Reflections on OA/TA coexistence
SPARC Open Access Newsletter, issue #83
March 2, 2005
by Peter Suber
Open access (OA) and toll access (TA) have coexisted for as long as there has been OA.  So the question is not whether they *can* coexist, but whether they will coexist forever or only for some transition period. 

I don't know the answer and won't offer a prediction.  One reason why prediction is difficult is that some of the variables suggest long-term coexistence and some don't.  Another is that this is nothing like predicting a force of nature.  We're talking about the actions of interested human beings, including ourselves.  Regardless of what we expect to happen, we're all busy making the future we prefer.

I don't even pretend to know all the variables.  But here are some of them along with my current take on which way they cut.  I expect to return to this topic in the future as I continue to think about it.

* There's an obvious kind of compatibility between OA and TA.  Among journals, we see it in the fact that two journals, even in the same research niche, do not directly compete with one another for readers, even if one is expensive and the other is free.  Since they publish different papers, researchers will have reasons to consult both.  If they are both TA, then subscribers will have reasons to buy both.  This is not a conspiracy; it's just a consequence of the fact that journals are not fungible like tanks of water or lumps of coal.  Individual journals do not directly compete with one another for readers, and TA journals as a class do not directly compete with OA journals as a class.

In archiving, we see a similar kind of compatibility in the fact that authors can publish in almost any TA journal and still provide OA to their articles by depositing them in OA repositories.

But these kinds of compatibility do not necessarily create long-term coexistence.  Their present compatibility might be a snapshot of two tectonic plates exerting complicated forces on one another.  In the rest of this piece I'll be looking for variables that affect long-term OA/TA coexistence, not just their near-term compatibility.

First, however, let me pause to show some of the complexity even in the simple kinds of compatibility I just mentioned.  For example, journals in the same niche compete for submissions even if they don't compete for readers.  Moreover, there are indirect ways in which they compete for readers even if they publish different papers.  For example, some readers look in the most convenient sources first and disregard the rest.  Even if this is bad research practice, it gives free resources a competitive edge.  (My position has always been that this is bad research practice, but that the solution is to make more good work OA, not to scold researchers for going to the most convenient OA sources first.)  In addition, there is clearly a tipping point, even if we haven't reached it yet, after which libraries will cancel high-priced TA journals because their niche is adequately served by high-quality OA journals.

If most journals were OA, that might create difficult pressures on the remaining TA journals, if only by changing author and reader expectations.  Or it might make life easier for the remaining TA journals, by freeing up money to pay for them.  Likewise, if most authors deposited their articles in OA repositories, that might create difficulties for TA journals by threatening their subscription rolls.  Or it might make life easier by increasing their visibility, their retrievability, and their citations.  Or it might to both in a way whose net effect is shifting or difficult to discern.

* Coexistence reduces the efficiency of both OA and TA models.  Hence, each side has reasons to keep trying to push the other off-stage. 

Coexistence reduces the efficiency of OA journals by making universities, libraries, foundations, and governments continue to pay for subscriptions, when that money could be spent on the OA alternative and neglected needs like books.  I don't see any sense in which coexistence reduces the efficiency of OA archiving.

Coexistence reduces the efficiency of TA by introducing OA archives and OA journals to compete with TA journals.  Either this cuts into their subscriptions, or will do so when the quantity and quality of OA reach a critical threshold.

* As long as OA and TA coexist, the paying institutions (universities, funders, governments) will have to pay for both, and the total cost of both seems to be higher than the total cost of either one alone.

I've called this the double-payment problem (in FOSN for 1/1/02) and argued that it is a burden on the transition to OA that doesn't affect the long-term economic viability of OA.  Even if that analysis is correct, what happens if we are doomed to live in the "transition" or "coexistence" period forever?

By itself, this cost doesn't make coexistence more likely or less likely.  But it gives the paying institutions a reason to make coexistence temporary or tolerable.  One way that libraries might act under this pressure is to lend their assistance to the OA movement, and we see a lot of this already.  Libraries don't have to work for the extinction of TA, merely for extending the scope of OA until the savings it brings offset the extra costs of coexistence. 

* If OA journal literature is less expensive to produce than TA journal literature, then we should expect its share of the total journal literature to grow steadily over time.  Certainly sticker-shocked universities and libraries will continue to explore the possibility of producing their own OA journals.  But even TA publishers may be attracted to business models that can produce literature of equal quality at lower cost.  If the prospect of lower profit margins is deterring them, then annual cancellation rates of 5-10% will eventually answer that objection. 

If the uncertainty of the business models is deterring them, then emerging empirical data from many journals in many fields will be relevant.  Insofar as the data show that some OA business models work in some disciplines, the objection will be answered to that extent.  Insofar as the data show that some OA business models are unsustainable in some disciplines, then the growth of that kind of OA on that path will have a natural limit and coexistence will become more likely at least until new OA business models evolve for those fields.

