Economic implications of alternative scholarly publishing models
Abstract.
Financial pressures – reducing budgets and increasing prices – have been one of the key drivers in the search for alternative, open-access-based publishing models, particularly in the higher education sector. Houghton’s case study describes the ways in which he used lifecycle and costing techniques in a macro-economic modelling approach to test the extent to which new methods of publishing were more cost-beneficial to research and development activity than existing ones. He found that there was a significant return on investment especially through cost-savings over a transitional period of 20 years in all the countries studied, and for research libraries in particular, even though there was a degree of variation as a result of regional factors. He describes and discusses the methodology used, which incorporated process mapping, activity costing and macro-economic modelling using a modified version of the Solow-Swan model. There remain questions of sustainability, though archiving policies, as described by Houghton, should do much to improve the likelihood of long-term viability, especially if projected savings can be achieved.
Introduction
This case study examines the findings of a series of studies that have explored the implications of alternative scholarly publishing models for researchers and research libraries, especially those in higher education institutions (HEIs).The primary study was funded by the UK Joint Information Systems Committee1 (JISC) and looked at the costs and benefits of alternative publishing models for the UK in general and UK HEIs in particular (Houghton et al., 2009a). That study generated a good deal of interest and there have been a number of follow-on projects, including national studies in the Netherlands and Denmark (Houghton et al., 2009b; Houghton, 2009a), and a three-country comparison of the UK, the Netherlands and Denmark that sought to compare the potential impacts of open access publishing models in one of the larger, a midsized and one of the smaller European countries (Houghton, 2009b).
During 2010 and 2011, there were four further projects. The first focused on Germany and brought the German National Licensing Programme (NLP)2 into the mix of alternative publishing and distribution models (forthcoming). The second was conducted by Alma Swan of Key Perspectives and used the online cost model developed for the original JISC study to examine the cost implications for a sample of UK universities (Swan, 2010). The third significantly extended one aspect of the method used in the original study to explore the return on investment implications of the proposed US Federal Research Public Access Act (FRPAA) (Houghton et al., 2010). The fourth project approached it from the user perspective, exploring the information needs and access levels of small high-technology firms in Denmark (Houghton et al., 2011).
All of these projects have examined the potential impacts of alternative publishing models on the costs faced by various stakeholders, and the potential direct and wider benefits of more open access to the findings of publicly funded research. Looking at the publishing models as alternatives, they suggest that:
Open access (OA) publishing models are likely to lead to cost savings and substantial wider benefits arising from increased access to research findings and efficiency gains in the performance of research activities.
Self-archiving alternatives (such as Green Open Access) appear to be the more cost-effective, although whether or not self-archiving in parallel with subscription publishing is sustainable over the longer term is open to debate.
Open access publishing might bring substantial activity cost savings in research libraries, as well as subscription cost savings, helping to free resources that could be used in meeting new needs and taking on new responsibilities.
Alternative publishing models
The UK, Dutch and Danish studies all focused on three alternative models for scholarly publishing, namely:
traditional subscription or toll access publishing, using individual subscriptions or the so-called Big Deal, where institutional subscribers pay for access to online aggregations of journal titles through consortial or site licensing arrangements;
open access publishing, using the ‘author-pays’ model, where publication costs are paid from the author’s side rather than by readers;
open access ‘self-archiving’, where authors deposit their work in open access repositories, making it freely available to anyone with Internet access.
Of itself, self-archiving does not constitute formal publication, so analysis focused on two publishing models in which self-archiving is supplemented by the peer review and production activities necessary for formal publishing, namely:
’Green OA’ self-archiving operating in parallel with subscription publishing;
the ‘deconstructed’ or ‘overlay journals’ model in which self-archiving provides the foundation for overlay journals and services (Smith, 1999, 2005; Van de Sompel et al., 2004; Simboli, 2005; Houghton, 2005).
Hence, each of the publishing models explored includes all of the key functions of scholarly publishing, including peer review and quality control.
Method
The approach taken to the three national studies involved a combination of process mapping, activity costing and macro-economic modelling, and the research involved four main steps.
Process mapping
To provide a solid foundation for the identification of activity costs we adopted a formal process modelling approach using the Integration DEFinition (IDEF0) modelling standard,3 which is often used in business process re-engineering. It is a development of the lifecycle model created by Björk (2007), which we extended to include five main activity elements:
Fund research and research communication.
Perform research and communicate the results.
Publish scientific and scholarly works.
Facilitate dissemination, retrieval and preservation.
