Click here for CCC’s blog where this paper was featured as a guest-post

  Download this paper

Abstract

Entering into transformative agreements requires careful planning. There is an enormous amount of data to analyze for publishers, institutions and consortia. All parties need to collate and examine the data held on subscription holdings and APC payments for all parts of the institution or institutions. Negotiating the content, structure, and pricing of a transformative agreement requires all parties to be working against the same data. Unifying the data on holdings and open access payments can be burdensome when the data is unconnected. Introducing persistent identifiers to unite the information reduces that effort. Adding organizational hierarchies enables publishers and institutions to better understand where subscriptions are held and the rate at which authors within institutional departments, research institutes, and other connected units will elect to publish under gold OA models. Ringgold’s Identify Database contains over 500,000 organizational entities each with a unique and persistent identifier, the Ringgold ID, and highly structured organisation data, consortia membership information, and hierarchical relationships. Using Ringgold data enables the components of a transformative agreement to be mapped for each part of an institution, rolled up to the main organisation and then to the consortium for negotiation, saving hours of staff time. Ensuring that negotiating parties are working from the same data provides efficiencies for all concerned. This paper examines some of the issues facing publishers and institutions and the considerations they need to make when structuring and negotiating transformative agreements and where good data governance supported by Ringgold data and services can make that process easier.

Introduction

Scholarly communication is under increasing pressure to transform the publication of research output. As a community, stakeholders have been discussing ways to improve the access to quality research and to make it as immediate as possible. One of the proposed paths to this is the move to open access publishing. In the last year or so, funders of research, some public, some charitable foundations, have successfully pushed the agenda for this transition to happen more rapidly. Mandates on the permitted path to publication have been introduced, and funders are rapidly changing the landscape for research publication. For those funded by, or publishing the research funded by the generators of this shift, the options available are diminishing; open access publication in subscription journals will no longer be permitted, and the deposit of authors’ accepted manuscripts must be immediate. These mandates are not universal, but they are growing in number. Not all paths will suit all stakeholders, but the requirements of navigating the changes to come will require identifiable, structured data, and for that, Ringgold is positioned to help.

What do we mean by transformative agreements?

Transformative agreements are designed to provide a pathway to the universal open access of scholarly journal articles. They enable institutions to reallocate funds from traditional subscription access to provide open access publishing routes for their faculty. In turn, publishers are making the transition to open access in response to market demand and funder mandates, such as Plan-S. Transformative agreements are an important means of mitigating risk for publishers while providing good value, predictability, and efficient administration for institutions.

Transformative agreements can take several different forms, one of the most commonly discussed is the Read and Publish, whereby publishers include both access to their content and the ability for authors to publish fully open access in their journals (Gold OA). These deals can be with individual institutions or consortia. In many respects, this represents a bigger deal than what have previously been described as “the big deal”.

There are other models that satisfy open access mandates. Such as permitting the immediate deposit of the author’s accepted manuscript or the version of record (the published article) in an institutional or subject based repository (Green OA), or organized transitions in which libraries, funders or other parties pledge to continue to fund the publication after the publishers has flipped the content to OA.

The desired route to open access is variable by funder, geography and discipline. In other words, no one size fits all, and we may need to navigate multiple open access models for some time.

For the purposes of this paper, we are going to use the Read and Publish model as the basis for transformative agreements and examine some of the issues facing publishers and institutions and how they can be addressed.

What are the problems facing publishers?

The transition to OA presents a range of challenges to publishers, affecting small and large publishers differently due to economies of scale and/or discipline focus. The decision to enter transformative agreements is dependent on a lot of different factors, such as:

  • Author funding sources and mandates
  • Nature of the subscription customer base (type of subscriber and geography)
  • Nature of the author base
  • Quantity and increase in OA publishing
  • Impact on adjunct revenue streams
  • Predictions about changes in author, institutional and funder behavior
  • Forecasting when the tipping point away from subscriptions to full OA publishing will occur.
  • Full financial modelling and cost considerations.

We can distill this to a simple description: navigating complexity and uncertainty.

Many publishers have been working with subscription-based transaction models for the entirety of their existence, more recently, many have introduced the option of paid open access publication within subscription titles (Hybrid Gold OA). cOAlition S, a group of over 20 funders coordinated by Science Europe, has defined the principles of Plan S, which include a period of transition to universal open access publication. Under these principles, OA publishing in hybrid journals will not be compliant with these funders’ mandates. Johan Rooryck of cOAlition S described hybrid journals as “the end point for publishers, rather than the method of transition to full open access” as was intended. We are entering a new era whereby outside forces, be they funders, institutions themselves, or authors, are demanding greater change to the model of scholarly publication. However, the US government favors a Green open access route. For example, the NIH mandate for the deposit of author’s accepted manuscripts in PubMed Central to be made openly available within 12 months of publication.

