2003). One critical assumption in the initial concept was that subscription and OA income would recover the cost of articles published in the subscription and OA mode, respectively. Publishers could then adapt to a changing publishing landscape by gradually replacing subscription income with income from OA fees as needed.

20 years on, the hybrid model has clearly failed to deliver on this promise. The percentage of OA articles in hybrid journals is not increasing fast enough to ensure full transition in a reasonable timeframe: a 2019 study by DeltaThink shows that the share of hybrid OA peaked around 2016, well before Plan S policy kicked in. Data[1] collected by Laakso suggests that just 69 subscription journals published by Elsevier and 23 subscription journals published by Wiley had transitioned from toll access to open access in 2016.  As these two publishers alone publish over 4000 journals, a full transition to OA via the hybrid model would take a very long time.

Hybrid publishing sustains the status quo: it fails to make a deliberate effort to transition to full Open Access, while enjoying an additional stream of revenue. Publishers do not ensure that OA income replaces subscription income for open access articles. Publishers now market hybrid journals to authors as providing desirable “author choice”. This shows that there is no intention to move the hybrid publishing model to full OA. A recent preprint by Piwowar et al. (2019) predicts the percentage estimates that by 2025: 44% of all journal articles will be available as OA and 70% of article views will be to OA articles. While this does represent growth, it shows that it could take a decade or more to reach full OA.

Publishers will argue that they cannot steer the uptake of OA, which is driven by the authors’ willingness to select the OA option and/or funders’ policies to mandate OA and pay for it.  But this ignores the fact that universities’ and funders’ budgets are limited and that the open access fees in hybrid journals come on top of the expenses for subscriptions. Thus, what limits the uptake of OA in hybrids is not researchers’ or funders’ lack of willingness to select OA, but the additional financial burden that this imposes on them.

Indeed, most hybrid publishers charge open access fees in addition to subscription fees, and do not ensure that open access income replaces subscription income for open access articles.  As a result, hybrid journals depend disproportionally on subscription income which subsidizes the open access option. Hybrid journals are thus no different from subscription journals in their obstruction to a full transition to OA because such a transition would undermine the current subscription business model of academic publishing.  Therefore, hybrid is a tactic that sustains the (subscription) status quo, allowing publishers to enjoy an additional stream of revenue without forcing them to make a sustained effort towards the implementation of a fully OA business model.

 

Argument 2: The research community pays twice (double dipping)

Hybrid OA publication fees are charged in addition to subscription fees, which results in publishers being paid twice for the same content.  Although subscription publishers argue that the two revenue sources are kept separate, and that no double-dipping occurs, Pinfield et al. (2016) shows otherwise.  Using data from the UK, Pinfield states that APC payments are now an important part of the relationship between universities and publishers, and are “adding to the total cost of publication markedly”. The hybrid model provides no transparent checks on double-dipping other than the lofty assurances of the publishers. If publishers really wanted to demonstrate that no double dipping occurs, they would need to be fully transparent on their pricing and revenue streams and be ready to open their books for an independent audit.

Often publishers will not make articles fully open access until APC payment has been received. This has created situations where an article is paywalled for several months until payment is finalized. In the meantime, those articles are only available to subscribers or those who pay individually. This practice blurs the claimed separation between the two revenue streams, and provides an opportunity for publishers to make money off articles intended to be OA. It also represents an unnecessary delay in access.

 

Argument 3: Hybrid journals are more expensive than fully OA journals

The average APC levied by hybrid journals is on average higher than that charged in fully open access journals. In addition, expenditure on APCs for hybrid journals has risen faster than expenditure on subscription fees. Libraries and other organisations are paying OA fees in addition to what they were already paying for subscriptions and total costs have risen, with hybrid the main contributing factor [UUK Fig 4.3]. Continued financial support for hybrid journals would reinforce this unsustainable model.

Data from Jisc shows that in 2014-15 the average APC for a hybrid journal was £1,882, as opposed to £1,354 for a full OA journal (some 39% more expensive). More recent data from OpenAPC show that the average APC in hybrid journals is £2,770 compared to £1,768 for full OA journals. A group of librarians recently made a study of APCs paid by the Bill and Melinda Gates Foundation and found that APCs paid to hybrid journals were significantly higher than APCs paid to OA-only journals. A 2014 study on hybrid APC data reaches similar conclusions. Publishers will argue that this is because of the perceived prestige of those journals, but the academic and funding community should be paying for actual services rendered, not the perceived prestige that is more typically associated with luxury products.

