In April 2014, The Andrew W. Mellon Foundation awarded a planning grant to Ithaka S+R to convene a panel of experts to develop a study methodology for determining in as granular a way as possible the true costs of publishing scholarly monographs. The workshop brought together some of the best thinkers on this topic and resulted in a research methodology for this project, also funded by The Andrew W. Mellon Foundation and conducted by Ithaka S+R from January through November 2015. This research project takes on a fundamental question at the heart of any potential new model to support OA monographs: What does it cost to create and disseminate them?
The goals of the research were to
- provide a comprehensive list of all of the activities needed in order to produce and disseminate a high-quality digital monograph;
- generate empirical data on what it costs presses today (what activities they are undertaking today) to produce those books; and
- offer recommendations of general principles to guide presses in seeking to establish price points for author-side payments for Open Access digital monographs.
Data gathered from twenty participating presses, all members of the Association of American University Presses, suggest that monograph publishing today is considerably more expensive than has often been reported anecdotally or in other studies, and certainly more expensive than current price points for publishers with OA models would suggest.
The study gathered costs of 382 titles published in Fiscal 2014. The data gathered included estimates of staff time, direct expenses (such as the cost of a freelance copyeditor), and press-level overheads (such as legal support or rent). The most important aspects of this study are the focus on developing a definition for the full cost of publishing, and developing cost estimates for staff and non-staff expenses in core press activities—Acquisitions, Manuscript Editorial, Design, Production, and Marketing—in part by having staff estimate their time spent on monographs and on certain activities involved in producing a book. By using the methodology of calculating the component activities of book publishing among twenty presses, we were able to examine the staff time allocations in bottom-up fashion, thereby separating academic monographs fairly cleanly from the presses’ other activities (including journals and trade publishing).
The study focused solely on the costs of producing the first digital copy of a “high quality digital monograph.” As such, the study excluded trade titles, textbooks and databases. It also excluded works in translation, edited volumes and paperback reprints.
Working with this data, the research team developed three cost “definitions”:
- BASIC: Which includes the cost of staff time and other non-staff direct costs in Acquisitions, Manuscript Editorial, Design, Production, and Marketing.
- FULL COST: Which includes the cost of staff time and other non-staff direct costs in Acquisitions, Manuscript Editorial, Design, Production, and Marketing; and also includes Press-Level Overheads.
- FULL COST PLUS: Which includes the Full Publishing Cost, and also includes “in-kind” contributions.
This study of 382 titles across 20 university presses from four category typesyielded a wide range of costs per title, from a low of $15,140 to a high of $129,909 and the range of costs is wide both within and across groups.
Table 1. Full cost of a High-Quality Digital Monograph (Excluding In-Kind Cost)
|Group||Group Average||Group Median||95thPercentile||5thPercentile||Highest Cost Title||Lowest Cost Title|
Studying the costs of publishing nearly 400 monographs across 20 university presses, provides a descriptive account of how university presses of different sizes and with different missions account for the costs of monographic publishing. As such, it offers a model for how a university press can study its cost structure. Our hope is that publishers and funders alike will be able to describe more accurately what they are including—and not including—when the elusive subject of publishing costs arises. Twenty presses joined this effort to dig deeply into staff, overhead, and other costs and shared data in a common way so that, as a community, we can begin to understand all of the variables in the equation.
Among the key findings:
- Regardless of group type, the largest cost item for university presses is staff time, specifically the time related to activities of acquisitions, the area most closely tied to the character and reputation of the press. This activity is least likely to be outsourced, and considered to be closely tied to its financial success: acquisitions editors being the ones with the skill, subject expertise, and relationships needed to attract the most promising authors and topics to the press.
- The working hypothesis at the outset of the study was that larger presses would demonstrate a lower per-book cost, presuming that larger houses are able to work more efficiently due to the economic benefits of scaling. Based on the data contributed by the individual presses, the small university presses in group 1 have been able to produce monographs at a lower cost than the other groups. It is impossible to determine if this signals greater efficiency on the part of the small presses or it simply means they underinvest in their publications.
- We looked for significant determinants of cost. While press size, page count, and number of illustrations showed a relationship to cost, other factors, including whether or not the title was a “first book,” whether or not the press was at an institution that required it to pay rent, or whether the press was at a public versus private institution, did not. An examination of disciplines was not conclusive, due to small sample size.
The study also captured costs concerning the Press-Wide overheads, defined here as departmental overheads and General and Administrative expenses (G+A). These figures are important to consider, since at the most basic level, they are integral to the press functioning as a business. And yet there is still real debate about the best and most equitable way to assign those costs at a product level. Further complicating the question is the underlying reason for the costs themselves: should higher per-book costs be interpreted as a sign of press inefficiency, or a reflection of a healthy press, in a position to devote and invest greater resources to its work?
In the months ahead presses, funders and university administrators may try to determine how these costs can be best used to construct a system to support author-side subsidy for monographs. This study suggests the need for further study and discussion of the implications of OA, not just on libraries’ materials budgets, but on:
- The role of marketing in encouraging wide-spread dissemination of scholarly works. The work of marketing departments serves today to sell books, but in its essence it is a machine intended to reach the widest audience for a given book. If “impact” replaces “sales” as the best way of framing success, what effect may this have on the strategies presses adopt to market monographs? Which specific metrics will the presses gather, and what systems will they need to harvest and aggregate these?
- The impact on revenue of introducing OA titles into the list. As publishers introduce OA titles to their lists, most, if not all, will still have print and perhaps even a premium or enhanced digital version for sale via consumer channels. What will the impact on revenue per title be as a result of this? While such sales for an OA title will likely decline, it may depend on the title and the formats offered.
- The credentialing role of publisher. In journals publication, credentialing is discussed primarily in terms of peer review. For monographs it is important to also articulate the role of the acquisitions editor in identifying and developing scholarly work, the place of selection by an editorial board, and the way in which processes such as copyediting, design, and marketing also assure the quality of a work. As monographs transform into works of long-form digital scholarship, the values behind these processes of selection and quality control may stay the same, but the nature of the actual processes is likely to change to reflect the new demands of multimodal works.
So many people contributed to this study in ways both big and small, but mostly big. First, sincere thanks to Don Waters and Helen Cullyer at The Andrew W. Mellon Foundation, for entrusting this project to Ithaka S+R.
The advisory group that participated in the planning grant and the Advisory Committee for the current study included press directors who are deeply engaged in running their own businesses and sharply alert to the significance of the study. They have from the start made themselves available, both in person and via email and their guidance in shaping the study and interpreting its findings has been invaluable.
The group of 20 press directors who volunteered to take this adventure with us was exceptionally generous in the time and thoughtfulness they applied to this work, both in talking with us, in welcoming us to their offices, and most of all, in supporting this project and granting us access to their most valuable resource, their colleagues. Our whole research team wishes to thank the 406 press staffers who shared their experiences with us. While our main objective was to develop sound financial data, the staff meetings offered an opportunity for us to hear from the field’s most experienced professionals and its newest members. We learned a great deal from them about the cost drivers of publishing, but also about the passion and expertise they bring to their work.
As funders and academic organizations have begun to address the implications of open access to scholarly resources, one question comes up repeatedly: What does it cost to create, produce and disseminate—to publish—a high-quality digital monograph? This study was conceived and designed as a means to capture not just a single bottom line figure, but to capture the component costs of creating a scholarly book.
The “crisis” of University Press publishing may well be, when viewed over the long term, more of a chronic condition. And yet, the current model faces particular challenges that require attention. The prevailing business model for many years, still largely in place today, assumed that a press would invest in creating a book, reproduce it in print form, and manage its distribution through relationships with scholars and scholarly institutions, in particular their libraries and the wholesalers who served them. Sales through wholesalers, retailers, institutions, and individuals would generate the revenue to cover the initial investment, if not more.
But the number of units sold per book has declined over time. University press directors describe a downward trajectory, perceptible over the past decade or so: what used to be a typical sale in the low thousands of books, is now often in the low hundreds. And yet the effort and investment needed to create that book has not decreased in step with declining revenues, leading to a financial situation many publishers understand to be unsustainable.
While publishers are actively seeking solutions to counter declining sales of monographs, there is a growing expectation from some corners—and a bona fide legal requirement from large federal funding agencies—that scholarly content be made openly available, free to all. Open Access publishing models, where a fee is paid by the content creator (or funder) to support the costs of publication have earned a strong foothold in journal publishing, where there is some evidence that an author-pays OA model can be economically viable. Most book publishers have stayed on the sidelines in this discussion, for reasons both practical and philosophical, among them that books cost more to produce and humanities authors have fewer financial resources such as grants at their disposal to cover those costs. This moment offers an opportunity to re-examine book publishing business models and the assumptions that underlie them, with an eye to developing new models that can both deliver sustaining revenues to the presses while affording greater access to end users. Indeed, university press directors, deeply aware of how declining monograph sales have had an impact on their revenues, have already begun in earnest to experiment with other models, with the hope of increasing both revenue and ultimately the reach of scholarly works. The University of California Press (UCP) launched the monograph-focused Luminos and journal-focused Collabra in February 2015. While Collabra is focused on the sciences, where open access is already an established presence, Luminos is humanities-focused and embraces a model whereby costs are shared by the parties who benefit from publication—author or institution, publisher, and libraries.Other established university presses, both large and small, have also launched OA models, including Oxford University Press and Manchester University Press; Amherst College Press was recently created explicitly to be an open access publisher, as were Ubiquity Press and Open Book Publishers. And commercial academic presses as well, including Elsevier, Routledge, Palgrave Macmillan and Wiley, have also entered this space. 