Rejecting certain types of new models may appeal to some, and may suit certain circumstances, but the market shift is not going away. New models, and particularly multiple new models to suit differing funder requirements, necessitates thorough data analysis to create estimations and develop predictions on which to make decisions.

Applying proper data governance enables publishers to understand and assess each component of the changing system. Only from a clear understanding of the entities and their relationships with one another within this space can models be developed to mitigate risk, predict the future, and manage change. Just as a watchmaker must understand not only the mechanism itself, but the individual parts of that mechanism to create a timepiece, scholarly publishers must develop a deeper understanding of who is producing what, from where, which is funded by whom.

What do publishers need to know to plan a transformative agreement?
  1. Which institutions are involved, including departments and other sub-units. There may be multiple institutions engaged in a single transformative agreement with a consortium.
  2. By institution:
    1. What do they subscribe to, and at what cost
    2. Are these package deals or individual titles or a mixture of both
    3. Is there any variability in the content of package deals between institutions
  3. How many articles were authored by corresponding authors from the institution
  4. How many of the authored articles were open access (Gold) whether in hybrid or fully OA journals
    1. Are any existing arrangements made for the institution such as waivers, memberships, pre-payment of APCs
  5. How much revenue was generated by open access publishing
  6. Who funded each article published and what is their current policy on OA publishing
  7. Has the institution developed an open access policy

All of these pieces of information should be gathered for a period of at least the last three years.

Calculating the proposed deal:

To take a simple view, publishers are adding together the subscription spend from a given institution with the open access spend to give a price for publish and read deals. However, deciding on whether this is an unlimited publish and read deal, or a more restricted read and publish deal involves more analysis. Some of the considerations need to include:

  1. Subscriptions
    1. Rise or fall in subscription spend in the last three years
    2. Change in the subscriptions held
    3. Were there any journals with no subscriptions
  2. Gold open access
    1. Rise or fall in the open access spend in the last three years
    2. Change in whether this was in hybrid or fully OA journals
    3. Were there any hybrid journals with no open access articles from this institution
  3. Changes in institutional policy on open access publishing, or a statement of intent
  4. Funders involved, percentage of funders in Plan S or with other known mandates
  5. Disciplines involved (this can matter due to the lack of specific funding in some subjects)

By looking at the changes over time, forecasting the expected transition to open access publishing (extrapolation given the centralized availability, institutional and funder policies) it is possible to estimate how many articles will be published as gold open access and how much the “read” component of the deal makes up. Considerations can also include whether to include all titles in the “publish” part or whether some titles would incur additional fees, for example journals with very high rejection rates, or that operate in disciplines which would be better suited to a green open access approach and thus stay in the “read” part of the deal.

Ultimately publishers need to decide: What is going to be negotiable and what is not?

Of course, all of this is after the publisher has determined the best route for their particular publishing program, which requires a deeper assessment across all titles, all customers and all revenue streams.

What are the problems facing institutions?

Many institutions and consortia are interested in a move to transformative agreements, although for the time being, that interest is heavily concentrated in Europe. One of the concerns among institutions regarding a complete switch to gold open access is the impact on research intensive institutions paying for the majority of all article publishing. As financial contributions (subscriptions) from companies, governments, and teaching intensive institutions falls away, the burden on those generating the most output intensifies. One of the intentions of Plan S is to remove the involvement of authors in payment for publishing. This would necessarily lead to payments directly from the funder, or from the institution itself, in a more centralized manner.

On the whole, institutions appear to be in favor of open access publishing. However, the current system of subscriptions, APCs payments, and monitoring of deposits in repositories, is becoming time consuming and costly. Paying individual APCs has frequently become the responsibility of the library. With library budgets already stretched, a desire to move to a more unified approach is prevalent. Consortia have long been instrumental in helping members manage costs and expand access; now, that mission is carried forward by facilitating collaboration among institutions and publishers to support open access. The economies of scale and centralized administration provided by consortia makes them a natural avenue to negotiate and manage transformative agreements.

