The starting point for a study of pricing mechanism for the delivery of materials in digitised form to HEIs must be informed by an understanding some basic economic market characteristics of information services in the digital age. Halliday and Oppenheim (1999) pointed out that economic models have tended to be tested after solutions have been found to the more pressing technical, cultural and legal issues involved in providing digitised materials. Even though we recognise the importance of considering all these issues sufficiently to provide the foundation for a sound information infrastructure, there is an urgent need to explore viable pricing mechanisms that would satisfy all stakeholders involved in the academic information delivery chain. Since the digital library is still in its early stage of development, a useful economic model with its corresponding pricing mechanism must be adaptable and responsive to change. The economics of digital library services are characterised by its dynamics and fluidity, making it unlikely that a static model will do justice to the evolution of this young market. In this article, we first review some basic economic characteristics of supply and demand for the digital library, before some specific economic models are considered.

From the supply side, one of the most fundamental features of information goods is that their cost of production is dominated by the 'first-copy costs' (Shapiro and Varian, 1999). With recent advances in information technology and more materials becoming available in digitised formats, first-copy costs are likely to comprise an even greater fraction of total costs than hitherto. This cost structure of high fixed costs with near-to-nothing variable costs leads to substantial economies of scale.

The last two decades have seen an explosion of scholarly information that has been mirrored in a consistent increase of scholarly journal prices above the general rate of inflation. With academic libraries facing significant budget restrictions, the advent of electronic publishing has been welcomed by some as an opportunity to solve these problems. Since electronic publishing is cheaper than traditional hard copy publishing (Bot et al., 1998; Fishwick et al., 1998), there was hope that the digital library would be the way out of the cost dilemma. Publishers respond that over 70% of the publishing costs are still fixed. Therefore the distribution channel used has little impact on the existing cost structure (Frey, 1997). In addition, users request improved functionality such as multimedia elements, hot links etc. from electronic publications, perhaps eliminating initial cost savings. Since the demand for digitised materials is increasing, publishers are forced to respond.

The electronic publishing industry is further characterised by a relatively inelastic price elasticity of demand, which has a considerable impact on the development of market structures. Since the substitution effect is currently low for the academic journals market, it is likely that the new technologies used in the digital library will only have a long-term effect on prices. In some subjects in the short-term, the academic community will remain dependent on existing literature that is published by the traditional publishing houses. With the advancement of new technologies and the availability of software that allows the easy publication of materials in-house, Universities and other research centres will be more likely to create their own substitutes for the traditional journals market in the years to come. In the long-term, this would cause the price elasticity to become more elastic. This, in turn, might lead Universities to recognise digital publishing as an additional source of revenue. Currently, Universities would face an administration problem to cope with the complexity of monetary flows but, considering the recent developments of micropayment systems, those initial obstacles could be resolved in the long term. A more difficult problem lies with the development of a transparent copyright clearance system. If copyright remains with each author and is administered by the institution that has published the material, it is likely that copyright violations will become a serious problem. Consequently, it will be in the best interest of all stakeholders to work jointly on an economic model that satisfies their needs and is responsive to change in the future.

Frey (1997) and Arms (2000) reviewed a number of economic models that could form the basis for the digital library. Rather than assuming that any of the models that will emerge as the solution for the digital library, it is more likely that mixed forms will emerge and co-exist alongside each other depending on user needs and their ability to pay for certain services. Probably the most prominent model of the world-wide-web is the open access model 'without costs'. Obviously, these websites are not free to maintain and costs have to be recovered either through advertising or external funding. To date, academic publishers work with only little advertising. This is surprising considering the wide-ranging socio-economic profile of the academic community and level of tolerance students have towards advertising on the Internet (Rowland and Rubbert, 2000). External funding sources for digital journals might become more widely available in the future with ever more research centres producing sophisticated, easy-to-use documentation on-line. It is possible that some of the working paper series of some institutions will turn into digital journals in the future. We suggest that publishers should effect the incorporation of advertising and external funding sources into their business models. Although initial costs for the creation of a fund-raising department are high because of the expertise required in this area, the long-term payback may be beneficial to the whole of the academic community.