The production costs of OA and TA journals vary from journal to journal, and averages may not matter much when we're wondering about the long-term prospects for models that might be very far from the average today.  But it seems to be a safe generalization that OA journals have or can have lower expenses than TA journals:  they needn't pay for subscription management (to solicit, track, and renew subscribers), DRM (to authenticate users and block access to the unauthorized), or licensing (to draft, negotiate, or enforce licenses).  They needn't pay for print, although they have this in common with electronic-only TA journals.  And they needn't pay for marketing, even if some choose to do so.  They needn't generate profits or surplus or, if they do, they needn't produce the same high margins we see in the commercial publishers today.

Finally, of course, even if OA and TA journals have the same production costs, OA archives have much lower costs than any journal and their use is compatible with any kind of journal production system.  That suggests that OA archiving will continue to grow.  Whatever the slope of the upward curve for future OA journal market share, the slope of the curve for future OA archiving is likely to be steeper.  While this is a signal victory for OA, it is compatible with TA coexistence --though we do not know for how long.

* OA and TA have coexisted in physics since the launch of arXiv in 1991.  This isn't just a little OA coexisting with a lot of TA.  OA archiving is the default in physics, and yet TA journals in the field are not only surviving but thriving.

For details on the experience in physics, see Alma Swan's 2/3/05 posting to SOAF, which includes responses from publishers to the "competition" from arXiv.

The question is whether the experience of physics will transfer to other disciplines.  We don't know yet and we won't know until the rate of archiving in other fields approaches the rate in physics.

* What if OA actually helps TA?  For example, the American Society for Cell Biology reports that subscriptions and submissions to _Molecular Biology of the Cell_ both increased after it adopted the policy to provide OA to its articles with just a two-month delay.

OA increases a journal's visibility, retrievability, and citation impact.  This can certainly increase a journal's submissions, and if it is TA, then also its subscriptions.  This is clearly part of the explanation for the MBC experience, as well as the experience in physics.  But what does it leave out? 

If OA helps TA, at least in some sectors, then that would tend to promote coexistence.  Fears that OA would undermine TA would be exposed as groundless, at least for those sectors.  Hindsight would reveal that the real problem was not economic ruin but panic and resistance.  (Remember the movie industry's response to the VCR.)

* Two years ago, the French financial analysts at B.N.P. Paribas predicted that in 10 years, OA and TA would continue to coexist --and that today's big TA publishers would be tomorrow's big OA publishers.  The current TA publishers may be driven to convert business models on some of their publications, but they won't be driven out of business.

If true, this would answer one fear of some publishers.  If they are more concerned about their own economic survival as enterprises than about the survival of the subscription model for funding journals, then the Paribas prediction suggests that the players can survive even if the models change.

* Some anti-OA arguments call not for actual coexistence but freedom to coexist.  For example, Principle #7 of the DC Principles says, "we believe that a free society allows for the co-existence of many publishing models...."

Insofar as this is a coded objection to OA, then it simply mistakes the argument that OA is superior for the argument that TA should be banned.  The former is a serious argument that critics ought to face; the latter is a red herring.  I don't know any serious OA proponent who has argued that TA should be banned.  We agree that a free society will permit coexistence and we agree that we ought to live in a free society.

The right policy question is not whether OA and TA should both be permitted by law, but which better serves the interests of science and scholarship.

(Just to forestall objections:  The call for mandated OA to publicly-funded research is obviously not equivalent to a call for banning TA.)

* The TA journals most likely to survive in a world of inadequate library budgets and growing OA are the high-prestige TA journals.  If the high-prestige journals decide to continue to charge subscriptions, they are much more likely to continue to find buyers than lower-prestige journals.  So if the population of TA journals dwindles over time, the high-prestige journals will be the last hold-outs. 

However, some OA journals are already prestigious and others are growing in prestige.  An OA journal has no intrinsic prestige handicap just because it is OA --or if it does (or did), this is a prejudice that is rapidly vanishing.  However, most OA journals are new.  And while new journals can be excellent from birth, it takes time for a journal's prestige to catch up with its quality.  Now here's the key:  it's only a matter of time before the prestige of excellent OA journals does catch up with their quality.  At the same time, as OA spreads, it will easier to recruit eminent scholars to serve on OA journal editorial boards.  In addition, we'll see more and more already-prestigious TA journals convert to OA, taking their reputations with them.  These are three reasons to think that OA journals will continue to rise in prestige as time passes. 

For authors, the only reason to submit work to a TA journal is its prestige.  In every other way, TA journals are inferior to OA journals because they limit an author's audience and impact.  OA journals will start to draw submissions away from top TA journals as soon as they approach them in prestige.  And by the time they equal them in prestige, the best TA journals will have lost their one remaining competitive advantage.  As authors lose their incentive to submit work, subscribers will lose their incentive to subscribe.  This suggests that coexistence will be temporary.

* Any TA journal that adds enough value to the basic text should be able to find buyers willing to pay for the added value.  That suggests that coexistence can continue indefinitely.  OA archives and journals will tend to stick to the distribution of texts and data, since they will always have reasons to keep their expenses to a minimum.  If free or affordable tools can add a layer of utility (like reference linking or XML tags), then OA resources could provide those layers of utility.  But for layers that could only be added at greater cost, TA journals may always have a role in providing them.

Note, though, that this is a sketch of diverging products, not just diverging business models for the same product.