The presses that are currently engaging in publishing OA monographs have necessarily needed to develop pricing models. UCP’s Luminos requires a fee of $15,000 per title, and others that have developed new models include Palgrave Open ($17,000), Routledge and Taylor and Francis (£10,000), and Brill ($6940 for CC-BY-NC). Some publishers offer different price points, depending on how restrictive (or not) the license is.
Publicly shared pricing models should not be confused with the actual costs of publication, which may be greater, the same, or less. A pricing model reflects strategic choices of the press: it could align with actual expenses incurred, include enough margin to be profitable, or be priced below cost, if the press can subsidize the expense in other ways, as a means to quickly enter the market with a competitive offer.
Ultimately, however, pricing models need to be premised on actual costs, and the “cost of a monograph” has been extremely hard to pin down. While on one hand, publishers carefully track and forecast many types of costs, particularly the variable costs related to printing and other “out of pocket” expenses, internal staff time is rarely allocated back to the cost of producing a book. Arriving at a figure that accurately represents the full costs of publishing a monograph may require publishers to undertake certain accounting practices that they have not in the past. Having a deeper understanding of the costs of all the activities that go into the creation of a book will be valuable to publishers, regardless of what business models they choose to engage in. Such an understanding is also important for libraries and other stakeholders who care about the preservation of a vibrant monograph publishing ecosystem, since relying on numbers based on evolving and idiosyncratic business models risks undermining long-term sustainability. Should Open Access publication of monographs expand, it will be important to funders who will expect transparency in accounting from the publishers whose work they support.
Over the years, several presses have developed their own means of determining the cost of publishing a monograph, with estimates for first-copy costs for print books ranging from $15,000 to $25,000 and higher. Presses participating in the Knowledge Unlatched pilot in 2013-2014 each had to develop their own price of a book in order to participate; the price range of participating publishers was from $10,000 – $15,000. A recent study conducted by Indiana University and University of Michigan identified the average cost of publishing a monograph at their two presses as being around $27,000.
Others have suggested that full publishing costs for an academic monograph could be as high as $50,000 if all overheads are considered. Some presses, like National Academies Press, are able to develop a per-book cost assessment, given its current practice of tracking staff time, though due to their particular organizational model, the costs do not include several categories of activity, such as peer review, that most university presses would include.
Industry-wide statistics are harder to come by. While the AAUP itself gathers statistics each year, these costs are captured at the press level. This is very useful for benchmarking, but assumes that the unit of measure is the press and not the book. Those industry-based studies that have taken on some of the complexity of the book publishing model, including variable categories of costs, have not yet resulted in models that US-based publishers believe sufficiently take into account the categories of costs that a sustainable model would need to account for. A report from October, 2013, from OAPEN –Netherlands, entitled, “A Project Exploring Open Access Monograph Publishing in the Netherlands,” starts to demonstrate this complexity. The study focused on a sample of 50 titles from nine publishers in the Netherlands and showed an average cost of EUR€12,000 (or about $16,700) and a range of costs for the 50 books studied from €6,082 to €20,660.
In 2014, during the planning phase of this project two major papers appeared, outlining new ways to fund the creation of scholarly work. “A Scalable and Sustainable Approach to Open Access Publishing and Archiving for Humanities and Social Sciences,” issued in February 2014 by Rebecca Kennison and Lisa Norberg, outlined a model for an institution-based approach. In June 2014, the Association of American Universities (AAU) and the Association of Research Libraries (ARL) offered a “Prospectus for an Institutionally Funded First-book Subvention.” This approach also takes an institutional direction, with funding provided to universities to then redistribute to faculty members who are publishing first books. The reports’ contribution was in developing models for supporting Open Access scholarly content subsidies on a large scale, by estimating the size of the market, the scale of the demand and potential sources of funding. Yet both reports acknowledge that the figure of $20,000 per book was considered a placeholder only.
The Andrew W. Mellon Foundation awarded Ithaka S+R a planning grant in April 2014 to study the costs associated with academic monograph publishing. Ithaka S+R convened meetings of experts and facilitated conversations aimed at developing an effective methodology to capture all of the activities and costs needed to produce and disseminate a high quality, digital monograph.
The group met virtually starting in May and in person on June 4, 2014 in New York City. Participants included Peter Berkery, executive director, and Brenna McLaughlin, director of marketing and communications, of the Association of American University Presses (AAUP); Charles Watkinson, at that time director, Purdue University Press and head of Purdue Libraries’ Scholarly Publishing Services (and now associate university librarian for publishing and director of University of Michigan Press); Ellen Faran, director, MIT Press (now retired); Mark Saunders, director, University of Virginia Press; Barbara Kline Pope, executive director, The National Academies Press; and Kim Schmelzinger, former associate director and chief financial officer of Northwestern University Press, who manages the annual AAUP Operating Statistics survey and analysis. The methodology recommended in this proposal, described below, was developed throughout the course of the planning grant.
Based on the recommendations of university press experts, Ithaka S+R adopted a bottom-up approach, addressing cost at the title level and the press level. When considering approaches for the current study, the task force explored several alternatives, including an industry-wide survey and a top-down assessment of current press costs. The approach developed during the course of the planning grant addresses cost at the title level and the press level. By collecting detailed, activity-based costs and overhead expenses at the press level, we have undertaken a careful examination of the costs of published monographs. By conducting this exercise with a variety of university presses, including representatives in each size bracket of the AAUP membership, we are able to analyze key press-level differences and also provide comparisons for non-participating presses to consider. The limitations of this methodology are discussed in the appendix, but we believe that one of the most important contributions of this report is the breakdown of costs by specific activities associated with monographic publishing.
For the purposes of this study, a scholarly monograph was defined as “a work of scholarship on a particular topic or theme which is written by a scholar (or scholars) and intended for use primarily by other scholars.” We did not address costs of other publisher-issued content, such as regional and trade titles, textbooks or databases. Our definition of “monograph” excluded works in translation and edited volumes. While these works have much in common in spirit with the single-authored works included in the study, the means of their production would have introduced more challenges to the study.
The study collected data on the costs in the US-market for all activities related to creating and disseminating the “first digital file” for a book. This excludes costs that are only related to printing, binding and distributing print copies of books. Also excluded are advance payments made to authors and royalties from sales. The study captured both staff and other non-staff direct expenses attributed to specific titles, as well as operational overheads necessary for the publishing of a monograph. We consider these direct and overhead expenses together to make up the full publishing costs for a scholarly monograph. They include:
- The costs required to support a title over time (and not just up until publication).
- Other costs considered vital to an ongoing publishing program, including those that serve to build and promote subject-related lists and authors (not just individual books) over time.
As the scope of the study does not include questions related to revenue, we did not capture cost of sales in its many forms including royalty, distribution/sales discounting/sales commissions, and so forth. Similarly, costs related to print production, warehousing fulfillment and distribution were not captured. However, because even an entirely free digital file will require work and cost in order to go from publisher to its readership, we have included, where possible, costs related to distribution of and access to the file, including metadata creation, Search Engine Optimization (SEO), and e-promotion.
Starting from the AAUP’s categorization of presses by size, we selected five presses in each of the AAUPs four size categories (Table 2) taking some care for diversity even within each category concerning geography and publishing focus (monograph-centric, versus those with a balance of monographs and other publishing activities). Directors agreed to permit the data gathered to be shared in aggregated, anonymized form in this publicly released report.
Table 2. AAUP Press Categories, FY 2014 (n=67)
|Group||Annual Revenue, Books Division||Average Full Time Equivalent Employees||Average # Titles Published|
|1||Under $1.5 million||10.7||43|
|2||$1.5 – $3 million||21.4||77|
|3||$3 – $6 million||33.4||126|
|4||Over $6 million||78.9||242|
Table 3. Profile of Participating Presses
|Group||Average Full Time Equivalent Employees||Average # Titles Published||Average # Monographs Published|
From March through May 2015, our research team met in person with staff from the 20 participating presses to gather data. In addition to meeting with the business managers and CFOs to gather press-level financial data, the researchers held meetings with staff in Acquisitions, EDP (Editorial, Production, and Design), and Marketing, to walk through the staff time allocation together.
Data from staff and the business manager/ CFO were entered into Excel spreadsheets and sent to the research team at Ithaka, where it was compiled and cleaned. Any anomalies that were spotted were reviewed with press directors in conference calls in May and early June 2015. Each press’s data were shared back with them in the form of Excel files at that time. Further detail on the method of data collection and analysis can be found in the Note on Methodology in the Appendix.