Study publications and apply the knowledge (see Figure 7.1).
Each of these activities is further subdivided into a detailed description of the activities, inputs, outputs, controls and supporting mechanisms involved, creating an activity model with some 53 diagrams and around 200 activities. The process model4 is available online.
Activity costing
This formal process model provided the foundation for detailed activity costing, using a spreadsheet-based cost model that included all of the elements in the lifecycle model, as well as the base data necessary for the study (relating to research and scholarly communications activities at the national and higher education levels). The costings relied primarily on existing sources and the collation of activity cost information from the wide-ranging literature on scholarly communication and publishing. Where necessary, these sources were supplemented by informal consultation with experts in the field. For the UK national and higher education data, we relied on national and international sources on research and development (R&D) expenditure and personnel by activity and sector, expenditure and employment trends. Detailed data on higher education were sourced from such agencies as the Society of College, National and University Libraries (SCONUL) and the Higher Education Statistical Agency (HESA). The resulting activity cost model included more than two thousand data elements.
Macro-economic modelling
To measure the impacts of alternative scholarly publishing models on returns to R&D expenditure, we modified the standard Solow-Swan model. The standard approach makes some key simplifying assumptions, including that all R&D generates knowledge that is useful in economic or social terms (the efficiency of R&D), and all knowledge is equally accessible to all entities that could make productive use of it (the accessibility of knowledge). Addressing the fact that these assumptions are not realistic, we introduced accessibility and efficiency into the standard model as negative or friction variables, to reflect the fact that there are barriers to access and limits to efficiency. Then we explored the impact on returns to R&D of changes in accessibility and efficiency (Houghton and Sheehan, 2006, 2009; Houghton et al., 2009a).
A step-wise approach
As noted, there were four main steps in the research process. In the first, we produced a detailed costing of all of the activities identified in the scholarly communication lifecycle model, focusing on areas where there were likely to be activity and, therefore, cost differences between the alternative publishing models. In the second, we summed the costs of the three publishing models through the main phases of the scholarly communication lifecycle, so we could explore potential system-wide cost differences between the alternative publishing models. In the third step, we used the modified Solow-Swan model to estimate the impact of changes in accessibility and efficiency on returns to R&D. The final step was to compare costs and benefits, for which we used the three elements outlined:
the direct costs associated with each of the models;
the associated indirect system-wide costs and cost savings;
the benefits accruing from increases in returns to R&D resulting from increases in accessibility and efficiency. Because the returns to R&D lag expenditure and accrue over a number of years, the cost-benefit comparisons were made over a 20-year transitional period.
Main findings
The analysis of the potential benefits of more open access to research findings suggested that open access could have substantial net benefits in the longer term, and while net benefits may be lower during a transitional period, they are likely to be positive for both open access publishing and overlay services alternatives (Gold OA) and for parallel subscription publishing and self-archiving (Green OA). For example, during a transitional period of 20 years we estimated that in an open access world:
the combined cost savings and benefits from increased returns to R&D resulting from open access publishing all journal articles produced in UK universities using an author-pays model (Gold OA) might be around three times the costs;
the combined cost savings and benefits from open access self-archiving in parallel with subscription publishing (Green OA) might be around seven times the costs;
the combined cost savings and benefits from an alternative open access self-archiving model with overlay production and review services (overlay services) might be around four times the costs.
While the benefits from unilateral national adoption of open access alternatives would be lower, they would be substantial – ranging from two to four times the costs.
In exploring the potential impacts of alternative publishing models in the Netherlands and Denmark, differences in the modelling were kept to a minimum, although some minor adjustments of the basic model were necessary to fit different national circumstances (Houghton, 2009b). Nevertheless, there are a number of factors that can affect the benefit/cost estimates for different countries. As modelled, these included such things as:
the number and size of universities and research institutions;
the implied number of institutional and other repositories, each with substantial fixed costs and relatively low variable costs;
the ratios of publicly funded and higher education research spending to gross national expenditure on R&D;
historical and projected rates of growth of R&D spending by sector;
relative national and sectoral publication productivity;
There were also inherent data limitations that varied somewhat between the countries. For example, in addition to cost differences between the countries, there were minor differences in the official national methods used to estimate the full cost of researcher activities.
Despite these influences, the different national studies produced very similar results and exhibited broadly similar patterns within the results. The cost-benefits of the open access ‘author-pays’ publishing model were similar across the three countries. In terms of estimated cost-benefits over a transitional period of 20 years, open access publishing all articles produced in universities in 2007 would have produced benefits of two to three times the costs in all cases.
