Sign In

Communications of the ACM


Push Versus Pull

Push Versus Pull, illustrative photo

Credit: Flickr

The academic publishing industry provides an invaluable service to the academic research community. This publication infrastructure allows for the sharing of knowledge, facilitates new discoveries, and supports a vibrant exchange of ideas. Universities require and support the publication infrastructure and the associated peer-review vetting process through access subscription fees. A by-product of such scholarly activities and dissemination is that research institutions use peer-reviewed publications to make hiring, retention, promotion, and tenure decisions. Open access pushes publication costs onto authors (and their funding agencies), which devalues access subscription fees paid by institutions. The weak link for research vetting and dissemination is the business model to sustain the academic publication ecosystem.

Nearly everyone can agree that the free exchange of research ideas via publications is desirable and beneficial. In a financially strained environment for public and private academic institutions in the U.S., as well as numerous international institutions, particularly in the developing world, there is a growing interest to make such access to publications available at little or no cost.

The access subscription fee model represents a "Push Model" for financing publications, whereby publishers push access onto institutions, for an access subscription fee. An alternative "Pull Model" positions publishers to pull submissions to their journals. In such a model, no access subscription fees are paid, and hence, anyone can access research publications. What institutions would pay are submission subscription fees, an institution-wide fee that permits their entire faculty to submit papers to a particular journal or a publisher's portfolio of journals.

By design, the Pull Model is a systemized version of open access, where institutional submission subscription fees can be drawn from multiple sources (libraries, indirect cost funds, gift funds, or direct costs from research grants). What this does, however, is empower institutions to provide input on publication outlets that are valued, and provide their faculty the mechanism to publish in such journals. If researchers choose to submit research to journals for which their institution does not pay a submission subscription fee, researchers can still submit by paying a single paper submission fee, akin to an open access fee. This positions the Pull Model as a natural extension of the current Open Access model.

The publishing ecosystem is complex. A simple shift from Push to Pull will not end the publishing business model debate. The quality of research is often associated with the perceived quality of the journals (often based on the rigor and integrity of the peer review process) where they are disseminated, as well as quantitative measures like impact factors and citation counts. Several professional societies publish their own journals, which provide a service to their members, and quite often, are a reliable stream of revenue, funded primarily by academic/research institutions based on access subscription fees. In some instances, professional society membership may be partially driven by access to their journals. For-profit publishers like Springer, Elsevier, and Taylor and Francis, among others, serve a similar purpose as professional societies, in regard to publications.

By inverting the business model for publications based on submission rather than access, anyone can access published material. This is the future for academic research publications. One consequence of the Pull Model is that it places more responsibility on institutions and faculty to choose which publications to pay institutional submission subscription fees. Such a free market publication model will also put pressure on for-profit predatory publications (with thinly veiled review processes) to reevaluate their role in the publication ecosystem, since if no institutions pay their institutional submission fee, they will rely on per-paper submission fees (as they do now), which may not be viable in this new publication environment. Moreover, the Pull Model will implicitly create a journal hierarchy (and ranking), based on the number and reputation of institutions willing to pay their submission subscriptions fees. Such a market-based approach to fund the publishing ecosystem will enhance the value of quality, peer-reviewed publications and devalue predatory journals.

The proposed Pull Model may be viewed as unfair to research institutions, who contribute the preponderance of journal content, while also paying a significant portion of submission subscription fees. To assess this conjecture, in 2016, the ACM Digital Library (DL) generated over US$20M in revenue, publishing approximately 24,000 papers by 44,000 authors. Assuming a 30% acceptance rate, this translates into approximately 80,000 papers submitted and US$250 per submitted paper. Given that there were over 7,100 institutions that had one or more authors on these papers, a case study can be constructed that parses these institutions into three tiers (see the table here), based on their submissions volume.

To determine what each of the three tiers listed in the accompanying table would pay as a submission subscription fee, a baseline of US$500 per submission was used, which is in line with the high end of ACM's open access fees and the assumed acceptance rate. Tier I and II institutions would pay submission subscription fees of US$16,000 and US$10,000, respectively. The total amount collected by ACM for their DL would be the same as what it is currently collecting based on access subscription fees. The result of this allocation would mean that the highest volume submitting institutions would pay a slightly higher fee than they current do (for example, the University of Illinois at Urbana-Champaign paid just under $15,000 annually in 2016 for their access subscription.) Schools that have few submissions would pay less than what they do today. Of course, these numbers represent rough estimates, since once implemented, equilibrium pricing will be reached based on supply and demand pressures. For example, it is difficult to predict how Tier III institutions will react to this pricing, since many of their authors may have co-authors in higher tiers (and hence, avoid any fee). However, by including all institutions with up to 20 submissions in this tier (based on the 30% acceptance rate), some of the revenue generation uncertainty is mitigated.