The question of which costs to include stimulated a lively conversation among members of the advisory committee. There are many types of costs that presses track and assess regularly, particularly those related to the “out of pocket” expenses needed to publish any given title. These are the costs that are considered on a new book project’s profit and loss statement, a document developed at the outset of any new book project to assess the investment needed to create the book and the sales expected/needed to cover its costs or better. The basic calculations done in early stages of a decision to publish will take into account known pre-production and production costs, including printing, for example, and will also include assumptions and estimates from the sales and marketing teams concerning future sales through specific channels.
It can be harder to determine how to account for fixed costs, those expenses the press absorbs, regardless of how many books they may produce and sell in a year. Staff time, which includes salaries and benefits, is critical when talking about the cost of producing a monograph. An editor’s time spent selecting and developing a new work for example, should be included. But accounting for expenses that are less directly tied to the production of a specific book is more difficult. Should a share of the cost to have a company website be “charged” to each book? How about the director’s time? How is rent factored in? Most publishers do have some way to account for these “overhead” costs (in what is referred to as a “margin” calculation) when considering a book’s profit and loss statement. There is, however, little agreement about the best way to do this, either concerning which costs to include or how to most “fairly” share them among the products or titles a press publishes.
A main premise underlying the study is that the core activities that together are required to create and disseminate a high-quality digital monograph include activities grouped within the categories of Acquisitions, Manuscript Editorial, Design, Production, and Marketing.
There are many ways to consider the elements that together comprise the full cost of publishing. The elements we considered included:
- Staff costs: This includes salary and benefits for those working in the following departments: Acquisitions, Manuscript Editorial, Design, and Production. Staff time is the accumulated salary and benefit cost of the time that staff in these departments reported spending directly working on monographs. Staff overhead refers to time that staff in these departments reported spending on work not directly related to monographs or other publishing activity excluded from this study.
- Title-specific direct costs: This includes non-staff expenses that are directly attributed to the creation of a specific title, and are paid “out of pocket” by the press. These costs were gathered at the department level, at each of the departments identified above; Acquisitions, Manuscript Editorial, Design, Production and Marketing. For example, any outsourced work or services directly related to the creation of a monographs would fall into this category.
- Press-level overhead: Press level overheads were divided into two categories, General and Administrative expenses (G&A) and Departmental overheads. G&A include staff and non-staff expenses in departments such as administration, accounting, legal and finance, as well as cost for rent and utilities. Departmental overheads include non-staff expenses that are not directly attributed to the creation of a specific title from the departments identified above; Acquisitions, Manuscript Editorial, Design, Production and Marketing. For example, the acquisition department’s travel budget would fall into this category as well as activities like advertising that some presses do not break out by title.
- In-kind contributions: These include the many contributions the press may benefit from, but are not paid for. This can include rent, staff time from other departments, and work that the authors themselves take on, like clearing permissions or generating an index.
When we refer to the average cost of monographs in this report, we developed three definitions of “Cost.” They are all legitimate expenses that the press incurs in the course of publication, but some are more directly tied to the publishing process than others. We anticipate that these definitions will be discussed and debated in some detail in the community, as pricing models are being developed.
- BASIC COST: includes just staff and non-staff expenses directly incurred when producing the book. Some have referred to this as the “incremental” cost of a press adding “one more book” to its existing publishing operation. Here, our definition includes staff time and staff overhead, from those who work in the core publishing departments of Acquisitions, Manuscript Editorial, Design, Production and Marketing. It also includes what we are referring to as direct or “out-of-pocket” expenses which can be captured by title.
- FULL COST: Represents the BASIC COST, plus the press-level overheads.
- FULL COST PLUS: Represents not only the overhead costs included above, but even adds in IN-KIND contributions, the reported value of resources contributed to the project, but not paid for. These generally included contributed staff time, author-paid fees and office space.
The choices of what to consider when talking about costs are philosophical as much as economic in nature. Arguments abound about what sorts of costs to count and which to not count, particularly when developing a pricing model. The methods used by the researchers are essentially choices, and publishers will need to decide how, ultimately they will choose to calculate the costs of their work, as well as the implications for their business overall. But the report is intended first and foremost to provide data from the study, so that presses and funders will have a firm basis on which to make those choices. The pages that follow will include not just final figures, but discussion of how these totals were arrived at.
University press monographic publishing requires a fairly high-touch process, with many individuals in a variety of roles helping to shape the scholarly work from concept through final execution and distribution.
This study of 382 titles across 20 university presses yielded a wide range of costs per title, from a low of $15,140 to a high of $129,909, and the range of costs is wide both within and across groups (Table 1). Regardless of which definition we use, the total, average cost of publishing a high-quality digital monograph is considerable, and in almost every instance higher than the figure of $20,000 that has gained popular attention in recent months. Among the 382 titles in our sample, the overall average cost per title was $28,747 (basic) and $39,892 (full cost).
From these data, several patterns emerge:
- Generally, the smallest presses demonstrated lower costs and larger presses higher costs. Yet this was not uniform across the sample. Some smaller presses had very high per book costs, due to low title output relative to staffing levels.
- The data for presses in groups 2 and 3 demonstrated an interesting reversal with smaller Group 2 presses having a higher average cost per book. According to AAUP statistician Kim Schmelzinger, this is somewhat in line with other financial trends noticed among AAUP member presses: “There seems to be an inflection point on the growth path as presses reach the $3.5-$4.5 million mark where they become more efficient, most likely due to systems and processes being scaled properly for the output of the press.” This study did not permit us to definitely explain why this is, but there does appear to be a value of scale or strategy with the Group 3 presses.
- Scale? There was a working hypothesis at the outset of the study that larger presses would demonstrate a lower per-book cost, presuming that larger houses are able to work more efficiently due to the economic benefits of scaling. We did not see this in practice, as larger presses demonstrated higher costs, on average, per monograph. This came as a surprise to both the researchers and some directors. A few possible explanations were suggested: The larger presses carry higher costs because they can. The largest presses participating were located in metropolitan areas, where costs are higher. The financial health of these presses and their generous institutional hosts (including among the best-resourced universities in the country) permit them to take on investment in certain areas—staff salaries, marketing expenses, contributing to major overhead expenses like rent. In other words, a higher per-book cost in this case might be understood as signaling the ability to continue to invest in the business, not an inefficient use of funds. Or, perhaps the higher costs come from the infrastructure necessary to acquire better authors and to sell more copies of the books.
Table 4. Average Cost per title by Group. Shown by Breakdown of Cost Type and by the Three Different Definitions of Cost
|Group||Staff Costs||Staff Overhead||Direct Costs||Press-Level Overhead||In-Kind||Basic||Full Cost||Full Cost Plus|
Across all presses, the largest expense by far was in the staff time spent directly on monograph publishing, whether on acquisitions, manuscript editorial, design, production and marketing activities (Table 4). (“Press-level overhead” also includes staff time for those in press-level roles, such as legal, accounting, and so forth, so should those press-level roles be counted here, the percentage of salary and benefit related costs would be even higher.) The split of time spent on monographs by staff, versus direct costs is very similar across presses of all sizes. Although outsourcing of services is a common practice, direct non-staff costs are still modest by comparison to the staff’s salary and benefits.
Table 5. Average Staff Costs per Monograph, by Activities and by Group
|Group||Acquisitions||Manuscript Editorial||Production||Design||Marketing||Total Staff Expenses|
Range of Staff Costs by Activities
Direct non-staff costs reflect the Title Specific costs incurred in producing each book at each of the core departments. Each press reported by department and by title on the actual direct non-staff expenses associated with creating that book. Direct costs illustrate where outsourcing is taking place. By comparing the same set of activities across staff time and direct costs, we can begin to see where at the press certain functions are being outsourced. Tasks requiring specialized technologies, like printing, have long been outsourced. Relevant to this study we are also seeing many at least partially outsourcing certain elements of the publishing workflow, from copyediting to compositing to graphic design. While some may choose to entirely outsource a function, it is more common to see presses outsourcing some of the work, on a title by title basis.
It is best to assess the sum of staff time and direct costs to get a full picture of the costs of each activity. Manuscript Editorial, for example, often outsources copyediting, and Design departments may choose to hire freelancers to design interior and covers. Still, this more holistic view of both staff and direct costs, shows that acquisitions costs are still significantly higher than others, and this holds across presses of all sizes. (Table 6)
Table 6. Staff Cost and Direct Non-Staff Costs
(Note: This chart only includes staff time and direct costs, by department, and not staff time overhead as in the BASIC definition. The charts below are intended to illustrate relationship of staff time spent on monographs, relative to direct expenses.)
Direct Non-Staff Costs by Department
|Group||Acquisitions||Manuscript Editorial||Production||Design||Marketing||Total Direct Costs|
Staff and Direct Costs by Department
|Group||Acquisitions||Manuscript Editorial||Production||Design||Marketing||Total Staff & Direct Expenses|
Acquisitions is the most expensive activity. Among the five core publishing activities assessed, regardless of press size, the greatest share of staff cost is spent on Acquisitions activities, with Marketing and Manuscript Editorial activities vying for second (Table 6).