One observable difference related to scale and the impacts of unilateral national adoption of open access, with the benefits of worldwide adoption being relatively larger for smaller countries as they produce a smaller share of the world’s journal articles. However, the most obvious difference between the results related to the ‘Green OA’ self-archiving and repositories model, which did not look quite as good in the Netherlands as in the UK and nothing like as good as it did in Denmark. This was due to the implied number of repositories, each with operational overheads. As modelled, the number of institutional repositories required in each country related to the number of institutions, and their operational overheads were shared across the number of articles produced and archived. For example, under the modelled assumptions, for 2007 outputs, the Netherlands’ 86 higher education institutional repositories might have housed around 26,000 articles (an average of 302 each from that year), the UK’s 168 higher education institutional repositories might have housed around 100,000 articles (an average of 595 each from that year) and Denmark’s eight universities’ repositories might have housed around 14,000 articles (an average of 1,750 each from that year). As modelled, these differences materially affected the implied per article cost of self-archiving. Of course, had we used an averaged per article lifecycle costing, these differences would not have been apparent.
Notwithstanding these differences, the modelling suggested that open access alternatives would be likely to be more cost-effective in a wide range of countries (large and small), with ‘Gold OA’ or author-pays publishing, the deconstructed or overlay journals model of self-archiving with overlay production and review services, and ‘Green OA’ self-archiving in parallel with subscription publishing being progressively more cost-effective.
Given the potential benefits identified in the three countries, we recommended a policy focus on creating a level playing field by reducing the barriers to innovation in scholarly publishing and raising awareness of the opportunities. We suggested that this might involve, perhaps above all else, ensuring that research evaluation is not a barrier to innovation by developing metrics that support innovation in scholarly publishing rather than relying on traditional evaluation metrics that tend to reinforce traditional publishing models and reward traditional behaviours. It might also involve:
ensuring that there is funding for author-side fees by encouraging funders to make specific provision for publication charges;
encouraging – and perhaps funding – OA repositories to enable self-archiving should authors choose;
supporting advocacy initiatives to inform all stakeholders about the potential costs and benefits of alternative publishing models.
Extensions and developments
The general approach used in the three national studies has been further developed and extended in recent work in Germany, the UK, the US and Denmark.
Germany
As a part of a much larger project, funded by the Deutsche Forschungsgemeinschaft (DFG), we worked with colleagues at Goethe University in Frankfurt on a study that brings the German National Licensing Programme (NLP) into the mix of alternative publication and dissemination models, and compares the NLP with the subscription and open access alternatives.
The German NLP provides enhanced access for researchers in Germany through an extended form of consortial purchasing and licensing. While it centralises a number of activities in the lifecycle process relating to facilitating dissemination, retrieval and preservation (for example, negotiation and licensing), the NLP does not fundamentally change the activities performed. However, the NLP does impact a number of activities in the scholarly communication lifecycle, thereby affecting costs.
There is one important difference between the comparisons in the German study and those in the previous studies. Subscription and open access publishing perform very different roles. To the limits of affordability, subscription publishing seeks to provide an institution’s researchers with access to the worldwide research literature whereas open access seeks to provide worldwide access to an institution’s research output. These are very different things, but to compare cost-effectiveness it is necessary to compare like with like. It is also important to note that subscriptions do not cover the cost of subscription publishing. There is also revenue from advertising and reprints, page and plate charges, membership and other subsidies to subscription journals. So, despite the fact that it is what most people do, it is not always correct to compare open access publishing costs with subscription expenditures. In the JISC study, we compared the costs associated with publishing UK article output under different models, including subscription. In contrast, the German study compares the costs of operating within alternative models. This does not compare the cost of using alternative models to achieve a comparable task. Rather, it compares the cost implications of alternative models for a particular actor, in this case for Germany.
We found that the benefits of open access publishing are likely to outweigh the costs in Germany, as elsewhere. Interestingly, the NLP returned the second highest benefit/cost ratio during a transitional period – after ‘Green OA’ self-archiving. However, the NLP is a long-term commitment, and there is a risk that new developments in publishing models may change the relative cost-benefit of the NLP over time.
The United Kingdom
Alma Swan, of Key Perspectives,5 undertook another follow-on study for JISC in which she applied the online cost model produced as a part of the original study to an examination of the cost and benefit implications of alternative publishing models for a small sample of UK universities (Swan, 2010). As in the German study, Swan compared the costs of operating within alternative models, in this case for a sample of universities, by setting the cost of publishing UK articles under alternative publishing models against the costs of access to that share of worldwide articles to which the institutions currently subscribed.