The current climate in digital publishing is possibly determined by restricted access models with its corresponding economic mechanisms. There are a number of subscription models that are evolving in the digital domain. Most subscription models are password based. Some students complain that access to digital services is restricted in a lot of cases to on-campus use only (Rowland and Rubbert, 2000). In particular, part-time and distance learning students are disadvantaged by these information provision practices. Advances in biometric user recognition systems are likely to bring a solution to the problem of secure data transmissions in the near future. These systems work on the basis of recognising biometric characteristics of the end user (e.g. fingerprints), allowing for the complete automation of user control. Even though these technical facilities will make the full introduction of the digital library into the academic community much easier, they do not solve the issues of an appropriate pricing mechanism that recovers costs for information services. Bide et al. (1997) point out that any pricing mechanism must be simple, inexpensive to administer and produce predictable costs for users and revenues for publishers. In their report, the authors recommended that publishers adopt one or both of two pricing models when agreeing to license the digitisation of texts: the "textbook purchase substitution" model and the "library purchase substitution" model. The former refers to those texts that students would be expected to buy and allows the rights owner to issue a limited term licence (probably covering an academic year) linked to the number of students. The latter model would apply to those texts libraries would be expected to buy, and allow the rights owner to issue a perpetual licence in return for a one-time fee, which would not be modified by any "usage" metric. This would provide the publisher with a single payment broadly analogous with a library purchase (Bide et al, 1997).

Frey (1997) draws further attention to control circulation models. Normally, they require affiliation with a professional society in order to obtain society published materials. Publishers would create subject gateway services for certain disciplines. Such services would than include access to certain types of journals and would be paid for by a general subscription charge from the end user. Even though such an economic model would be relatively easy to administer, it causes considerable concern since the costs for the use of information services is ultimately shifted to the end user. Library substitution payments are not accounted for in this model, and a mechanism would need to be found to fairly integrate these payments over all disciplines, a task that is likely to bring some upheaval among academics and publishers alike. Therefore, transactional models are more likely to establish themselves within the digital library. The philosophy driving these models is that the user will only have to pay what has actually been used. The economic mechanism behind these models is normally referred to as 'pay-per-view'. The end user pays a fixed charge for every page or article he has downloaded. It appears that the academic community currently favours this model since actual use of materials can be easily monitored. Nevertheless, micropayment systems are technically not very advanced yet and it may take some years before such an economic model can be described as profit maximising. The most controversial model that is currently discussed among information professionals is a payment system based on submission fees. Since academics are increasingly forced to publishing in relevant journals, some authors suggest charging the academic for publication. In our view, such an economic model would have a long-term impact on the social structure of the academic community and should only be considered if any other model fails. Young academics, interdisciplinary research centres, and the new Universities are already facing tight budget restrictions; the introduction of such a model is likely to further divide the HE community.

Halliday and Oppenheim (1999) evaluated six economic models of the production and delivery of specific digital library services that derived from the general models above. Four of the models simulated the production of a single journal under varying conditions and the other two models evaluated were a resource discovery network and a National Electronic Reserve Service (NERS). The study revealed that the variation of costs for overheads and initial production costs had a considerable impact on the journal price. This could be of particular interest if the impact of a reduction of these costs is considered over the market as a whole. It can be assumed that the marginal costs of large publishing houses are most likely to decrease over all journal titles, in particular, if the level of functionality is decreased. Odlyzko (1999) claims that the last 20% of functionality account for 80% of the costs. With an increase in technological innovation to produce digitised materials, it is most likely that further cost reductions can be achieved and passed on to the end user. Consequently, it is suggested that the Halliday and Oppenheim (1999) study be expanded to include a dynamic representation of the market as a whole. In particular, it needs to be explored whether the differentiation between 'born' digital texts and those digitised after print publication has an impact on the viability of a pricing mechanism.

Since the publication of the Bide et al report, a number of key events have occurred. HERON (see below) has been established to continue the momentum of the on-demand/electronic reserve (OD/ER) strand of the eLib programme, and the Copyright Licensing Agency (CLA) established the Higher Education Digitisation White Paper concerning the licence terms for digitisation (CLA Higher Education White Paper, 1998). Following this, the CLA launched its HE digitisation license, known as eCLA and has adopted models similar to those recommended by JISC/PA. In 1999 the CLA launched it's "first scheme to license the digitisation of print material" (CLA Annual report 1999) which provides for both digitisation of printed textual material and its subsequent use.

