MustRead - Fee Statement
Author Fee, or the Article Processing Charge (APC), is always a complex and controversial topic. There have been probably hundreds of relevant articles or blogs, formal or informal, if one searches it online.
Below we first share some general perspectives, and then disclose the rationales behind ATC's APC policy.
1. Traditional Library-Fee Model
For decades, authors have been publishing for free and libraries or institutions must pay for subscriptions.
For those familiar with the university library systems, this has been snowballing into a huge annual financial burden. For instance, this cost survey in 2019 published by UCSF – “Journals Cost How Much,” states:
“In 2019, the average price of a health sciences journal was $2,156 and for a chemistry journal was $5,950.”
(From the bar graph) In 2017, universities such as Stanford, University of Pennsylvania, Washington University spent more than $4 Million on Health or Medical journals alone.
Let be examined what the main cost sources could be or can be justified – (a) Rental of office spaces (for professional publishing houses such as Elsevier, Springer Verlag or those associated with societies or organizations); (b) Printing costs; (c) Staffs: contracted art designers for designing logos, covers, or special issues, full- or part-time secretaries to the Editor-in-Chief, copyediting staffs (full time or contracted), printing managers, marketing staffs who fly to major conferences to advertise or who must successfully engage library clients, etc.
The main cost source is clearly from humans – the salaries, health and retirement benefits, or contracted compensations of art designers, secretaries, copyeditors, printing managers, and marketing staffs. Publishing multiple journals can dilute the direct burden on any specific journal.
Human costs merely reflect the varieties of services that a quality journal has traditionally relied upon.
The most common complaint ignores the above costs by exclaiming that “Editors, AEs, and referees are all free!”
2. Open-Access Fee Model (APC)
The APC model for open-access journals has partially alleviated the financial burdens on non-profit libraries or institutions around the world. The burden can be entirely dissolved only if all journals become open-access.
Following the discussion in Section 1, for online only open-access journals, other than the elimination of the printing cost, all the other costs still stay for publishing houses:
- Contracted art designers for designing covers, logos, or special issues
- Secretaries to the Editor-in-Chief to address all BAU (business-as-usual) issues such as author inquiries
- Copyeditors for the final article productions
- Marketing staff – still trying to advertise in conferences or to have the journal(s) enlisted in libraries (for free), mainly to help boost the so-called impact factors and journal reputations, which in return can attract more authors to submit and potentially generate more revenues.
Yes, the main cost source is still from humans, which of course represents the cost of services they contribute. Digitization does offer cost reduction, e.g., no more printing costs, and much cheaper online advertising services such as via Google Ads.
For most small-scale open-access journals unassociated with the major publishing houses, marketing staffs can be completely substituted by cheaper services such as Google Ads (another unfortunate example of “good for cost reduction, but bad for daddies or mommies in need of a job”).
Digitization, however, does entail one new type of human cost – IT supporting staff, unless it can merge with other roles mentioned above:
- For decades, manuscripts, rounds of revisions or referee reports are all stored in file folders, manually by the Editor-in-Chief or more commonly, the secretary.
- Behind a modern online journal, however, is an evolving database (e.g., via MySQL). All the above pieces of information are collected through web forms and stored into the database in the backend. They can then be queried through various APIs and presented in human-friendly forms (e.g., an html form for an article).
- Digitization eliminates the physical printing process. Instead of reading physical copies, readers now access the journal exclusively via its webpage, which becomes almost the only identity of the journal. Hence a professional design of the journal must be strategically developed and maintained by someone.
Thanks to non-profit initiatives such as the Public Knowledge Project (PKP) and unifying efforts such as the Open Journal System, these IT burdens have been substantially alleviated for individual publishers. Still, at least one IT-savvy staff who also knows web design must work behind for proprietary designs and support.
To summarize, for online open-access journals, the main cost source is still from humans.
The loudest complaint ignores the above costs by exclaiming that “Editors, AEs, and referees are all free!”
Notice that herein the following relatively lighter costs are not mentioned: annual web hosting (sometimes via more expensive Virtual Private Servers or Dedicated Servers for more versatile server-side programming), online Ads fees, etc.
Listed below are some studies or statistics that can be easily pulled out via internet search engines:
- (October, 2018, by the University of Cambridge, UK) “Generally the range is between $2000-3000…;” The average open access processing charge … is £2147 (an average across the entire university).