Swan showed how universities could compare the impacts of alternative publishing models for themselves, and that by looking at whole-of-system costs we can start to question the somewhat simplistic arguments that suggest that in research-intensive universities author-pays fees may be higher than current subscription expenditures. While that may be true in some cases, it is also apparent from this study that the potential savings in research time, library handling costs and so on that could arise from more open access would also be greatest in the more research-intensive universities. To understand institutional cost impacts, we need to take a fuller account of system-wide costs than has been typical to date.
The United States
During 2010, the Scholarly Publishing and Academic Resources Coalition (SPARC) supported a feasibility study that sought to outline one possible approach to measuring the impacts of the proposed US Federal Research Public Access Act (FRPAA) on returns to public investment in R&D (Houghton et al., 2010). The aim of the study was to define and scope the data collection requirements and further model developments necessary for a robust estimate of the likely impacts of the proposed FRPAA archiving mandate.
The project involved a major shift from previous studies in that its focus was on the modified Solow-Swan model rather than the scholarly communication lifecycle model or the associated activity cost model. That focus enabled further development and refinement of the modified model, particularly in relation to the most appropriate lag and distribution over time of returns to R&D, the most appropriate depreciation rate for the underlying stock of R&D knowledge arising from federally funded R&D, and metrics to measure potential changes in accessibility and efficiency. To establish plausible base case values for these parameters we drew on the extensive literature on returns to R&D (Salter and Martin, 2001; Martin and Tang, 2007; Sveikauskas, 2007; Hall et al., 2009).
The other piece of the puzzle is the input data required for the modelling. These include the implied archiving costs, the volume of federally funded research outputs (such as journal articles) and the levels of federal research funding and expenditure trends. For the purposes of preliminary analysis we used publicly available sources and published estimates. Data relating to federal research funding, activities and outputs were taken from the National Science Board Science and Engineering Indicators 2010 (NSB, 2010), and we explored three sources for archiving costs: the LIFE2 Project lifecycle costs (Ayris et al., 2008) and submission equivalent costings from arXiv (such as the former Ginsparg Archive at Cornell) and the National Institutes of Health PubMed archive (arXiv, 2010; NIH, 2008).
Preliminary modelling suggested that over a transitional period of 30 years, the potential incremental benefits of the proposed FRPAA archiving mandate for all federally funded R&D might be worth around four times the estimated cost using the higher-end LIFE2 lifecycle costing, eight times the cost using the NIH costing, which is likely to be the best guide, and as much as twenty times the cost using the historical arXiv costing. Perhaps two-thirds of these benefits would accrue within the US, with the remainder spilling over to other countries. Hence, the US national benefits might be of the order of five times the costs and be worth more than $1 billion a year.
Exploring sensitivities in the model in order to prioritise areas for further data collection and model development, we found that the benefits exceed the costs over a wide range of values. Indeed, in the US model, and in the other national models, it is difficult to imagine any plausible values for the input data and model parameters that would lead to a fundamentally different answer.
Denmark
In a study funded by the Danish Ministry of Culture and Ministry of Science, Technology and Innovation, we approached the issue from the user side and explored access to, and use of, academic research among small high-technology firms in Denmark (Houghton et al., 2011). It was a small, non-random sample: we received 98 usable responses to an online survey and conducted 23 in-depth interviews. Nevertheless, we found that:
Comparing responses on the importance and ease of access suggests that the things that are both important and difficult to access include research articles and market survey research and reports (and, for some firms, patents):
48 per cent rated articles very or extremely important;
38 per cent said they always or frequently had difficulty getting the articles they needed; and
72 per cent reported using institutional or subject repositories;
56 per cent used open access journals on a monthly or more regular basis.
Asked how long they spent trying to access the last article they had difficulties accessing, the average among researchers was 63 minutes, and an average of 17 articles presented access difficulties during the last year. So, access difficulties could be costing around DKK 540 million a year among specialist researchers in Denmark alone.
Looking at the value of academic research for the firms, we found that an average of 27 per cent of the products and 19 per cent of the processes developed or introduced during the last three years would have been delayed or abandoned without access to academic research. The value of academic research to sales was around DKK 16 million per firm per year and the value to cost savings around DKK 95,000 per firm per year. So, on a simple pro rata basis, academic research is directly contributing the equivalent of around 12 per cent of sales revenue.