In a consortial environment transformative agreements present challenges. Most importantly, what price shall be paid by each member? This isn’t new, consortia that operate all-in deals have always had the challenge of distributing fees. The difficulty for consortia operating opt-in models is similar; in order to utilize their scale and centralized administration, consortia must treat those institutions participating as an all-in group. Many are ready to take on this challenge and have experience in these calculations. However, the debate about who pays for what (“read” or “read and publish”) within a transformative agreement will continue to exist until the transition to open access publication is complete.

In terms of negotiating transformative agreements with publishers, institutions and consortia are faced with similar challenges to publishers. It is effectively the other side of the same coin. The primary concerns will be affordability and value for money. For consortia, this may be across a diverse range of institutions with different needs and depths of pocket.

Institutions need to know the specifics involved in any potential deal, and not just regarding content, but also the extent of the licence to its constituent parts. For example, is the university hospital included? What about jointly owned research institutes or institutes that cohabit the campus but which are separate from the main institution? They also need to calculate what the expected research output is and how many open access articles are likely to be generated. So, the core question being asked is: Is this a good deal and is it affordable?

What do institutions and consortia need to know to map a transformative agreement?
  1. Which institutions are involved including departments and other sub-units. Are there any co-located organizations involved. By institution or department:
  2. What is subscribed to, and at what cost
    1. Are these package deals or individual titles or a mixture of both
    2. Is there any variability in the content of package deals
    3. If subscriptions exist within departments do they permit access for the department or the whole institution
    4. What are the overall entitlements, does the licence with this publisher restrict access in any way
  3. How many articles were by corresponding authors from the institution
  4. How many of the authored articles were open access (Gold) whether in hybrid or fully OA journals
    1. Are any existing arrangements made with this publisher, such as waivers, memberships, pre-payment of APCs
    2. Does the institution pay APCs on behalf of parts of the organisation that are not included in the subscription licence
  5. How much was spent on open access publishing
  6. Who funded each article published and what is their current policy on OA publishing, would they be willing to finance OA publication
  7. Are other parts of the institution, such as the research office going to contribute to the costs, either by diverting funder’s contributions or directly.

All of these pieces of information should be gathered for a period of at least the last three years.

Calculating the negotiation points:

Firstly, what are the objectives: to maintain spending; reduce spending; and/or benefit from the reduction in staff time involved in individual APC payments and subscriptions.

For most institutions, there is a desire not to increase costs until new mechanisms are found to fund open access publishing. It may be important to maintain access to the content currently provided and streamline the process for handling APC payments. Important points to consider are:

  1. Subscriptions
    1. Rise or fall in subscription spend in the last three years
    2. Change in the subscriptions held (was anything cancelled or added)
  2. Gold open access
    1. Rise or fall in the open access spend in the last three years
    2. Number of APC payments in each journal
    3. Current and likely future institutional policy on open access publishing
    4. Any predictions of change in author intentions about open access publishing
  3. Funders involved, percentage of funders in Plan S or funders with open access mandates

Institutions need to look at their current spending, both on subscriptions and APCs and what they predict the mix to be in the next three years. This will inform whether the provision for open access publishing is sufficient for their needs and whether it is inclusive of the titles in which their authors wish to publish. What should the proportion of the fees being attributed to the “read” part of the deal look like? Should it be included at no extra cost, are there titles which operate green open access due to discipline or general lack of funding for publication? Has the publisher ring-fenced any titles to be excluded from the deal? Are these fully OA journals? What impact does this have on any calculated savings on staff time on individual APC payments with the particular publisher.

Is this a deal that moves the relationship with the publisher to an open access environment? What are they laying out as the roadmap for transition? Is it affordable? Does it meet the remit of the institution on publishing research?

Research impact assessment

Institutions usually assess the impact of the research that their authors create, this can be the impact within academia and broader societal impact. As scholarly communication moves to open access publication, monitoring the success of transformative agreements in improving impact requires yet more data analysis. The Library and the Research Office frequently collaborate on the mechanisms for research impact assessment and will need to track open access publishing (whether gold or green) versus publication behind paywalls. Impact assessment can involve more than looking at the impact of the article itself. It may involve examining output by departmental or discipline or require data on collaborations between authors and their affiliated institutions.

Where has open access improved research impact? What is the nature of that impact? Are there differences between open access models? All of these components will require data, and the more it is structured, identifiable and granular, the better. Entering into any transformative agreement with that consistent data structure in place will feed into the impact assessments further down the line.

How does Ringgold data and hierarchies help?

