The licensing battlefield: consortia as new middlemen between publishers, agents and libraries: a view from the continent

Based on a presentation given at the 2007 ASA Conference: ‘Policy, Pricing and Purchasing’ Consortia are everywhere nowadays. Every country and every region is now covered by at least one consortium. International co-operation is taking off as well, both formally and informally. Nevertheless there is a wide variation in the nature and size of consortia. One of the most distinguishing features is the role that national governments are assigning to themselves. Consortia are vulnerable to conflicts of interests that are more likely to occur when a government’s role is weak.The article deals with pricing issues that haunt the minds of librarians.The mixed feelings of librarians about usage-based pricing schemes are explained. An alternative business model is outlined that allows for differentiation and transparency in the pricing of digital products and that restores flexibility in purchasing decisions for libraries.

The title of this article, which began life as a presentation at the 2007 ASA Conference: Policy, Pricing and Purchasing, contains just the right amount of belligerency that nowadays haunts the minds of librarians when confronted with vendors, publishers and agents alike.
It seems that nowadays all libraries are organized and represented by at least one consortium, and there is hardly a country left without at least some consortial organization. Recently there have been examples of international co-operation between consortia in a more formal way. The UK, Netherlands, Denmark and Germany have founded Knowledge Exchange to bring together initiatives, capacities and expertise from the JISC, SURF, DEFF and DFG (http://www.knowledge-exchange.info/). In Southern Europe we have already had, for some years, SELL, the Southern European Libraries Link (http://heal-l.physics.auth.gr/SELL/). Consortia can take a bewilderlingly wide range of forms and I can imagine that vendors have difficulties understanding what each consortium stands for. Not only do they differ in number and type of organizations they represent, but they also show a wide variety of attitudes, internal organization and decision-making powers and procedures.

The composition of consortia
All these differences reflect underlying variations regarding the organization of higher education in European countries and the role that national governments play or do not play in the strategic development of universities and polytechnics. In those cases where national governments play a relatively important role, consortia seem to be better organized, to act more formally and to cover a wider range of institutions than in consortia in countries where institutions are more or less left on their own. There is some logic in this, because in the first case a government has its own objectives and is willing to pay a price for it. In the latter case institutions have to look after themselves and will tend to minimize both costs and organizational effort. So they choose wisely to work on a smaller scale and use already existing organizational platforms. Consortia are everywhere nowadays. Every country and every region is now covered by at least one consortium. International co-operation is taking off as well, both formally and informally. Nevertheless there is a wide variation in the nature and size of consortia. One of the most distinguishing features is the role that national governments are assigning to themselves. Consortia are vulnerable to conflicts of interests that are more likely to occur when a government's role is weak.The article deals with pricing issues that haunt the minds of librarians.The mixed feelings of librarians about usage-based pricing schemes are explained. An alternative business model is outlined that allows for differentiation and transparency in the pricing of digital products and that restores flexibility in purchasing decisions for libraries.

NOL VERHAGEN Director University Library University of Amsterdam
The JISC, FinElib and Bibsam are good examples of the first type of consortial organizations in, respectively, the UK, Finland and Sweden. My own consortium, UKB (in The Netherlands), is a good example of the second type. That is why, for instance, UKB only covers Dutch research universities (plus the Royal Library), and does not include the polytechnics, the independent not-forprofit research institutions, or the public libraries. The broader the membership of the consortium, the more likely conflicts of interest between groups of members will occur, and if there is no central body to reconcile these conflicts or to mediate and to interfere, the more difficult it is to stick together as a consortium.
The composition of the consortium is an important issue for vendors anyway. Universities have rather higher level and more international information needs than polytechnics, whilst research institutions wish only to access those journals and databases that are of prominent relevance for their specific discipline. That is, in fact, a good explanation why research institutions often do not participate in consortia, unless they are covered by government driven organizations such as FinElib.
Vendors must, however, be aware that there is always tension between individual institutions and collective bodies, whatever their nature. To give an example: UKB has, like many other consortia, developed a cost division model that reallocates the amount of money that is being spent collectively at each consortial agreement. It is important to remember that the total amount of money involved is still based on the print portfolios that universities had in the past -a past that is drifting away ever further. In theory, it is an absolutely justifiable model, because it measures very adequately the relative relevance of a specific publisher's portfolio for each university separately. But, of course, after reallocation of costs it turns out that some universities have to pay more and others have to pay less. Each time we calculate the outcome of the model for a specific purchase, at least one librarian will object the application of the model in this one case, because the outcome does not meet his or her expectations or preferences.
Many problems with respect to both selection of content and division of costs arise from differences in the nature and size of member institutions. For instance, if a consortium is dominated by one or two large universities with traditionally high levels of expenses for journals, these universities will see smaller universities with rather low levels of expenses getting access to the same amount of content for a much lower price and will ask compensation for their high expenses from the smaller universities.