Implications for research libraries
Throughout these studies, analysis of the potential benefits of more open access to research findings suggested that open access could have substantial net benefits in the longer term, and while net benefits may be lower during a transitional period, they are likely to be positive for both open access publishing and overlay services alternatives (Gold OA) and for parallel subscription publishing and self-archiving (Green OA).
Given a capacity to enhance access at very little cost, self-archiving alternatives appear to be the more cost-effective. Although whether or not self-archiving in parallel with subscriptions is a sustainable model over the longer term is, perhaps, debatable, nevertheless the evidence from these studies suggests that archiving policies and mandates, be they at the national, sectoral, funder or institutional levels, can enhance accessibility and improve efficiency at relatively little cost, and with no immediate disruptive change to scholarly publishing practices and traditions.
With an increasing number of policy directives and funder and institutional mandates calling for more open access to the findings of publicly funded research, it seems inevitable that the share of material available in open access will increase. The most recent count suggested that more than 20 per cent of the journal articles produced during 2008 were already freely available online the following year (Bjork et al., 2010). So, open access is already a reality and an increasingly important part of the access landscape.
At the same time, of course, traditional publishing models live on, and the mix of access and publishing models increases. This has brought, and is likely to continue to bring, a shift in research library activities and responsibilities, adding the operation of institutional or subject repositories and a new role in the dissemination of the institution’s research publications (and data) to existing and increasingly challenging duties to assist in information literacy and access. These shifts also bring a wider range of users to research libraries, as users outside the traditional subscription world can and do seek open access materials – be they open access journals that may be supported and hosted through a research library or open access articles and other materials produced by the institution and hosted on an institutional repository.
While the studies reported in this case study suggest that open access might bring substantial subscription and activity cost savings for research libraries (such as in negotiation and licensing, library handling and access control and so on) there are also new challenges to be met. The savings that might be realised in some areas may release resources to address the new challenges (curating and sharing data, improving information literacy, building tools to enable researchers to do their work better and more efficiently, and do new kinds of work based on text mining, data mining, complex research workflows and so on). The benefits of being able to address some of these challenges with the resources released by open access and, of course, the shift toward electronic-only holdings could be considerable.
Acknowledgements
The author would like to acknowledge the support of the UK Joint Information Systems Committee (JISC) in the development of the modelling approach underpinning the studies reported, the SURF Foundation and Denmark’s Electronic Research Library (DEFF) for enabling its application in the Netherlands and Denmark, SPARC for enabling its further development and application in the US, and DfG for extending the work in Germany. Thanks are due to the research team from the original JISC project, including principal collaborator Charles Oppenheim of Loughborough University, Bruce Rasmussen and Peter Sheehan of Victoria University in Melbourne, and Anne Morris, Claire Creaser, Helen Greenwood, Mark Summers and Adrian Gourlay of Loughborough University, as well as members of the project advisory group. Thanks are also due to the research teams for the Netherlands project, including Jos de Jonge and Marcia van Oploo of EIM/Research voor Beleid and to the many people who assisted in both the Dutch and Danish studies, Bruce Rasmussen and Peter Sheehan of Victoria University in Melbourne, Alma Swan and Sheridan Brown of Key Perspective, and Wolfgang König, Berndt Dugall, Matthias Hanauske, Julia Krönung and Steffen Bernius of the Goethe University in Frankfurt.
References
arXiv. org. arXiv Business Model White Paper. Cornell University. Online at: http://arxiv. org/help/support/whitepaper, 2010.
Ayris, P., Davies, R., McLeod, R., Miao, R., Shenton, H., Wheatley, P. The LIFE2 Final Report. Bristol: JISC. Online at: http://discovery. ucl. ac. uk/11758/, 2008.
Björk, B. -C. A model of scientific communication as a global distributed information system. Information Research. 2007; 12(2):307.
Björk, B. -C., Welling, P., Laakso, M., Majlender, P., Hedlund, T., Open access to the scientific journal literature: situation 2009. PLOS ONE. 2010 e11273. Online at;5(6): http://www. plosone. Org/article/info:doi/10. 1371/journal. pone. 0011273
Hall, B. H., Mairesse, J., Mohnen, P. Measuring the Returns to R&D, NBER Working Paper 15622. Cambridge, MA: NBER; 2009.
Houghton, J. W. Economics of publishing and the future of scholarly communication. In: Gorman G. E., Rowland F., eds. International Year Book of Library and Information Management 2004–2005: Scholarly Publishing in an Electronic Era. London: Facet Publishing, 2005.