The largest presses have higher costs for the acquisitions work than do the presses in groups 1, 2, and 3. This does not appear to be due to higher numbers of people doing the work – the larger presses reported fewer people contributing to acquisitions per title than many of the smaller presses. It may be due to a few factors contributing to higher salaries, including that the largest presses are in metropolitan areas, and that they are more likely to offer attractive salaries to recruit potential editors.
Of the types of acquisitions work, the most time consuming (thus most expensive) tended to be author communications and support, followed by developmental editing, and the peer-review process. Acquisitions editors described working with authors, or potential authors, over the course of many years, sometimes meeting them as graduate students, or seeing a project in very early stages take shape over time. The peer-review function was described as especially time consuming, particularly from less senior staff, who are often tasked with identifying scholars willing to conduct the review, and coordinating the process.
Selection and taste are difficult to define, but central to the role of acquisitions. Acquisitions editors exert editorial judgment in finding those manuscripts that fit the press’s mission. While some of the smallest presses make a limited investment in acquisitions, acquisitions editors of the larger presses described a major investment in immersing themselves in the fields in which they specialize. As one acquisitions editor from Group 3 described, “A common misperception is that we are simply reviewing things that come in through the door. That is not what we do. We are actively out there recruiting projects. A heck of a lot of our time goes into our “never published” category. Some of the things I am working on today will not be published until 2020.”
While personal taste certainly plays a role, the final decision to publish at a university press depends on the work passing through a process of peer review.Identifying reviewers for a work is a complicated and time consuming task. Editors research the field by reading articles, asking the author, and using their own contacts, to build lists of potential reviewers. When a press ventures into new areas, developing lists of these contacts, and new reviewers is a substantial time commitment. For acquisitions editors, a list of trusted reviewers is an invaluable asset. The peer review process itself can be time consuming; editors and their assistants must choose to send out the book, invite the reviewers, send out the manuscript, and then wait for feedback, which can take months. Some reviewers, not surprisingly, are better and more comprehensive than others. Some reviews come back too lightly considered, or too negative. Some editors spoke to us about reviews that are “more about the reviewer than about the book.”
Acquisitions Table 1. Acquisitions BASIC Costs (Group Averages)
|Group||Staff Time||Direct Costs||Total|
As with Acquisitions, a major cost in the Manuscript editorial function was author communication and support. One production editor from Group 2 noted, “You just oversee everything… the copyeditor and the author constantly cc you on everything. When the copyeditor sends the first run through to the author, you can go over it, to make sure it is going alright.” When viewed through the lens of value-added and competitive advantage, a publishing enterprise well-known for effective author relations may have a definite advantage over its peers. Most departments indicated that author relations was one of their most time-consuming – and by extension their most costly – activities.
Copyediting and proofreading reside within this function: copyediting is the process whereby a manuscript is shaped to conform to or meet house style standards and ensures that grammar, punctuation and spelling are correct and consistent. The copyediting function may also be charged with ensuring that headings, data, references, footnotes and other content elements are correct and consistent. Copyediting was among the most commonly outsourced functions at the press. This is reflected in substantial costs appearing as direct expenses.
Manuscript Editorial Table 1. Manuscript Editorial BASIC Costs (Group Averages)
|Group||Staff Time||Direct Costs||Total|
Changes in technology and the advent of XML workflow processes mean that display markup and composition—annotating documents in preparation for software transformation—can readily be combined with activities that have traditionally been undertaken at the copyediting stage (in what is considered file preprocessing). Interior design, once aimed solely at providing a quality reading experience, is now linked to the need for distribution across multiple formats and platforms, serves many masters, and because of the use of templates and the leverage provided by technology can be performed in conjunction with many of the Manuscript Editorial and Production functions previously described.
Cost drivers for design, according to staff interviewed, included complexity issues related to illustrations and supplemental and other materials. Heavily illustrated monographs were consistently and often cited as using up vast resources in the preparatory stages. Because of digital portability, the need for marketing activity in support of titles, and much (real or hoped-for) crossover between monographs and the rest of publishers’ lists, full-color covers are now the norm for scholarly monographs. Much of this work is outsourced and performed by freelancers, with staff who are not performing cover design functions managing the work of others. Presses see design as a key differentiator and of real appeal to both authors and audience. A deputy director in a Group 3 press commented, “All of our books get the same design treatment… even books in series get the same effort, they all go through that process, not just the trade books. We will spend as much time working on a…monograph as working on a frontlist trade book. The distinction doesn’t apply to the covers. We have a reputation in the academy of being good at this, and it helps us to attract prospective authors.”
Design Table 1. Design BASIC Costs (Group Averages)
|Group||Staff Time||Direct Costs||Total|
Production functions in a traditional sense include digital file preparation activities such as managing fonts, images, and templates, the preparation of photos and other images, and the like. Digital asset management activities, or the processes and systems used to find, organize, convert, and store large content collections, typically reside side by side with digital file preparation activities. File preservation and file distribution activities, while often performed by third parties, must be managed, and that management occurs here.
In many presses, Production and Manuscript Editorial processes described above are merged in a single department. Because of the compressed workflow timelines and fluid relationships between these functions, those activities we think of as Production activities in the traditional print publishing environment are often undertaken by the same staff members (regardless of the department’s name or reporting lines).
Throughout the Editorial, Design, and Production (commonly referred to as EDP) functions, we observed the greatest amount of outsourcing. Almost all (if not all) presses engage in some amount of freelance outsourcing in these stages.
While work may be outsourced, there is still plenty of time devoted to project management, as press staff manage the freelancers and their work. Quality and high production and design standards are seen by many presses as real differentiators. Areas where they add value and are able to compete for top manuscripts in their fields of specialization. The need to control quality – particularly with respect to aesthetics – was expressed by many presses to be a binding constraint. Several interviewees noted that they would not embrace a fully OA environment, or even a fully digital environment, because they perceived that this would compromise the visual aspect(s) of the content.
Production Table 1. Production BASIC Costs (Group Averages)
|Group||Staff Time||Direct Costs||Total|
Marketing departments mentioned repeatedly their early and ongoing involvement with many activities aimed at promoting monographs. Pre-publication activities aimed at publicizing, or creating talk about and therefore consumer demand for monographs; the creation of detailed advertising plans, copywriting for seasonal and specialty catalogs (as well as for jackets and covers). Marketing staff engage in extensive planning aimed at generating review coverage. Detailed award submission plans are put together, as well as plans for exhibiting books at conferences and exhibits. Many of the direct expenses associated with these activities have been aggregated and included here as departmental overhead, but the pre-publication staff time was directly captured during staff interviews at the site visits. Seasonal catalogs, though expensive and time consuming, are still very much part of the marketing process.
Title-specific web marketing has become quite important. These activities—routinely undertaken for monographs and trade books alike—include such activities as managing authors in creating book websites or blogs; generation of email newsletters and promotional emails; web-based publicity for review generation, tracking of web mentions, and the like. Search engine optimization and metadata creation and distribution have become critical tools in pre-publication content marketing.
Among the things that tend to drive the cost of marketing work at the participating university presses are that there are multiple streams of work taking place in each department, and each member tends to touch every book in some way. (The exception to this tends to be people in publicity roles, who might focus on certain lists.) The specialized roles within the department include Conferences and Exhibits, Web-based marketing/Social Media, producing the annual and disciplinary catalogs, and advertising. Rarely is this work outsourced. As one press director at a Group 3 press described, “Acquisitions is core to the press. It is the management of the selectivity, relationship building. The other core function is strategic marketing, with the emphasis on strategic. It’s making decisions about understanding particular fields and making appropriate marketing choices. Everything else can be outsourced.”
Author relations, also seen in other departments, can demand a great deal of time from staff. Authors, even authors of academic monographs, might have expectations for reviews appearing in major mainstream publications. Marketing staff are often the front line of discussion with authors, to get the most from their expertise and existing contact base and to manage their expectations.
Marketing Table 1. Marketing BASIC Costs (Group Averages)
|Group||Staff Time||Direct Costs||Total|
Staff-level overhead includes the time staff in the core department reported working on monographs that were not published and on other activities. While the former is almost entirely borne by the Acquisitions department, this activity on occasion may have an impact on some other roles, as well, such as marketing, typically called in to evaluate or even develop plans for prospective titles. (Table 6)
Time spent on books not published was often characterized as a critical step in the path to discovering the best quality works that the press eventually does sign and produce. Staff discussions yielded some important context for understanding these figures:
Senior staff reported that acquisitions is more of a 24/7 job, with pitches and new ideas coming to them not just in the workplace and at conferences, but everywhere, including as one press director noted, “in line at the grocery store.” The deluge of ideas, well before they are filtered through the editorial lens can equal significant time spent.
However, it is not just the senior editors who report much time spent on seeking new work. Junior staff, whether assistants or editors, spend a great deal of time developing new leads, both for authors and for reviewers. Should they decide to enter a new discipline, or a field that crosses over multiple disciplines, significant time is needed to identify the right people, and encourage them to work with the press.