Pricing issues
What causes these problems is what I like to consider the Achilles' heel of the current journals business model: its lack of transparency, its lack of flexibility and its adherence to former print spends. We are beginning to realize that our newly developed access and pricing concepts have their drawbacks. 'Big deals' are inflexible, in the long run expensive, and are squeezing out small notfor-profit publishers, who are going to pay the bill for the inability of libraries to step out of big deals or to manage their budgets via cancellations to journals that form part of big deal arrangements. Moreover, libraries become aware of the intransparency and incomparability of the pricing of big deals, internally within the consortia and externally between consortia. It is therefore worth taking time to consider pricing issues, the first issue being usage-based pricing and why libraries do not like it, and the second being an attempt to develop a new approach to pricing digital content.

Usage-based pricing
Both publishers and librarians talk about usage statistics as an indicator for pricing. In a 2005 report by JISC Collections Journal Working Group 'Business Models for Journal Content' (http://www. jisc.ac.uk/uploaded_documents/JBM.pdf), both parties expressed some interest in this pricing method. The JISC study confirmed earlier findings with respect to opinions amongst librarians on preferred business models. Of course, usage statistics can teach us a lot, and in some cases they can be used as a parameters for internal cost division, or for internal comparisons of the relative relevance of a specific portfolio for a specific library. But it can be argued that it is dangerous and inadequate to use usage as an important basis for charging libraries.
In the JISC survey, both librarians and publishers were asked about their main concerns, priorities and preferences regarding usage-based pricing. The outcome of the study was that Serials -20(2), July 2007 Nol Verhagen Consortia: a view from the Continent predictability is important for both libraries and publishers. For librarians reducing costs was a high priority. For publishers the highest priority was continuity, which is mainly perceived as continuity of cash flow, which, of course, could very well mean no reduction of costs for libraries at all. Remarkably enough, publishers mentioned simplicity -librarians did not. Indeed, I sometimes have the feeling that librarians do not consider complexity to be a problem. In other words, librarians tend to not take into account increasing complexity so much when they are considering pros and cons of specific arrangements.
In the study, usage-based pricing mainly took the form of pay per view (PPV). It seems rather easy to understand why publishers would like to have usage as a parameter for charging. It is simple and flexible, and indeed it looks like a perfect instrument for cost attribution, and making negotiations with libraries and consortia a lot easier. Since usage is always going up, it certainly guarantees financial continuity for publishers. Even for libraries these are, at first sight, welcome features. However, PPV as a pricing mechanism would create ultimate unpredictability for libraries, as even publishers understand. So what librarians typically do is look for mixed models that guarantee access to specific titles, the so-called core collection, and eventually apply pay per view to others, in such ways that the risks are limited.
However, these mixed models have a strong tendency towards complexity, both in their design and in their execution. Especially in a consortial setting, it will be almost impossible to agree on what should be the core collection and what the peripheral collection. One of the experimental models was that of PPV-converting-intosubscription. It seems to protect libraries against unexpected price increases, but it rather easily will cause a situation in which libraries have to pay the same amount of money as previously for their big deal, but have to manage a complex administration to monitor usage. Other publishers have offered 'token' programmes that safeguard against unexpected expenditures, but that at the same time can cause extra price increases in the next contractual period. So we can question whether usage-based pricing schemes actually do foster flexibility and cost reduction, or if it only looks that way.
Moreover, applying usage-based pricing schemes supposes that libraries/consortia as well as individual users are able to identify the value of usage, so that they can really choose if they want to read the article or not. It supposes also that libraries can attribute usage to specific users or groups of users, to prevent some users consuming the usage rights of others. To make it work we should create a system in which a decision to use has at least some financial consequences for the user -if he is not going to pay for it himself, he at least should be aware of consuming his usage budget.
To be able to ascribe value to usage we should know a lot more about usage than we actually do. At the moment, we do not know who is using what, if our user is a student or a staff member, how much of usage represents actual reading, let alone how often the information is actually (re)used for teaching, research and study. The paradox is that as usage becomes easier, not only the costs but also the value per usage seems to decrease dramatically. Most usage does not have any value at all -representing not more than a glance at the content. So the real value seems not to be so much in the information itself as in the availability and accessibility of the information.
For me, it is obvious that we should not take that road. Libraries are there to encourage and to enable usage, not to monitor and restrict usage. Even if we wished to and were able to, it is hard to imagine that we could find a fair way to divide usage rights among our users. The value of usage for different kinds of users cannot be defined in any imaginable way. So besides not improving transparency or reducing costs, this pricing mechanism seems not to be very realistic in its execution either.