Houghton, J. W. Costs and Benefits of Alternative Publishing Models: Denmark. Copenhagen: DEFF. Online at: http://www. knowledge-exchange. info/Admin/Public/DWSDownload. aspx?File=%2fFiles%2fFiler%2fdownloads%2fDK_Costs_and_benefits_of_alternative_publishing_models. pdf, 2009.
Houghton, J. W. Open Access: What Are the Economic Benefits? A Comparison of the United Kingdom, Netherlands and Denmark. Brussels: Knowledge Exchange (June). Online at: http://knowledge-exchange. info/Default. aspx?ID=316, 2009.
Houghton, J. W., Sheehan, P. J. The Economic Impact of Enhanced Access to Research Findings. CSES Working Paper No. 23. Melbourne: Victoria University (July). Online at: http://www. cfses. com/documents/wp23. pdf, 2006.
Houghton, J. W., Sheehan, P. Estimating the potential impacts of open access to research findings. Economic Analysis and Policy. 2009; 39(1):127–142.
Houghton, J. W., de Jonge, J., van Oploo, M. Costs and Benefits of Research Communication: The Dutch Situation. Utrecht: SURF Foundation (June). Online at: http://www. surf. nl/en/publicaties/Pages/CostsandBenefitsofOpenAccessPublicationlTheDutchSituation. aspx, 2009.
Houghton, J. W., Rasmussen, B., Sheehan, P. Economic and Social Returns on Investment in Open Archiving Publicly Funded Research Outputs. Washington, DC: SPARC. Online at: http://www. arl. org/sparc/publications/papers/vuFRPAA/index. shtml, 2010.
Houghton, J. W., Swan, A., Brown, S. Access to Research and Technical Information in Denmark: Report to the Danish Agency for Science, Technology and Innovation (FI) and Denmark’s Electronic Research Library. Copenhagen: Forsknings- og Innovationsstyrelsen and Denmark’s Electronic Research Library. Online at: http://eprints. soton. ac. uk/272603/, 2011.
Houghton, J. W., Rasmussen, B., Sheehan, P. J., Oppenheim, C., Morris, A., Creaser, C., Greenwood, H., Summers, M., Gourlay, A. Economic Implications of Alternative Scholarly Publishing Models: Exploring the Costs and Benefits. Bristol: JISC. Online at: http://www. cfses. com/EI-ASPM/, 2009.
Martin, B. R., Tang, P. The Benefits of Publicly Funded Research. SWEPS Paper No. 161. Brighton: Science Policy Research Unit. Online at: https://www. sussex. ac. uk/webteam/gateway/file. php?name=sewp161&site=25, 2007.
National Institutes of Health (NIH). Analysis of Comments and Implementation of the NIH Public Access Policy. Washington, DC: NIH. Online at: http://publicaccess. nih. gov/analysis_of_comments_nih_public_access_policy. pdf, 2008.
National Science Board (NSB). Science and Engineering Indicators 2010. Arlington, VA: National Science Foundation (NSB 10–01). Online at: http://www. nsf. gov/statistics/, 2010.
Salter, A. J., Martin, B. R. The economic benefits of publicly funded basic research: a critical review. Research Policy. 2001; 30(3):509–532.
Simboli, B. Subscription Subsidized Open Access and the Crisis in Scholarly Communication. Bethlehem, PA: Lehigh University; 2005.
Smith, J. W. T. The deconstructed journal, a new model for academic publishing. Learned Publishing. 1999; 12(2):79–91.
Smith, J. W. T., Open access publishing models: reinventing journal publishing Research Information, May-June. Online at. http://www. researchinformation. info/features/feature. php?feature_id=79, 2005
Sveikauskas, L. R&D and Productivity Growth: A Review of the Literature. Washington, DC: Bureau of Labor Statistics Working Paper 408. Online at: http://www. bls. gov/osmr/pdf/ec070070. pdf, 2007.
Swan, A. Modelling Scholarly Communication Options: Costs and Benefits for Universities. Bristol: JISC. Online at: http://ie-repository. jisc. ac. uk/442/, 2010.
Van de Sompel, H., Payette, S., Erickson, J., Lagoze, C., Warner, S., Rethinking scholarly communication: building the system that scholars deserve D-Lib Magazine, 10 (9). Online at. http://www. dlib. org/dlib/september04/vandesompel/09vandesompel. html, 2004