The CLA Annual report states that "by October 1999 over 700 publishers had...added their works to the new scheme including most of the major UK publishing houses...by the end of October over 100 HE institutions had signed a licence" (Annual report, 1999). The CLA uses its Rapid Clearance Service (CLARCS) to allow CLA licensees to quickly obtain permission for such digitisation. But it was not only in the UK that a need for identifying an economic model for the electronic distribution of materials in HEIs has been recognised. Other countries developed similar initiatives alongside those in the UK. In the following, we will review some of the initiatives that developed in the UK before we turn to an evaluation of international practices.

HERON (Higher Education Resources ON demand) is funded jointly by JISC and Blackwells Retail Ltd.

HERON provides a one stop copyright clearance and digitisation service for the HE sector. It currently undertakes the role of an intermediary between the HEIs and publishers, rightsholders and the CLA and helps in their respective quests for a means by which the academic sector can provide access to digitised learning materials for students. HERON also ensures that publishers can be reassured that this is being done under secure and controlled conditions (Curry, 2000).

In October 2000, the CLA, which is now authorised by over 800 publishers to offer the community access to digitise their works (press release, 2000), signed an agreement with HERON for the clearance of print material for digitisation to be used by the HE sector. In working with the CLA, HERON is also creating a resource bank of digitised materials.

The success of HERON, and indeed any other form of electronic access to learning materials, is dependent on an economic model that makes the availability of materials to students no more expensive than current methods. At the same time, there must be a fair rate of return to those in the supply chain, including authors, publishers and service suppliers such as HERON.

The PLS (Publishing Licencing Society) mandate, with the CLA, has set the 'textbook model' as the default, although it stated that the "library substitution" model is appropriate for scholarly monographs and journal articles, and has suggested that a default rate of five pence per page be adopted as common practice amongst publishers. While this is generally accepted publisher practice, many HEIs refuse to pay such a high rate for permission to digitise.

HERON is not the only body that had struggled with the implementation of a workable pricing model for the distribution of digitised published materials in HEIs. Many of the problems the British scholarly publishing industry faces are echoed worldwide. A viable economic model will ultimately depend on its fit to the corresponding national copyright management and clearance system, but to date, there is no uniform approach to deal with copyright issues around the world. Carmel and Collins (1997) show that the market is divided between collective management through mandatory clearinghouses (e.g. Norway, Finland, Japan), and individual management approaches that favour market solutions (e.g. US, UK). A majority of countries have special arrangements that depend on the types of work that are being protected, using voluntary clearinghouses. Even though the collective management approach has, in principal, the advantage to reduce transaction costs, there is a danger that centralised pricing will cause pricing convergence, which leaves poor works over - and top works under priced. Market solutions, on the other hand, work principally in favour of the publisher, with fair dealing clauses remaining a nebulous concept without solid definition. As the majority of countries operate in the middle spectrum of both management approaches, there is evidence from all countries that there is no agreed pricing mechanism for the distribution of digitised published materials in HEIs.

Nevertheless, a report by the Media Group for the European Commission (2000) identifies a number of opportunities for the academic periodicals market that could be utilised for the successful introduction of increasing numbers of digitised texts in HEIs. Because of the international nature of academic publishing, it is possible for firms to globalise their operations and market bases, which allows the spread of costs for the development of digitised materials. Since academic institutions and readers have the information technology, infrastructures, and financial resources to access electronic publications, publishers have the security that their clientele will most certainly use the introduction of new titles in digitised forms. On the other side, it must be realised that the highly developed IT infrastructure of academic institutions might equally pose a threat to publishers since some organisations may decide to return to publishing their own titles directly because of cost savings and lower risk of operations. Consequently, there is an urgent need to develop a pricing mechanism, which has the potential to develop into a strategic tool that operates globally for the benefit of the whole industry. Even though national concerns must form the basis for such a model, it must integrate and critically assess those lessons that have been learnt from other projects worldwide.

For example, TULIP, one of the earliest US projects involving Elsevier and nine US Universities that conducted research into the use of digitised materials in HEIs, concluded in 1996 that even though good advances have been made with regard to hardware and software development, 'economic issues were least conclusive'. TULIP examined a number of pricing models, which included internal charging at Universities, consortia models, subscriptions to electronic-only material, and pay per use article delivery. As an agreement for best practice could not be found, Elsevier became involved in PEAK (Pricing Electronic Access to Knowledge). This project introduced three possible pricing models (pay per view, traditional subscriptions that include electronic links, and generalised subscription) in an experiment with the University of Michigan. Key developments of PEAK were the recognition of a non-linear pricing model for the distribution of electronically published materials in HEIs, and a distinction between an institutional and individual access model. The latter provides the individual with greater autonomy to choose between journal and article titles while reducing costs at the same time.