Based on this data, institutions with few submitters may be viewed as free-loaders in the system, gaining access to everything while contributing no content. I would counter that such thinking focuses entirely on a single dimension of value (cost), and does not recognize the high value that the publication ecosystem provides to research institutions and their internal faculty review process. One can even argue that research institutions have been underpaying for this service, and that the Pull Model rights this transgression, while stabilizing the current peer-review publishing ecosystem. These points suggest that beyond economic factors, the Pull Model has more potential benefits than detriments to research institutions. Moreover, the current access subscription fee paradigm is unsustainable and being challenged by sites like Sci-Hub,1 which allows anyone to access scientific research articles at no cost.

Figure. Pull Model case study.

The case study given here is but one possible scenario; specifics of the proposed subscription-submission-fee based Pull Model must be worked out among the stakeholders. The transition from a Push Model to a Pull Model will likely be traversed via incremental steps from one paradigm to the other. In fact, the transition period may be highly problematic, since no institution wants to pay both submission and access subscription fees. One option will be a phased transition, whereby some journals are Push and some are Pull. The transition can also be executed on an institution by institution basis. Both these transitions can be done by subject area, or across all areas, balancing the volume of submissions so that the transitions affect a comparable number of submitters in different communities. During such transition periods, submission subscription fees may be instituted for certain journal, while access subscription fees will be reduced by a commensurate amount. For the transition to be effective, numerous questions must be addressed. What factors will go into setting submission subscription fees? Would it be based on historical submission data or size of institution (number of faculty, number of students, size of funded research programs)? Would such fees be sufficient for the publishers to continue to deliver their products? It is reasonable to expect that the actual aggregate amounts paid by institutions will be similar to what is being expended today? At first glance, it may be that research institutions will end up paying more during this transition period, until some type of steady state is reached. Clearly, the pathway from Push to Pull will be fraught with an endless stream of questions, with each stakeholder protecting their position. The Pull Model may also lead to for-profit publishers abandoning the academic publishing market (if it becomes unprofitable for them to exist), relinquishing this responsibility to not-for-profit entities who would fill the void. Laasko et al.2 provides an overview of approaches to convert to open access, some of which can be used to guide the Pull Model transition and mitigate pitfalls along the way.

Time is becoming a factor in this debate, as the seeds of such a transformation are already in motion. Vogel5 notes that universities in Germany decided to not pay subscription fees to Elsevier, effectively challenging the subscription fee model. Park and Seo3 outline a Korean publishing service that facilitates open access. Satlow4 provides a commentary on separating the review process from the dissemination process. By separating the key facets of the publishing ecosystem, the proposed idea provides an à la carte menu for financial support, which shares some of the free market aspects of the Pull Model.

Reputable society publications and publishers should welcome such a business model shift. Some academic institutions may find ways to avoid paying submission fees while having their researchers still publish their research; however, the sheer diversity of faculty and their publication needs will make this a challenge. In the Push Model, researchers who gain access without paying access fees can remain anonymous. In the Pull Model, with submission and eventual publication, researchers who avoid paying submission fees will be exposed by who they list as their institution affiliation and/or co-authors. Therefore, the Pull Model creates a transparency (for submissions) that the Push Model is challenged to achieve (for access).

The best consequence of the proposed Pull Model is access for all. It also introduces a free market mechanism for scholarly publications, whereby publishers must compete for institution submission subscription fees, by establishing themselves to be worthy outlets for dissemination, maintaining their reputation for quality, and preserving the integrity of the peer-review process. Lastly, it encourages institutions and their faculty to work more closely in assessing publication quality. With these ends in mind, the future of publications will continue to change, and the Pull Model, though disruptive to the existing publishing ecosystem, is one step to initiate a discussion on such a transformation.

Back to Top


1. Bohannon, J. Who's downloading pirated papers? Everyone. Science 352, 6285, 508–512.

2. Laasko, M., Solomon, D., and Bjork, B.-C. How subscription-based scholarly journals can convert to open access: A review of approaches. Learned Publishing 29, 4 (Apr. 2016), 259–269.

3. Park, M. and Seo T.-S. Creating a national open access journal system: The Korean journal publishing service. Journal of Scholarly Publishing 48, 1 (Jan. 2016), 53–67.

4. Satlow, M. Academic publishing: Toward a new model. Commentary, Chronicle of Higher Education 62, 38 (June 2016).

5. Vogel, G. German researchers start 2017 without Elsevier journals. Science 355, 6320 (2017), 17.

Back to Top


Sheldon H. Jacobson ([email protected]) is Founder Professor in Computer Science at the University of Illinois at Urbana-Champaign, USA.

Back to Top


The author thanks two anonymous reviewers for their comments, resulting in a significantly improved Viewpoint, which was also written with support to the author from the National Science Foundation (CMMI-1629955). Any opinions, findings, conclusions or recommendations expressed in this material are those of the author and do not reflect the views of the U.S. government, or the National Science Foundation.

Copyright held by author.
Request permission to (re)publish from the owner/author

The Digital Library is published by the Association for Computing Machinery. Copyright © 2018 ACM, Inc.


No entries found