Indeed the more senior staff tended to spend less time on manuscripts that were not published. Some even noted that as a list becomes better known and better respected, it is less likely to receive manuscripts submitted in error, for example, or those that are just a bad fit, as authors are more attuned to the requirements of the list. The time spent on books that don’t come to fruition tends to be at the early stages of possible publication; while editors report winnowing a list from 100 to 1 up front, once they have identified the projects they feel are viable, the success rate increases significantly. By the time a project makes it to the editorial committee/board, many acquisitions editors reported there may be close to a 90% chance it will make it through to publication.
Table 7. Staff-Level Overhead: Percentage of Time Spent on Books Not Published
|Group||Total Overhead Costs Staff Time||Staff Time on Monographs Never Published||Staff Time on Other Activities||Cost of Acquisitions Time||Time on Monographs Not Published as % of Acquisitions Cost|
In order to develop an accurate picture of the full costs of publishing monographs, it was necessary to include some of the costs of running the business, even though not directly attributable to producing a specific title. It was important in this study to take overhead into account; editing and printing a book may be more directly relevant to publishing a new title, but no publishing house exists without support in areas like legal, accounting, and other necessary services.
University presses may have non staff-related expenses that are managed at the departmental level, for example, travel and expenses for acquisitions editors attending conferences and costs of marketing efforts that are otherwise not broken out by title. The data below reflect averages by group. Note that in some cases, presses did not indicate any costs for certain departments. This may not necessarily be because they do not spend money on these activities, but rather that they were able to assign costs directly to the titles themselves, or it might be due to the way the press’s departments are organized. A press that has a combined Design and Production department, for example, may have reported the departmental cost on just one of those lines. Among the responses concerning departmental expenses for acquisitions were: business entertainment, travel, receptions, readers’ fees, free books and postage, training, memberships and telecom.
Table 8. Average by Group of Departmental Overheads, by Department and by Group
Marketing was the only department for which we captured specific line items for Departmental level overhead costs. The categories requested in the questionnaire include those shown in Table 9.
Table 9. All Presses Breakdown of Departmental Overheads: Marketing Detail
|Email & Social Media||Website||Advertising & Promotion||Seasonal Catalogs||Conferences & Exhibits||Other|
General and Administrative CostsIt is not uncommon for a press to develop a method for allocating a share of its overhead costs to the products it creates, however, there is tremendous variety in which costs a press actually pays itself, and which are covered by its host university. Almost all the presses reported paying directly for accounting; many also listed finance positions. HR, however, was often an in-kind contribution at small and medium-sized presses. Very few presses listed costs for legal support, with the exception of one press with a strong focus on art and art history which pays a substantial sum for legal staff. Among the smallest presses, IT staff consisted of contributed staff time at all five participating presses. Almost all the other presses reported paying for IT support directly, but with significant increase in cost from Group 2 (average = $14,743), Group 3 (average = $75,856) to Group 4 (average = $527,486).
Table 10. General and Administrative Costs
|Group||G+A per Monograph||Full Cost per Monograph||G+A as a Percentage of Full Cost|
It may seem obvious that if a press does not need to pay for something, it should not be factored into any calculation of cost, but there are two important reasons to take these costs into account. First, things change. What the university or library might cover one year, might not be included the next and having a clear notion of what the actual resource needs are is important. More to the point of our current study, some of the costs that are not paid by presses today, but that are integral to the book’s quality, are routinely covered by the author, often by grants and other funds the author is able to obtain.
A worthy example exists in the case of indexing. While most presses and authors still feel this is a worthwhile activity, few presses today are paying for it themselves. They may provide guidance to authors, or may edit the authors’ work, but it is the authors who take responsibility for hiring someone, or doing it themselves. A similar case exists for image rights. Presses expect authors to deliver manuscripts with rights cleared. These costs can vary widely, depending on the number and type of images or other rights needed.
This is also the case for image and other rights and permissions clearance. Presses routinely expect authors to have done permissions research, and to pay any related rights fees as part of the completed manuscript package they deliver to the publisher. Yet, we also heard numerous times just how time consuming the work of press staff is in vetting, guiding, and correcting the work done by authors in this activity.
An examination of the least and most costly titles offers a taste of the data at a more granular level. While average page count across the full sample was 296, the most costly titles tended to be much longer works, with page counts well above the average (356 pages on average for the top 5% titles). In Table 11 below, the most costly monographs had an average illustration count of 99 images (or tables or charts) per book, while Table 12 shows that the least costly had an average of 24 (and almost half had none). That page count and number of images/illustrations/tables drives cost is unlikely to come as a surprise.
Table 11. Highest 5% of Titles, by FULL COST
|Group||Discipline||Pages||# Illustrations||Cost from Staff Time||# People Who Worked on Title||BASIC COST||FULL COST|
|2||African American Studies||200||19||$34,246||14||$50,578||$68,506|
|3||African American Studies||238||0||$47,834||17||$61,982||$71,624|
|4||Art & Art History||184||12||$43,423||19||$54,685||$76,215|
Table 12. Lowest 5% of Titles, by FULL COST
|Group||Discipline||Pages||# Illustrations||Cost from Staff Time||# People Who Worked on Title||BASIC COST||FULL COST|
Considering the averages as well as the extremes prompts a slew of further questions, as well: How strong are the correlations between page length and cost, or illustrations and cost? What role—if any—does discipline play? While the sample size, at 382 titles, is neither a particularly large sample nor completely representative of all university presses (or of all academic publishers in general), these data did allow us theorize about—and test—the strength of some of these connections.Candidates for so-called “predictive variables” emerged from several sources. The data themselves, of course, suggested several possible areas for further analysis. During our interviews with press directors and staff, book-specific factors such as number and complexity of illustrations, whether authors were publishing their first book or were farther along in their publishing careers, were mentioned as important factors in driving costs. Discussions with our advisory board members highlighted other potential types of cost drivers; some of the likely variables discussed included regional cost-of-living, public or private ownership, parent institution subsidy, press size, the number of monographs as a percentage of a press’s book publishing program, and the number of staff.
Below are some findings from this work examining correlations with various factors:
Page count. There is a moderate, positive relationship between the number of pages per monograph and the basic (and full) cost per monograph.
Illustrations. There is a moderate, positive relationship between the number of illustrated pages per monograph and the basic (and full) cost per monograph. While our count included all sorts of illustrations—black and white, halftone, and color images, tables, charts, and figures—EDP staffers frequently cited this as a cost driver.
Disciplines. Our sample was not structured to include a representative sample of different disciplines, but we tried to see what we might find by grouping the disciplines represented into groups that reflected possible cost drivers. Table 13 shows average cost for titles in each discipline listed, using the BASIC cost definition.
Table 13: Average Basic Cost per Title by Discipline
First Book? Do first books, cost more to produce than books by more experienced authors? In this sample, we found that there was no statistically significant difference between the cost per monograph of “first books” versus other books.
Table 14. Comparison of the Basic Cost of First Books and Not First Books
|Count (n=342)||Average Basic cost|
|Not first books||172||$28,997|
In addition, we examined characteristics of the presses themselves:Press size. Average cost, by press size demonstrated that the smallest presses have the lowest average costs per title, and the largest have the highest costs. But a closer look at the numbers suggests that this is not a simple progression. That Group 3 presses have lower costs than Group 2 presses led some in our advisory group to suggest that there is in fact an element of efficiency at work here; the Group 3 presses have perhaps reached a scale at which they are able to produce books more efficiently, through higher title count per staff and some outsourcing. Why Group 4 presses, then, do not demonstrate even lower costs is unsure, but this may have something to do with the range of activities they engage in and how they chose to report their costs.
Table 15. Average Cost per Monograph by Press Group
Press Location. Presses in the Northeast and South had significantly higher average costs than presses in the Midwest and West. It is also the case that four of the five Group 4 presses are in the Northeast. Explanations for the high cost of presses in the South is less obvious.Rent – no Rent. There is no statistically significant difference between the cost per monograph at presses that pay rent and the cost per monograph at presses that don’t pay rent. Average Basic cost per monograph at presses paying rent was $31,066 and was $27,465 at presses not paying rent.
Public v Private. There is no statistically significant difference between the cost per monograph at presses at public versus private institutions. The average Basic cost per monograph at public institutions was $28,090, while the average cost per monograph at presses at private institutions was $29,883.
Authors. Perhaps the subtlest and most interesting cost driver cannot be discussed using strictly quantitative terms. While speaking with staff about what they felt made one title more time consuming to produce than another, staff in every department in every press, came back to the author. When staff allocated additional time to working on a particular book, it was often because they could recall—often in some detail—the difficulties of working with a specific author on a specific project. This is not to discount the very real costs associated with complex production issues like illustrations, charts, or page count, but the theme arose so often that we feel it is worth exploring here.
Authors themselves, without regard to subject matter, can be significant cost drivers, based on the time they require at every step of the process. This may be because they are new to the publishing process and require a good deal of hand holding along the way (one manuscript editorial staffer described a major value in the work she does as “training scholars to be published authors.”) A less appealing scenario occurs when an author is unwilling or unable to hold to the format or deadline requirements of the press. This can result in press staff spending time managing the situation internally, and can also trigger higher external costs, if for example, an author insists upon making significant changes once a manuscript has already been to the compositor.