In this context, the “opportunity cost argument” arises for cOAlition S organisations. Research funding and performing organisations only have limited funds available. They must therefore make careful choices on how to best spend their money. Spending money on open access publication means that less funds are available for research proper. cOAlition S organisations have jointly concluded that they should not spend their scarce funds to support a failed and needlessly expensive publication model.

cOAlition S organisations have jointly concluded that they should not spend their scarce funds to support hybrid journals, a failed and needlessly expensive publication model. Share on X

 

Argument 4: Hybrid journals provide a poor quality of service

Since 2014, Wellcome has published an annual report detailing its OA spend and highlighting any issues and concerns.  Every single report over the last 5 years has expressed concerns about the quality of service provided by hybrid publishers in terms of making articles freely available from the Europe PMC repository with a CC BY licence.

Looking at the most recent data from 2018-19, Wellcome reported that some 184 articles – for a total APC cost of £417k – had either not been deposited in Europe PMC or had been assigned a licence other than CC BY.  163 (89%) of the “problem articles” were published in hybrid journals.

In an OpenAire report from 2018, Najla Rettberg argues the same point: “But due to a lack of discoverability, hybrid OA might not offer authors the best value after all. Discovery and link resolution services traditionally work on the journal level, which means that hybrid OA articles might not be indexed by search engines or picked up by library electronic resource management and discovery systems, but remain hidden behind a paywall. In addition, there have been repeated cases where hybrid OA articles have been incorrectly licensed, sold as closed access content, and their reuse has been restricted. See Paywall Watch for recent examples”. As also argued by Ross Mounce here, hybrid OA is unreliable.

 

Argument 5: Hybrid journals crowd out new, full OA publishing models

As data on hybrid publishing becomes available we see that most of the money paid for hybrid OA publications goes to the Big 5 publishers. That money is therefore locked up in a dead-end for OA, and leaves less funds available to support publishing models that do provide full and immediate OA. Hybrid publishing also holds traditional publishers back from adopting alternative models.

Thus, the hybrid model is now acting as a major obstacle against innovation in the scholar publishing industry. It represents one of the main barriers to entry for competitors who could provide more innovative and full OA solutions. Why would a publisher want to take the riskier route of offering full OA outlets, when the hybrid option is a much safer and more profitable model?  Why would a new publisher be incentivized to enter the market with innovative OA publishing models if incumbent competitors can continue to run hybrid journals at low risk?  With the invention of the hybrid model, legacy publishers have found an astute way to shuffle off the entire business risk of transitioning to OA to the scholarly community.

Paying for OA fees in hybrid journals will therefore make publishers complacent. They will hold on to the lucrative hybrid model, and postpone the implementation of new OA business models that the scholarly community desperately needs.

The hybrid model is now acting as a major obstacle against innovation in the scholar publishing industry. It represents one of the main barriers to entry for competitors who could provide more innovative and full #OpenAccess… Share on X

 

Argument 6: Reader access: a hybrid journal is a “random OA” journal

To a reader, a hybrid journal appears as a collection of research papers in which only a random subset of papers is available OA. The reader is unable to predict which papers are available OA and which are not. Thus, hybrid journals are really random OA journals: they contain a haphazard mix of OA and paywalled papers. It is impossible for the reader to predict which papers will be OA. This means that any reader who wants to make proper use of a hybrid journal is obliged to have a subscription to it, so they can have access to all papers in the journal.  And it is difficult for libraries to surface this content as their systems manage content at the title, not the individual article, level.

These different levels of access to a hybrid journal often impede proper scholarly discourse. There are documented cases where an article in OA received a paywalled reply, while the resulting author response was again published OA. Clearly, for the readers, a hybrid journal is not OA in any useful way; they remain subscription journals for all intents and purposes. The hybrid model only benefits publishers: they enjoy both the subscription income and the extra layer of APC income paid by authors, institutions or funders who care about OA. The latest iteration of hybrid, like Elsevier’s ‘mirror’ journals (now also sometimes denoted as “companion” journals) – which carry the same title, editorial board as the subscription journal – are just hybrid by another route.

Publishers will argue that the individual paper, rather than the journal, is the relevant unit.  Yet, they continue to sell subscriptions to journals or bundles of journals rather than individual articles. The hybrid model allows publishers to sell the same product twice: the journal as a relevant unit to readers, and the individual paper as a relevant unit to authors.

 

Conclusion

The arguments developed here provide the rationale for why cOAlition S Organisations have decided not to financially support hybrid journals unless these journals adopt a transformative arrangement that will lead them towards full and immediate OA. Funding by cOAlition S members for these transformative arrangements is finite and will cease at the end of December 2024. In the meantime, these arrangements must ensure that OA publication fees are properly offset against subscription fees, so as to avoid any double-dipping. Publishers will only “bite the bullet” and start exploring new OA publishing models when they realize that the hybrid model is no longer a viable option.

 

References


[1] Full dataset available on request from Mikael Laakso, Hanken School of Economics, Helsinki, Finland.