* Even if individual TA journals are able to find subscribers indefinitely, the system in which all or most journals are TA cannot survive.  The reason is simply that the TA system will not scale with the explosive growth in published knowledge.  This would be true even if TA journal prices were low now and guaranteed to remain low forever.  It would be true even if there were no monopolies or anti-competitive practices in the journal industry.  It would be true even if the growth of published knowledge slowed down greatly.

The current system is already dysfunctional and has been for 10-20 years.  Charging for access to each piece of an explosively growing body of knowledge will only make the access gaps for individual researchers larger and larger over time.  Pressures on libraries to cancel some priced journals, and to refuse to subscribe to others, will only increase over time.  The only way to avert these outcomes is to make library budgets grow at the same pace as published knowledge, and I don't know anyone who thinks that will happen.  (Publishers who say "don't fix what isn't broken" have not been listening to librarians.  They can start now or they can wait until their journals are cancelled.) 

The fact that published knowledge is growing much faster than library budgets --the scaling problem-- is a reason to think that most TA journals around today will either not survive or not remain TA.  It's an argument that OA is superior, because it will scale, but not by itself a reason to think that the TA journals that lose subscribers will convert to OA.  They might fold up instead.  Still, it's a reason to think that coexistence will be temporary, or that the percentage of OA journals will steadily rise.

(Every time I raise the scaling argument, someone counters by saying that the TA system provides a welcome check on information overload.  My brief reply:  Information overload is a problem, but don't be satisfied with crude solutions when there are much better ones.  I'd rather have access to all knowledge and use increasingly sophisticated tools for carving out the subset I need than to have access only to a shrinking subset of knowledge determined by what I can afford rather than what I need.)

* OA archives do not provide peer review and depend on the continued existence of peer-review providers.  Publishers of TA journals have argued that widespread archiving would kill TA journal subscriptions and thereby kill the peer-review providers.  The fallacy here is to assume that the only peer-review providers are TA journals.  That's untrue.  Not only are there 1,440+ OA journals today providing peer review, there are likely to be many more over time.  Moreover, there's no reason why peer-review couldn't become a service decoupled from publishing and provided by organizations (like free-floating editorial boards) that look very different from anything we call "journals" today.

In short, OA needs peer-review providers but it doesn't need TA peer-review providers or TA journals.  By itself this fact doesn't cut for or against long-term coexistence.  But it does undercut the policy argument that the survival of peer review depends on the survival of TA journals.

* Open source and closed source software coexist, and seem likely to coexist forever.  At first this looks like the difference between journals and books:  if authors choose to write royalty-free software rather than royalty-producing software, that's their choice, and over the long term we'll continue to see choices of both kinds.  But on closer inspection, open-source tends to win against closed-source when they compete in the same application niche, when quality is roughly equal, and when monopoly power does not distort adopter decisions.  If the analogy to OA is strong (is it?), then this argues for continuing growth of OA at the expense of TA.

* Some TA publishers argue that TA journals are better for science.  But many concede that OA journals would serve science better if only they had secure funding models.  Those in the second camp that have not adopted OA are still looking at the evidence for financial viability or still running their own experiments.

Those in the first camp object to OA journals for anyone, but those in the second camp are objecting only for themselves or only for now.  Publishers in the second camp will not convert without a secure source of funding, and don't expect anyone else to do so either, but they tend to acknowledge that a secure source of funding would answer the major objections and make the case for OA compelling.

I won't try to guess the relative proportions of TA publishers in the two camps.  But let me focus on those in the second camp.  Note the asymmetry between them and OA providers.  Most OA providers argue that OA is best for science and scholarship in every discipline, not just for themselves. 

What does this asymmetry mean for long-term coexistence?  To me it suggests that OA proponents will continue to push for OA, in all disciplines and all languages, but that many of those not yet persuaded will only oppose it in their own cases.  If so, OA will spread to the extent that it successfully addresses individual circumstances --this research niche in this country in this decade.  If there are circumstances in which OA will never be viable, or in which finding viable funding models is unusually difficult, then at least there won't be circumstances in which OA proponents will stop trying.


Read this issue online

SOAN is published and sponsored by the Scholarly Publishing and Academic Resources Coalition (SPARC).

Additional support is provided by Data Conversion Laboratory (DCL), experts in converting research documents to XML.


This is the SPARC Open Access Newsletter (ISSN 1546-7821), written by Peter Suber and published by SPARC.  The views I express in this newsletter are my own and do not necessarily reflect those of SPARC or other sponsors.

To unsubscribe, send any message (from the subscribed address) to <>.

Please feel free to forward any issue of the newsletter to interested colleagues.  If you are reading a forwarded copy, see the instructions for subscribing at either of the next two sites below.

SPARC home page for the Open Access Newsletter and Open Access Forum

Peter Suber's page of related information, including the newsletter editorial position

Newsletter, archived back issues

Forum, archived postings

Conferences Related to the Open Access Movement

Timeline of the Open Access Movement

Open Access Overview

Open Access News blog

Peter Suber

SOAN is licensed under a Creative Commons Attribution 3.0 United States License.

Return to the Newsletter archive