Two factors make this situation particularly difficult for presses to manage. The first is that it is harder up front to assess the likelihood of the author creating problems for the presses than it is to analyze the complexity of her manuscript. Second, staff at many if not all of the presses we spoke with, at some point, referred to themselves as being “author centered” and considered the care, time and access they offer authors to be a key differentiator. The degree to which this is actually the case might be difficult to assess, but is an area worth further exploration, as well.
It is clear that there is a very real and valuable set of activities that publishers offer, throughout the process of taking their original raw manuscript through to final publication. Whether offering guidance in developing the manuscript and its ideas, formatting and designing layout, or considering how best to make sure the work reaches the right audience, these steps involve many individuals with specialized knowledge and connections. Presses in the future might make this value more explicit, as a means of attracting authors; as some commercial presses already do, they might consider offering a basic service for those whose manuscripts come in “ready to go” and add-on services for those desiring (or requiring) additional support. Or they might consider what the sweet spot might be between a fully hands-off template model of publishing, and one that offers a deep level of engagement.
Data gathered from the twenty participating presses suggest that monograph publishing is considerably more expensive than has often been reported anecdotally or in other studies, and certainly more expensive than current price points for publishers with OA models would suggest.
The source of these costs is largely staff time, and specifically the time related to the activity of acquiring content, an area closely tied to the character and reputation of the press. This function is the least likely area to be outsourced at a press, and the most closely tied to its financial success: the acquisitions role, after all, is to bring skill, subject expertise, and relationships to bear in selecting and developing the most promising authors and topics to the press.
Another major cost is the overall operations of the press itself, the press-level overhead expenses. These figures are important to consider, since at the most basic level, they are integral to how press functions as a business. And yet, there is still real debate about the best and most equitable way to assign those costs at a product level. Further complicating the question is the underlying reason for the costs themselves: should higher per-book costs be interpreted as a sign of press inefficiency, or a reflection of a healthy press, in a position to devote and invest greater resources to its work?
These questions get not just to the heart of what it costs to produce a book, but what it costs to run a university press. Though this study did not take on that particular question, it certainly did end up pointing up some important issues worth further investigation.
For most university presses, monographs are rarely profitable on a per-title basis. Some may achieve sustainability at the “list” level—a line of books, whether by discipline or by product type (monograph or trade books, for example) or perhaps at the level of the “books division.” Further complicating this, certainly, is that while the financial success of certain books might end up subsidizing the weaker sales of others, choosing the winners in advance is just not possible.
While the study did not address the presses’ financial performance, it did appear that presses that report being on good financial footing tend to be larger and have multiple streams of revenue that end up cross-subsidizing the monograph list. For some presses it may be a journals list, like Chicago’s; others, like Yale University Press, have a strong textbook program; University of Minnesota Press has a professional testing program in its portfolio. To more fully understand to what extent this is the case, a fuller assessment of the economics of the university press—the whole press—would be needed.
This study suggest some important next steps for university presses, as well as some open questions. How can these data be used to develop pricing models for open access funder subventions and publisher fees? Now that we know what it costs to produce and distribute a monograph, what (portion of the) costs will publishers need to obtain as an up-front subvention, in order to make the book openly available at publication? Will publishers look to recover full costs, or be able to develop estimates that model just the revenue forgone with OA models? In other words, is the basic, full, or full plus cost the one which publishers will seek to recoup?
Unlike the retail book market, where publishers establish prices for the products they sell, it is not entirely clear where this pricing model will develop, or what figures will take hold. Who, in the end, will be setting the prices: presses or funders? How might a change in funding model result in changes to the publisher landscape? Will new, institutional sources of support have the impact of incentivizing other, newer entrants into the field of scholarly book publishing? And will new entrants into the publishing field find ways to offer some, if not all, of the current list of services at prices lower than those publishers can support?
We heard from press directors and editors just how difficult it is to accurately guess which titles will be the bestsellers and which will underperform. How will this uncertainty influence the way that publishers make choices about implementing an OA model? Will they act cautiously, considering which books are likely to sell few copies, and therefore require the subsidy? How will they place those bets, and might it have the unfortunate effect of keeping the potentially most valuable material tied to the pay-for-access model?
While there has been reference to publishers’ ability, in theory, to continue to monetize the monograph content in other ways, alongside an openly available digital version, how will this work, and what sort of revenue is likely? What licensing would be needed to insure continued revenue generation from other formats? How can publishers model future revenue across a monograph list, across the press, given certain titles most likely to become OA? What will happen to subsidiary rights and permissions income? What are the prospects for continuing to sell premium ebook versions and print copies?
Libraries and their patrons stand to gain quite a lot should open monographs become the norm. In the short term, this would remove a certain category of cost from the library collections budget (though this might be a short-lived benefit, should university-based models end up shifting funds from acquisition to subvention). However, some existential questions about the collection development role also arise in an environment where many monographs are OA, questions that also challenge the business models of the vendors that supply libraries. Today, while multiple channels exist to carry, distribute and otherwise disseminate content to libraries, the primary routes of this distribution are premised on a sales model, where vendors, aggregator platforms, wholesalers and retailers each expect to see some gain from the sale of a book. Absent the notion of “sales,” how robust will the notion of “distribution” and therefore discovery be? It is likely that fee-based models will emerge to support hosting OA content. Unless the resulting OA books are as easy to identify, find and access, cataloged and indexed as their paid versions would have been, the aim of providing access to all will only be realized in theory.
There are some interesting additional points of entry for libraries as publishers in this space, as well: For those institutions already undertaking peer-reviewed publishing, we suspect new models of subsidy might encourage some library presses to further pursue monograph publishing. We hope this report is useful for those new to this work, in that it outlines the categories of activity that are currently taking place at the high-quality university presses that are publishing monographs. For those in libraries already supporting the scholarly communications process, we hope that the description of the cost and value of the many stages of work that university presses currently undertake will be helpful in highlighting the degree of investment currently being made by established presses. New publishers or publishers-to-be can use this report as a way to consider all these phases of work, and to determine which to take on, which to outsource, or which they may choose to forgo.
In the months ahead, publishers will be considering not just what it costs to produce a monograph, but how much they would need a subvention to cover, in order to make that book available in an Open Access model. Our findings suggest that the “basic” cost (including staff time and other direct non-staff costs in the five functional areas of the press) represents the bare minimum needed to develop, create and disseminate new works. We would also suggest that presses examining this for themselves look closely at what we have called here “departmental overhead.” While for our purposes, it was not possible to fully break this out, there may well be elements that presses will want to include as well.
We cannot estimate costs for models and book formats that do not yet exist, but it is worth mentioning that we are at a moment of experimentation with digital formats, which suggests that in the future, even “simple” monographs are unlikely to consist of just an uninterrupted text. Innovative projects underway today include the University of Minnesota Press’s Manifold Scholarship; the efforts of the University of Michigan Press and its partners to publish humanities monographs linked to rich media primary source materials using the Hydra/Fedora software framework, and Stanford University Press’s initiative to create a “peer-reviewed process for …interactive scholarly research projects.”
For monographs to have significant reach and impact, absent a sales-based distribution channel, where will the books “live”? How will readers and institutions know they have been published? Aside from the work of marketing teams, the act of distribution via sales channels, it is assumed, would need to be replaced by something similar. Would today’s aggregators want to host open content alongside paid content? If not, where would the OA titles be hosted? If so, what would the hosting cost be? These are issues that have critical implications for the discoverability and impact of long-form digital scholarship.
The humanities and social sciences Open Access movement has been stalled for a while in the United States, at least, with university presses uncertain of the payoff, authors not sufficiently demanding the change, and little funding to support the new model in the system. The success of this author-side model will lie in the ability of funders, and ultimately the universities, in continuing to support this into the future. Studies also funded by the Mellon Foundation at Indiana University and the University of Michigan and Emory University suggest that provosts may be willing to financially support a producer-pays model for monographs, although many questions remain; for example, about what will happen at smaller institutions and for faculty not on the tenure track. The Lever Press collaboration was recently formed to introduce a “platinum” OA business model for the arts, humanities, and social sciences, with a commitment to issue works that will be both free for the author to publish and for the researcher to read, suggests a somewhat different set of opportunities and challenges.
Imagining a landscape where all new scholarship is quickly and freely available to anyone with an internet connection is deeply appealing. Some of the questions concerning the mechanics of how to produce works of high quality, with digitally-enhanced features and functionality will continue to evolve in the months and years ahead. Some of the economic questions may be helped along with the data in this report and others like it.
But the questions will not all be financial ones. What do we see today as the real value of publishing? If it is defined as a means of quickly disseminating current research in a timely fashion, certain production-minded OA models are likely to prevail. They may provide a quick and efficient means of getting information into the world, and they may focus on certain production and dissemination activities primarily.