Alternative business models
So if usage-based pricing is not a realistic alternative for existing models, which model is? Basically, we now have two different pricing models, one that is based on individual title-pertitle subscriptions, and the other that essentially is a subscription to a database. The price for a single title is essentially the same for all types of libraries all over the world, while the price for a database subscription shows a wide variation, mostly depending on the library's previous costs for print titles. In a way, the latter model reflects at least some of the differences between libraries, but it is far from transparent and it rewards libraries that in the past have been economical or that have been too poor to build a representative collection. Or that have just neglected their obligations to their users.
One could imagine a model in which several parameters are used to identify the relative relevance of an information product for a specific institution, where relevance can be translated in a sort of tiered pricing. I know tiered pricing has a bad reputation, but if we could succeed in making the model sufficiently sophisticated, it might perhaps become more fashionable. At the same time such a model would enable a return to more flexible arrangements and would bring back institutional selection as an element of collection management. Of course I am aware that tiered pricing is being used already on the basis of the Carnegie classification in the USA, and the JISC has developed its own well-known banding system. These classifications use, in one way or another, the nature and size of an institution as parameters. My proposal is just to add the relative buying capacity of the region in which the institute is settled as a parameter and to broaden the applicability of the model to every single product of a publisher. This would result in a limited set of parameters that together indicate the intrinsic as well as the monetary value of an information product, be it a journal, an e-book or a database, for each separate library. A well-considered system of discounts depending on the amount of titles a library is subscribing to would help to safeguard the progress in availability of information that has been introduced by the big deal model.
The underlying assumption is that the digital format of an information product permits a publisher to almost endlessly fine-tune the pricing of it to the needs and the possibilities of a customer in a way that will be felt justifiable by all other customers. By using this type of price setting we can acquire a new form of transparency and justice and regain flexibility for libraries.
A start has been made by ACS, who announced during the ICOLC Conference in Rome in October 2006 that they were developing precisely this kind of pricing model and that they will introduce it later this year. To my surprise the librarians were not that enthusiastic -most of them believed that it was a new trick by ACS to get us to pay more than we did before. Another consideration could be that this will turn out to be a new source of administrative complexity. But librarians did not consider complexity so much as a problem, did they…? For a link to the table of contents for the issue of Serials in which this article first appeared, click here: http://serials.uksg.org/openurl.asp?genre=issue&issn=0953-0460&volume=20&issue=2