- (November, 2019, by Sustaining the Knowledge Commons, Canada)) “Average APC price by publisher ranges from 246 to 2,851 USD. ”
- (2013, by Nature; Probably the most well-known article) “…an average revenue per article of roughly $5,000. Analysts estimate profit margin at 20-30% for the industry, so the average cost to the publisher for producing an article is likely to be around $3,500-4,000.” “The largest open-access publishers – BioMed Central and PLoS – charge $1,350 – 2,250 … in many of their journals, although their most selective offerings charge $2,700-2,900.” “Outsell estimates that the average per-article charge for open-access publishers in 2011 was $660.”
3. ATC's APC and Its Rationale
As elaborated separately in “MustRead – Journal Model,” for ATC, the Dedicated Editor (DE) integrates all the roles aforementioned. The notion of DE is the unique invention of ATC, adventurous, and probably risky, but only time will tell! It is a bit difficult to generalize, and actually could become detrimental to academic reputation and integrity if being abused.
An enlightening analogy is the following. When one needs a new glasses prescription, instead of turning to big chain stores such as Lens Crafters or Cohen’s full of staffs, one step into a small and old-fashioned optical shop in Harvard Square, where the owner does all the job for you – from eye examination to frame selection. This is the exact parallel of DE, and hence the DE has to be a true expert or someone understanding the field sufficiently well.
The DE not only designs all the proprietary properties such as the logo, the cover, and even the painstaking tex class file for a unique journal-specific latex style, but also constructs the entire web backends/frontends and creates all the web contents. Each designed component must be proprietary in order to avoid any potential copyright infringement.
Most importantly, the DE has to review all the submitted manuscripts, with the journal’s 5D5W commitment. Each review itself is an individual project which demands deep diving, online research, and real-time communications with the authors, etc.
Finally we are at a position to disclose the assumptions and calculations for our APC policy.
- The DE must be a Ph.D. who has attained sufficient expertise in the field, in order not to reject too many submissions under the journal’s “discretionary” principle. The DE must also be very familiar with the general refereeing process.
- Assume that this is the full-time job of the DE whose compensation target is what an Assistant Professor (AP) is paid on average.
- The DE’s quantitative skills should be trained in Math/Stats, and programming and web skills via Computer Sciences. Hence the DE’s compensation is mainly benchmarked against Math and CS APs in major universities in USA.
- According to the most recent salary survey by the American Math Society (AMS) in 2018-2019, the average salary of new Math APs is about $95,340 for big public universities that we are all familiar with (who are usually treated as state employees). (The 9-month figures of course do not include summer pays from grants, nor health or retirement benefits.)
- For CS APs, according to the most recent 2018 Taulbee Survey by the Computing Research Association (CPA), the median 9-month US CS AP salary is $97,874 or higher (for different categories).
- To facilitate calculation, the compensation target of the DE is hence rounded to $96,000 (a multiple of 12), which amounts to $8,000 per month (before tax, and without health or insurance benefits).
- As elaborated in the journal’s “MustRead – Journal Model," for quality control, the DE limits reviews to about 5 submissions each month (in order to fulfill the 5D5W commitment and avoid piling up of backlogs). Hence each submission should contribute about: $8,000/5 = $1,600, per submission. (In practice, rejected submissions are not charged.)
- As of Q4, 2020, we plan to launch two journals in parallel in the expertise scope of the DE.
Hence the APC is finalized at: $1,600/2 = $800.
That is all behind our APC pricing policy: $800 per accepted article. It could fluctuate in the future, but any change does not impact the ones already published, be it higher or lower. (We charge $400 for Commentary Articles without much technical details. The journal mainly publishes Research Articles, though.)
We also remind about these facts: (1) the DE, self-employed, must pay health insurance out of pocket, (2) the DE does not earn retirement benefits (e.g., 401(k)), and (3) on top of the regular federal and state (and city for NYC) taxes, Uncle Sam charges 15.3% for Social Security and Medicare for self-employment (2020). Hence in reality, the annual compensation of the DE (as hypothetically designed above) is way south to the average total compensation of the benchmarking APs.
(The above detailed but heuristic rationales will be removed from this page after an initial period. They are presented here to give our readers and authors some comfort and transparency behind the journal's APC configuration.)