If, however, there is real value in publishing as an act of curation, selection and author development, then most of the activities included in this report must continue to be supported. The value of creating not just one book, but a sustained contribution to the development of a discipline through developing works that advance the field is something editors feel is a mainstay of their work. Publishers also provide important support to authors, whether first-time writers or not, by framing their work in the context of broader disciplinary trends, identifying constructive peer reviewers, by polishing their prose, packaging it, and positioning it to reach its intended audience, and generally in supporting their work in turning a manuscript into a polished final monograph.
What models will emerge to financially support these high-quality, freely-available digital monographs? Many issues remain to be resolved not least of which concern how publishers balance the different business models emerging in the OA space.
Perhaps it is possible to have it both ways, after all. It may be that we are at a moment when the line between “scholarship distribution” and publishing—implying a deeper investment into an author and her book project—must be more clearly defined. Unpacking the core elements that generate value—value to authors, to readers, to the academic disciplines – is likely to be the subject of debate and discussion as publishers and those they publish with and for must work together to create a sustainable ecosystem of high-quality, easily accessible, scholarship for all.
Peter Berkery, Executive Director, AAUP
Brenna McLaughlin, Marketing and Communications Director, AAUP
Barbara Kline Pope, Director, The National Academies Press
Darrin Pratt, Director, University Press of Colorado
John Rollins, Chief Financial Officer, Yale University Press
Mark Saunders, Director, University of Virginia Press
Rebecca Schrader, Director of Finance and Operations (retired), The MIT Press
John Sherer, Spangler Family Director, The University of North Carolina Press
Charles Watkinson, Associate University Librarian for Publishing and Director, University of Michigan Press
Baylor University Press
Colorado University Press
Columbia University Press
Johns Hopkins Press
Indiana University Press
Northwestern University Press
Rutgers University Press
Texas A&M University Press
The University of Chicago Press
The MIT Press
The University of North Carolina Press
University of Arizona Press
University of Arkansas Press
University of Georgia Press
University of Michigan Press
University of Minnesota Press
University of Nebraska Press
University of Virginia Press
University of Washington Press
Yale University Press
In developing this methodology, we encountered certain challenges, some of which we understood going in to the process, others of which became apparent as work progressed. Many directors praised the approach, in offering them an understanding of costs, particularly concerning staff time, that they did not have before. Some staff were less convinced, and shared their concerns—about the value of staff allocations in general, about our method in particular, about potential uses of the data. Below is a short outline of the issues encountered in gathering data directly from staff, and the steps taken to mitigate them.
- We designed our staff allocation tool to be completed on-site, by staff using desktop computers or laptops. This was a problem for several presses, whose staff did not have access to laptops. Even where presses were able to set up computers for our meetings, some staff were new to using them, and basic set-up took additional time. Where technology was an issue, we used a printed worksheet, allowing staff to write in their answers, which our research team entered into Excel sheets back at the office.
- The question of how to allocate one’s time is not an easy one in general, and asking staff to come up with accurate figures proved difficult. The basic question asked was quite simple: how would you say you spend your time on monographs (as defined in the study), versus work on other types of publishing versus time spent on other non-publishing things? Staff in certain departments—marketing comes to mind—were more comfortable with making estimates. In other departments, staff were less comfortable with this approach. A good deal of our on-site time consisted of group conversations, where as a group we discussed and debated what reasonable allocations might be. Doing this as a group setting was invaluable: if one person was able to offer their approach to the group, others were then able to consider how their numbers might be different. This group discussion helped to normalize the responses, at least at the departmental level.
- Gathering a comprehensive set of data required extensive follow up with each press. Not all staff could be there for our visits, and so our follow up process involved reviewing the staff lists with the responses received and working with our press contacts to make sure all data were recorded.
- Once the data were received from the presses, we reviewed and scoured it for anomalies. If most staffers in a department, for example, indicated spending little time in prior years on a title, but one staffer reported significant time in prior years, we might check with that person to make sure the response was accurate.
- In some cases, human error intervened. Three questions asked people to assign values to different books or activities, adding to 100%. The spreadsheets included a check field, showing if indeed the figured entered equaled 100%. Many sheets came back without reaching this figure. If the error was 1-2 points off, we simply recalibrated all responses to equal 100%. If, however, the gap was greater than that, we went back to the staff to ask them to have another look at their response, and to re-submit.
- The study definition of “monograph” was itself an issue of discussion. The exclusions (translations, edited works) did not map directly to what presses consider to be their monograph list. This posed some challenges, when asking staff to develop estimates for books falling into our definition of monographs.
- While presses were asked to submit 20 titles each, three presses had fewer than 20 titles. Two presses had such low title output that they ended up reporting on two years’ worth of data. In those cases, costs for the 2-year span were allocated to the monographs.
We gathered three categories of data, using different tactics:
- Staff time allocations
- Direct costs per title (“title specific costs”)
- Press-Level Overheads (G+A and Departmental Costs)
Staff time allocations
On-site visits included 90 minute sessions with staff in Acquisitions, EDP, and Marketing departments. The sessions involved a round-table discussion, followed by time spent filling out the allocations grid together. The discussions were framed in a way to guide the group through three main areas:
- The process that they engage in
- What they see as the elements of their work that tend to drive cost
- What they see as the elements of their work that deliver the most value to the publishing enterprise
Each staffer who responded to the study (406 in all) completed a Staff Allocation form. This was completed in Excel by most respondents. For some presses, without the ability to use laptops or otherwise complete the Excel forms, we permitted staff to handwrite in the responses, which Ithaka S+R staff later keyed in.
Direct costs per title
The non-staff costs directly incurred by each title were gathered at the press, and entered into an Excel workbook provided by Ithaka S+R, with one tab for each of the twenty target titles. Our main contact was asked to gather data on these costs (some did this themselves and others asked department heads to participate). The categories of direct costs were grouped in five core categories, corresponding to the key departments (Acquisitions, Manuscript Editorial, Design, Production, and Marketing).
To gather data on overheads, Ithaka S+R provided the Director and CFO with a sheet labeled “Press-Level Costs.” This included two sections. One section was for the press-level G+A expenses, such as legal and accounting. The second section was for department-level overhead. This section was intended to capture any relevant non-staff costs not already captured on the “title-specific” sheets. Indeed, the financial lead at each press was asked to make sure that the title-specific sheets were given priority over the department level expenses; all costs should first be assigned at the title-level, where possible. Any remaining expenses could be included them in the Departmental costs, which we later allocated per title.
To arrive at staff cost, we had press employees in each of the five departments estimate the time they spent in several ways. First, they were asked to estimate time spent on monographs in 2014, the year the study covered. Staff had a list of the relevant titles in hand, for reference, and were guided by a member of the research team in thinking about ways to best estimate time. They were also asked to consider time they might have spent on the 2014 list in earlier years (2012, 2013) and in later years (2015), and to assign percentages.
The calculation of staff cost involved adding the percentages of time people reported spending on monographs published in 2014. Each staff response was matched to the corresponding person’s salary and benefits. This cumulative percent was multiplied by the person’s salary/benefits to get a total cost for this person’s work, over several years, on 2014 monographs.
Staff were then asked to consider which of the 2014 titles they worked on took more (or less) time than others, based on recollections of the work they had done. Some reported spending the same amount of time on all titles; this might be the case for a marketing person, for example, whose role is to present all titles at conferences. But most were able to distribute 100% of their cost among the titles, based on which took more or less time. (Their resulting percentages were also displayed to them, by showing an estimate of equivalent days of work.) This total cost pool was then distributed across all the 2014 monographs they worked on. The cost of each staffer’s time on a specific title was added up across the press, resulting in a total cost for staff time on a specific title.
Staff were also asked to report on those activities they performed at the press, again, assuming 100% as the total amount of time they devoted to monographs. They were given a full list of activities to consider. Though presented in “departments” including Acquisitions, Manuscript Editorial, Design, Production, and Marketing, all staff had the opportunity to assign a percentage of time to activities in any department. This was necessary due to the great variety among presses in organizing the work, particularly of the EPD functions. For this reason, the figures offered for “staff cost” are best understood as the cost of staff time devoted to those activities, rather than the cost of the specific staff members in a specific department.
Note: Since many directors also do acquisitions work, we privileged their work in acquisitions, charging a portion of their reported time to that line. Their remaining time was then included in the G+A calculation.
Staff Overhead Costs
At any publishing house—at any office, really—people spend time on things other than their primary activity, here producing books. When reporting on how they spent time in FY 2014, staff were permitted to enter a percentage of time for “other” activities, including general staff meetings on things like health benefits, administration, and so forth. They were also permitted to enter a percentage for time spent on “books never published.” This was a category specifically introduced for acquisitions editors, who may devote a great deal of time to identifying new works. This time spent on “general” and “books not published” together comprise what we are calling staff overhead.
To arrive at a per-book staff overhead by press, the costs were pooled at the press level and then distributed across other all books and monographs published by the press in 2014, though the methodology is different for these two different types of costs. Staff costs on general activities are distributed across all books based on the relative time spent on monographs versus non-monograph publishing activities in 2014.
Time spent on “books never published” is calculated by pooling costs at the press level, then dividing it amongst other monographs only. Thus 100% of these costs are then divided amongst all of the monographs.