In Europe, TECUP emerged as an important meta-project that analysed the development of business models for licensing in the digital future of a number of national and international projects. Project partners working with TECUP pointed out difficulties in developing a suitable pricing mechanism for the distribution of electronic published materials for HEIs (TECUP, 1999). Models used were wide-ranging from author financing of specific electronic journals to more complex pricing mechanisms as implemented by HERON. In particular, the NESLI (UK), and Elektra (Finland) projects revealed that the biggest obstacles to implementing a simple pricing mechanism involve lengthy negotiating processes with publishers through managing agents, and a lack of decision-making power on the individual project basis. In the case of LAURIN (Norway), the question of a business model was not even reached because European copyright issues could not be resolved. Others such as Decomate II had explicitly addressed the question of an appropriate business model, but problems emerged in negotiating with certain publishers. Only EZUL (Germany) had the explicit project aim to set up an appropriate pricing model. It arranged for a pay per view model with the library fixing the end-use price after negotiating a standard charge with the publisher. As the standard charge exceeds the end-use price that could be charged to the individual, it is not clear yet who will finance the difference. Other projects (e.g. LINK) co-operated only with certain publishers, which made it easier to implement a pricing strategy. Nevertheless, it is doubtful whether these models are useful for an universal application as in the case of HERON.

Even though PELICAN will not be able to deliver a worldwide solution, it recognises the importance to develop its model within a European framework. If a solution that satisfies all stakeholders in the HE community can be found for the UK market, there will be scope for adapting the model for use in other countries, as well.

It is clear that the delivery of electronic information to HEIs depends on the price being right. The development of a suitable pricing mechanism, satisfying all stakeholders in higher and further education, including publishers and authors, is a pressing and immediate task, that could provide the foundation for a holistic distribution model of academic literature in digitised form. Naturally, "the level of fee passed to the CLA and to rights holders is a crucial factor in determining the price set" (Curry, 2000).

Amongst other things, the "bookshop substitution" model requires payments for all students on a course. Most HE libraries would prefer a "library substitution" model, as it better reflects methods of text provision and usage. However, rightsholders are reluctant to adopt this model since they wish to ensure that sales of textbooks are not jeopardised and because there is no precedent for establishing a "library substitution" fee.

Although various technical developments are underway, at present there are no systems appropriate for HEIs which can satisfactorily monitor usage (e.g. print-outs, loading onto disk), including at remote sites, and most payment systems within HEIs are not sufficiently sophisticated to handle differential charging and payment. Publishers and authors are keen to encourage electronic delivery. However, publishers' priority seems to be not damaging present income streams. There also seems to be a lack of awareness and understanding of issues between UK HEIs and rightsholders of each other's needs and views.

Publishers are concerned that the impact of digitisation could threaten sales of journals to libraries and textbooks to students. "Furthermore, the integrity of digital copies can be compromised whilst maintaining the appearance of an authentic copy" (Halliday, 2000), thus placing publishers' reputations at risk and affecting the moral rights of the authors. A fair rate is hard to establish in an untested market (Halliday, 2000). Academic staff want easier access to copyright materials in digitised form for teaching and research, and would like to see a clear pricing model for their secondary use of publications (Bide at al, 1997). Libraries, already working under financial constraints, prefer to work with a fixed cost model, as this would allow for long-term planning. As far as students are concerned, since the introduction of tuition fees for undergraduates, and increasing prices for postgraduate programmes, they are more likely to want easy access to suggested readings in digitised form and reject additional payment for library resources (Bide et al 1997).

PELICAN (Pricing Experiment Library Information Co-operative Network) is another JISC-funded project. It is intended to develop an understanding of possible charging mechanisms for distributing commercially published electronic texts to students. The project will assess existing usage patterns and current electronic delivery of text to students. It will promote awareness of the different policies and pricing strategies that are available within the various stakeholder groups while satisfying the diverse interests and needs of all stakeholders. It will identify the models most likely to be acceptable to authors, the Copyright Licensing Agency (CLA), Higher Education, publishers, and students. This, in turn, will help the relevant stakeholders to develop appropriate business plans so as to ensure medium to long-term viability of any systems developed for this purpose, whilst at the same time ensuring an adequate return to the rights owners. PELICAN will act as a facilitator to provide a communication network for the development of a suitable pricing mechanism recognising the diverse interest of all stakeholders involved in the HE/FE community.