The additional administration caused by transitioning to new models can equate to increased cost. Well-structured, and consistently identified data, is the best means of reducing pain points and supplying datasets for reliable and accessible analysis. As research is conducted by individuals who are supported by institutions, linking organizational and individual identifiers, should be provided as part of the official publication metadata.

The Ringgold Identifier, and supporting Identify Database, is uniquely adapted to the complexity of academic affiliations. The identification of individuals, and their respective institutional affiliations has always been desirable, but with transformative agreements is now more important than ever. For publishers and institutions to navigate transformative agreements they need to be able to unambiguously identify institutions, their subscription holdings and the corresponding authors of gold open access articles that are affiliated with the institution or part thereof. While submission systems have become sophisticated, complete with mechanisms to facilitate accurate affiliation management, there can be barriers to capturing information in a useful manner. In most cases, authors make every effort to properly identify their institution. They face choices, to affiliate at the main institution, a department, or college, hospital, satellite campus, or other division of an organization. When a non-hierarchical organisation identifier is used, or no identifier at all, the challenge to find the “main” or “central” institution of affiliation becomes much more difficult. Ringgold ID’s and full institutional hierarchies enable the connection of this data.

In the case where an author has made a self-affiliation with what is considered to be a subdivision of the “main” institution (often referred to as the ‘top-level institution’), the connection between the author’s specific choice (e.g. Department of Medicine) may conflict with a requirement to affiliate at a higher level (e.g. University of Washington). While the granular designation is preferred, resolving that level of specificity to the more general “main institution” requires hierarchical connections between nodes in that particular organization’s structure. The operation of making a more general affiliation is often referred to as a “roll up”- meaning, the specific designation is rolled-up to the “top level” or “main institution”. This is useful to publishers, institutions, consortia, and funders alike when performing analysis on where are submissions coming from, how is it being funded, and what traffic is generated against publications.

Capturing the deepest practical level of granularity in an organizational hierarchy is highly desirable. When organisation identifiers are assigned to the components of the hierarchy it is always possible to roll information up to a higher node, where organisation IDs only exist at the top level, it is extremely difficult to subsequently bring the data down to a departmental level. Once the respective elements are connected in a structured manner any needs for analysis at a deeper level are easier to satisfy. Ringgold holds metadata identifying the top level of an institution and can provide services to match subscribers and author affiliations to the correct level of the hierarchy and then provide the data with Ringgold IDs, metadata and the corresponding top level for the institutions concerned.

In transformative agreements the “roll-up” function becomes even more important. Publishers and institutions must be able to accurately identify all of the transactions that they have with each other, both for subscriptions and open access payments. They must be able to estimate the open access publication requirements for the whole institution. They must also be able to identify the access permissions for the “read” component of the agreement.

Furthermore, they must be able to identify any conflicts in the arrangements that they currently have, for example where APCs are not paid by the top level institution for a unit, such as a research institute or hospital which is included in subscription access to content, or where subscription access is not included but APC payments have been made centrally on behalf of that unit. All of these considerations will be important when negotiating the deal’s structure and price.

Conclusion

As we have laid out, transformative agreements may not suit every publisher or every part of the publisher’s journal portfolio, but where they do, the need for structured and accurately identified data is more important than ever. Publishers need to assess their market and all of the component parts of their revenue stream before entering into these arrangements. Once the decision has been made to pursue read and publish transformative agreements the need for clean, well-structured data is even more critical. Ringgold’s unique and persistent organisation identifiers enable the connection between subscription transactions and open access fees within the same institution. Ringgold’s hierarchies enable the roll up of all information to the main or top level institution for analysis, not only during the assessment and negotiation process, but into the future when the results of transformative agreements are analyzed both in terms of cost effectiveness and the effect that model transition has had on the impact of published research.

Rather than a single new model to contemplate, we are faced with several. All with differing transaction types, and reporting requirements. At the same time, the emergence of new publishing models does not eliminate the old. Some regions of the world are transitioning to OA as a function of funder requirements and mandates, while others are moving more slowly. During this period of transition, all models must be accommodated, tracked, and accounted for. As a result, the utility and necessity of well-formed data increases. The ongoing transformation of traditional models will be a driving force behind demanding well-structured data and access to rich metadata via connective identifiers. Only when participants are clearly identified can analysis be reliably performed against the emerging dimensions of scholarly communications. Implementing good data governance is the first step to provide stakeholders with a pathway to understanding the who, what, and where regardless of model. Ringgold’s services and supporting data is specifically designed to make that possible.

  Download this paper