Costs on Activities
Each staffer who reported on time provided data on the percent spent on each activity related to their monograph work. We estimate the cost of an activity for a monograph by multiplying the percent of time someone spends on a particular activity by the cost of that person’s time on a particular monograph. For example, if one person’s cost for book A is $10,000 and 10% of her time is spent on author communication, that book has a cost of $1,000 on author communication from that person. These are added up across all people in the press to get a total cost by activity for each book and activity by department.
The title-specific costs were added up across all titles. The worksheet used to gather direct costs by department also included a line to enter any estimates of cost for in-kind contributions.
In this study “Press-Level Overhead” includes General and Administrative expenses as well as Departmental overheads. Departmental costs, where possible, were allocated to specific titles. For some non-staff direct costs at the department level, such as exhibits for the marketing department or travel for the acquisitions editors, presses were able to assign costs to specific titles. Where that was not possible, the cost was included as “departmental overhead,” added to press-level overhead and allocated as described below. Neither was particularly easy to allocate, as presses had a wide variety of methods in use. We explored many ways to do this, and three are described below. Only the third, the one based on “effort spent” is shown in the data tables for the report.
- An even split. The first method is to distribute Press-Level Overhead evenly across all books. Here we calculate all costs on the press information, tab, and total them. Then we divided it by the total number of books the press produced in 2014.
- Weighted by title. The second method uses weights. Starting with the total Press-Level Overhead, we took a percentage of that cost, based on the percentage of the total number of 2014 titles published that the 20 target titles represent. This new pool was then distributed across the 20 target titles based on the weights described in the Staff Level Data. These weights reflect relative time spent on each title. In this model, books that required more time (and expense) from staff are also charged a greater share of overhead. (The weights are generated by counting the total number of days staff spent on each target title. Then the percent of time spent on each specific title is the number of days on that title divided by the total number of days on target titles. If this number is 5%, the monograph gets 5% of the overhead costs for target monographs.
- Weighted by press effort on monographs. The third method is to determine the relative effort spent across the press on the monograph list, as a means of determine what portion of the overhead ought to be allocated to monographs, as opposed to other product lines at the press. We did this by calculating a percentage of time spent by all reporting staffers, and assuming that non-reporting staffers spent no time on monographs. The average time spent on monographs was used to determine the pool of G+A to be charged to the monographs list. This pool was then divided evenly by total number of monographs, to arrive at a G+A allocation per monograph for each press.
- Ithaka S+R used the categories established by the Association of American University Presses: Group1, presses with annual revenue for the books division under $1.5 million; Group 2, annual revenue for book division between $1.5 and $3 million; Group 3, annual revenues between $3 and $6 million; and Group 4 with annual revenue over $6 million. ↑
- While these averages provide a good sense of the range of costs involved in publishing, note that individual presses will want to develop their own figures. In the study, some presses reported elements as “departmental overhead” that might be best considered “basic” or direct costs. ↑
- John Thompson, Books in the Digital Age (Cambridge: Polity Press, 2005): “In the 1970s academic publishes would commonly print between 2,000 and 3,000 hardback copies of a scholarly monograph . . . Today many academic publishers say that total sales of hardback-only monographs are often as low as 400-500 copies worldwide” (93-94). ↑
- While focused primarily on the UK, the HEFCE sponsored report and annexes offer a very good assessment of the current environment and issues facing monograph publishing today. Geoffrey Crossick, “Monographs and Open Access: A Report to HEFCE,” HEFCE, January 2015, http://www.hefce.ac.uk/pubs/rereports/year/2015/monographs/. ↑
- Other initiatives announced recently include Getty Publications, whose open-content initiative aims to make available, without restrictions, as many of the Getty’s digital resources as possible. Johns Hopkins University Press has been awarded a grant from the Andrew W. Mellon Foundation to support the development of MUSE Open, a distribution channel for open access monographs through Project MUSE. Minnesota is exploring hybrid OA titles with print sales. ↑
- Brill: http://www.brill.com/about/open-access/publication-charges; Palgrave McMillan Open Access Books: http://www.nature.com/openresearch/publishing-with-palgrave-macmillan/publication-charges/; Taylor and Francis’s Routledge Press refers to a policy on “research monographs”: http://www.tandfebooks.com/page/openaccess. ↑
- NPG offers a comparative list of OA monograph fees here: http://www.nature.com/openresearch/publishing-with-palgrave-macmillan/publication-charges/ ; Ubiquity Press: http://www.ubiquitypress.com/site/publish/; Taylor and Francis Instructions to Authors (2013) https://s3-us-west-2.amazonaws.com/tandfbis/rt-files/docs/instructions+for+authors.pdf. ↑
- Jennifer Crewe, “Scholarly Publishing: Why Our Business is Your Business too,” Profession 2004, 25-31, http://dx.doi.org/10.1632/074069504X26368. ↑
- This appears to be the same range of costs in the current Round Two model, presented in December 2015 and launching in 2016. ↑
- Appendix E of Carolyn Walters et al, “A Study of Direct Author Subvention for Publishing Humanities Books at Two Universities: A Report to The Andrew W. Mellon Foundation by Indiana University & University of Michigan”, September 15, 2015. ↑
- Joseph Esposito, “The Natural Limits of Gold Open Access,” The Scholarly Kitchen, November 19, 2013, http://scholarlykitchen.sspnet.org/2013/11/19/the-natural-limits-of-gold-open-access/ . ↑
- It is difficult to directly compare almost any two studies’ figures, since the categories that researchers choose to include within the cost differ. Eelco Ferwerda, Ronald Snijder, and Janneke Adema, “OAPEN-NL: A Project Exploring Open Access Monograph Publishing in the Netherlands. Final Report,” October 2013. Also see Guide to Open Access Monograph Publishing (a basic primer, but includes table citing current OA monographs pricing from several commercial presses), http://oapen-uk.jiscebooks.org/oaguide/. ↑
- Rebecca Kennison and Lisa Norberg, “A Scalable and Sustainable Approach to Open Access Publishing and Archiving for Humanities and Social Sciences,” April 2014. ↑
- Raym Crowe, “AAU/ARL Prospectus for an institutionally funded First-book Subvention,” June 2014. ↑
- John Thompson, Books in the Digital Age (Cambridge: Polity Press, 2005), 84-85. ↑
- For a good discussion of accounting approaches, see Thomas McCormack’s article, “Book Publishing Accounting,” AAUP Wiki. ↑
- It is worth pointing out that the additional expense in the Group 4 press overheads, is somewhat idiosyncratic. One large press carries the cost of legal staff; another counts its digital division as overhead in monograph publishing, since its staff handle all metadata and formatting for monographs, as well; others pay significant rent. ↑
- We looked at this question in two ways, trying to capture both the time spent by staff in certain departments; and also time spent on certain activities, regardless of the actual department people were assigned to. (The two versions are very close; the department-based version may be a little lower, since some non-department affiliated staff, like directors, would not be included.) ↑
- Recent studies have demonstrated that peer review processes are a critical aspect of establishing trust. See David Nicholas, Anthony Watkinson, Hamid R. Jamali, Eti Herman, Carol Tenopir, Rachel Volentine, Suzie Allard, and Kenneth Levine, “Peer Review: Still King in the Digital Age,” Learned Publishing 28, no. 1 (2015): 15–21, DOI: 10.1087/20150104. ↑
- Other departments had much lower overhead costs and these were not broken out, but reported by the press as one aggregated figure per department. ↑
- Further detail on the method used in this study is found in Note on Methodology in this report. ↑
- Jason Weidemann, “Thoughts from Editors and Authors on What Makes a Good Manifold Project,” January 25, 2016, http://manifold.umn.edu. ↑
- “Hydra/Fedora Mellon Project,” Michigan Publishing, http://www.publishing.umich.edu/projects/hydra/ ↑
- Gabrielle Karampelas, “Stanford University Press Awarded $1.2 Million for the Publishing of Interactive Scholarly Works,” Stanford University Libraries, January 12, 2015, http://library.stanford.edu/news/2015/01/stanford-university-press-awarded-12-million-publishing-interactive-scholarly-works. ↑
- Michael Elliott et al. “The Future of the Monograph in the Digital Era: A Report to the Andrew W. Mellon Foundation by Emory University,” July 1, 2015, https://pid.emory.edu/ark:/25593/q4fd0; and Carolyn Walters et al. “A Study of Direct Author Subvention for Publishing Humanities Books at Two Universities: A Report to the Andrew W. Mellon Foundation by Indiana University & University of Michigan,” September 15, 2015, http://hdl.handle.net/2027.42/113671. ↑
- See http://www.leverpress.org/ for details. ↑
- “Economic Analysis of Business Models for Open Access Monographs: Annex 4 to the Report of the HEFCE Monographs and Open Access Project,” http://www.hefce.ac.uk/media/hefce/content/pubs/indirreports/2015/Monographs,and,open,access/2014_monographs4.pdf. ↑
- The one exception to this was Yale University Press, where the press chose to give us department-level data, rather than having staffers complete staff allocation forms themselves. ↑