At the end of the project in late 2001, we will publish a summative evaluation report. This will be distributed widely among stakeholders through the World Wide Web and through publication in appropriate journals. The summative evaluation will allow for stakeholders to engage in further discussions beyond the lifetime of the PELICAN project.

As the research project is exploratory in nature, its success will ultimately depend on the communication infrastructure, which allows for open discussions with all stakeholders involved in the project, and the willingness for all parties to engage in such a full and frank exchange of views. To this end, we will be approaching a number of academic and Learned Society Publishers in the first half of 2001 to explore their views on the pricing of licences to digitise texts. All these explanations will be off the record and confidential, of course.

The project is under the direction of Professor Charles Oppenheim of Loughborough University, and the project team is working closely with its partners and its Advisory Board, which represents all stakeholders. The aim of PELICAN is to achieve a consensus economic model that will appeal to all stakeholders. Libraries in HEIs need such a consensus model urgently as an increase in the electronic distribution of materials for students is expected. PELICAN's modus operandi will be communicating and exchanging ideas with all stakeholders and suggesting innovative pricing strategies. We look forward to co-operating closely with the scholarly publishing industry in this exciting project.

Arms, W.Y. (2000) 'Economic Models for Open Access Publishing', iMP, March, http://www.cisp.org/imp/march_2000/03_00arms.htm

Bide, M., Oppenheim, C. & Ramsden, A. (1997) Charging Mechanisms for Digitised Text. Second Supporting Study for the JISC/PA working party, http://www.ukoln.ac.uk/services/elib/papers/pa/charging/

Bot, M. et al. (1998) 'The cost of publishing an electronic journal: a general model and a case study', D-Lib Magazine, http://mirrored.ukoln.ac.uk/lis-journals/dlib/dlib/dlib/november98/11roes.html

Carmel, E. and Collins, E. (1997) 'The impact of international copyright management and clearance systems on multimedia markets', Telematics and Informatics, 14(1), pp. 97-109

Copyright Licensing Agency (1998). Higher Education Digitisation Licensing Scheme. White Paper, http://www.cla.co.uk/have_licence/he/he_digitisation_oldwhite.html

Copyright Licensing Agency: Press Release, (11^th October 2000) CLA supports production of digital learning material, http://www.cla.co.uk/media/press_releases/press51.html

Copyright Licensing Agency (1999). Higher Education Digitisation Licence, http://www.cla.co.uk/have_licence/he/he_digitisation_oldagreement.html

Copyright Licensing Agency (1999) A Year of Growth, http://www.cla.co.uk/media/review-99.pdf

Curry, S. (2000) NAG Conference Sept 13-15^th HERON: Partnerships in practice: why we need each other.

Fishwick, F. et al. (1998) Economic Implications of Different Models of Publishing Scholarly Electronic Journals for Professional Societies and Other Small or Specialist Publishers, London: South Bank University

Frey, K.L. (1997) 'Business Models and Pricing Issues in the Digital Domain', Journal of Library Administration, 24(4), pp. 27-37

Halliday, L. and Oppenheim, C. (1999) Economic Models for the Digital Library, London: South Bank University

Halliday, Leah (2000) 'Copyright and Digitisation in UK HEIs'. VINE 118. p. 27-34

Media Group (2000), Competitiveness of the European Union Publishing Industries, Final Report prepared for the European Commission, Office for Official Publications of the European Communities, Luxembourg.

Odlyzko, A. (1999) The Evolution of Electronic Scholarly Communication, http://www.research.att.com/%7Eamo/doc/evolution.communications.txt

Rowland, F. and Rubbert, I. (2000) Information needs and practices in a lifelong learning environment, Research report presented at the Ingenta Institute Conference 2000, London, available at http://www.ingenta.com/home/pdfs/Rowland.pdf

Shapiro, C. and Varian, H.R. (1999) Information Rules, Boston: Harvard Business School Press

TECUP (1999) 'Providing and controlling access to digital documents', Workshop, 29 October 1999, hosted by the Deutsche Bibliothek, Frankfurt am Main

Back to PELICAN Home Page