Communicating the results of research: How much does it cost and who pays?


Published on

Presentation by Michael Jubb at the Bielefeld Conference, Germany February 2009.

Published in: Education, Technology, Business
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide
  • This a description of the core functions that was developed by the RIN, as part of an exercise in collaboration with researchers, funders, publishers and librarians to set out the key goals of public policy under each of those seven heads. Statement endorsed by all the key publisher and library bodies in the UK, inc STM So there is agreement across the key groups of players as to what the functions and the strategic goals of public policy are But there is much less agreement, of course, as to how those functions and goals are best fulfilled.
  • So who are the key groups of players in the scholarly communications and digital libraries world? They come from both the public and the private sectors
  • First time that this has been attempted, so far as we are aware All costs based on conservative assumptions, and we are aware of the impact of the exclusions from the scope of the study (and in terms of activities, only limited coverage of preservation (in libraries,) and no coverage of evaluating usage and impact costs) Important to emphasise that we include here not just cash costs, but non-cash costs in terms of the time that (especially) researchers spend on various critical activities in the system as a whole So what is presented here is a minimalist view of the costs of the system.
  • So, £175bn for the system as a whole, the vast majority for the doing of research itself; just under £60bn (£58.9bn) for everything else, from publication to reading of journal articles Research costs covers that part of overall R&D investment as recorded by the OECD that is conducted by active researchers and authors who are part of what in the UK is called the research base (mainly in universities). So what we are trying to capture here is only that part of the investment in research that results in papers in journals.
  • Of that £58.9bn, £6.4bn spent on publishing and distribution Will look in next few slides at the key components of that cost
  • Peer review the largest single element in costs of publishing, and of course it is a non-cash cost Based on very conservative assumptions as to time and as to salary costs of peer reviewers Note that this is only one part of the peer review system: estimate excludes costs of peer review at grant application stage ( UK Research Councils estimate £200m a year, and you should probably multiply that by at least 20 to get a global figure)
  • Next largest element: everything that publishers do to get a submitted article to publication stage
  • So first copy costs amount to c 50% of all publishing and distribution costs
  • Variable and indirect costs, together with need to generate a surplus together account for the remaining 50% of the costs of publishing and distribution under the current system
  • Now shift back to the broader picture of the £175bn costs of the system as a whole Note that provision of access by libraries Costs to users of time spent in search and print Costs of time spent in reading together amount to £52bn, much larger than the £6.4bn spent on pubication and distribution Will look at each of these in turn
  • First, access provision by libraries [go through each of the three bullets]
  • Second, a non-cash cost, but a real one
  • Third, another non-cash cost And as Don King stresses, this is a measure of value as well as of a cost (something we might want to explore further)
  • Now we shift focus again, to how these costs are being met, and by whom Remember that we have cash (revenue) costs and non-cash costs, and the non-cash ones are very significant Publishing and distribution: a complex picture which I’ll deal with in a moment Access: costs borne by libraries Search, downloading and reading: largely time costs that are borne by readers/users
  • These are the costs that publishers are most concerned with Note that Over 50% of the costs of publishing and distribution are met by academic libraries in the form of subscriptions Taken together, HE and other library subscriptions, plus the contributions made by peer reviewers, meet well over 90% of the costs Across the system as a whole, advertising, membership and individual subscriptions, and author-side payments, together relatively trivial funders of the system But there are differences across different types of journal, eg advertising revenue for Nature very significant; for niche journals not so academic subscriptions much less important for Science/ Lancet etc than for major discipline journals
  • This covers the £58.9bn cost of publishing, distributing, providing access, searching and printing, and reading (ie everything except the doing of the research itself) £46bn (78%) of costs met by the HE sector And remember that while HE budgets have been relatively buoyant (or at least have shown significant increases) in recent years, they are likely to be under increasing pressure as we look forward So it is not surprising that the HE sector is looking for efficiencies and cost savings
  • A question that the RIN’s funders do ask us: we do not gather and analyse information of this kind just for the sake of it. Don’t want to defend every last figure here, but hope I have demonstrated that the exercise has been done rigorously And we are working to make the model available so that others can try different sets of figures and assumptions But would be surprised if the overall dimensions of the picture presented here were to be changed significantly. So at least we have some impression of what the overall picture looks like Efficiency and effectiveness are important issues, and cost-efficiencies and effectiveness are part of what we need to look at The model allows us to do at least three kinds of things as set out on this slide Will say a little about a couple of the scenarios we looked at in my final few slides Shd add that we do present figures in the report on the UK contribution, but the model cd be used to make similar calculations for other countries, I’ll say no more on that issue here, but glad to answer questions
  • These figures based on continuation of trends we have seen over the past ten years: annual real terms increases of 2.5% in research funding and 1.5% in the annual production of articles Focus here back simply on publishing and distribution costs Assume no other changes (for example in formats of delivery) But note that the increases in costs shown here for publishing and distribution are dwarfed by the increases in costs for search and print, and for reading, which would, if researchers were to keep pace, rise by £14bn
  • Challenges here for peer review, and for libraries, esp in the HE sector: how are they going to meet these costs?
  • Assume here that we have moved to a position where 90% of journals publish only digitally Cost saving of £1bn, largely by libraries in not having to deal with print But there is a VAT issue in EU. (note also that the percentage saving in publishing/distribution is much higher (~12%) if we calculate solely on the basis of current variable costs ************************************************************************************* Other scenarios presented in the report relate to a large-scale move to author-side publication fees, where we estimate that a further saving of £561m cd be made And the introduction of payments for peer review on the basis of the time costs incurred, where the result would be a rise in the costs of subscriptions of around £1.4bn (c 43%) Other scenarios could of course be developed.
  • And if anyone wants to make use of the model, do get in contact
  • Communicating the results of research: How much does it cost and who pays?

    1. 1. Communicating the results of research: how much does it cost, and who pays? Michael Jubb Director Research Information Network Bielefeld Conference 4 February 2009
    2. 2. Core functions in the research communications system <ul><li>Doing research to generate new knowledge and understanding </li></ul><ul><li>Assuring the quality of information outputs </li></ul><ul><li>Ensuring appropriate recognition and reward </li></ul><ul><li>Presenting, publishing and disseminating information outputs </li></ul><ul><li>Facilitating access and use </li></ul><ul><li>Assessing and evaluating usage and impact </li></ul><ul><li>Preserving valuable information outputs for the long term </li></ul><ul><li>Research and the Scholarly Communications Process: Towards Strategic Goals for Public Policy. RIN 2007. </li></ul>
    3. 3. Some key groups of players <ul><li>Researchers as creators, disseminators and users </li></ul><ul><li>Research funders </li></ul><ul><ul><li>Public, charitable and commercial sectors </li></ul></ul><ul><ul><li>National policy-making bodies </li></ul></ul><ul><li>Research institutions </li></ul><ul><li>Publishers (and secondary publishers) </li></ul><ul><li>ICT service providers </li></ul><ul><li>Libraries and publicly-funded service providers </li></ul><ul><li>Commercial information service providers </li></ul>
    4. 4. So what does it all cost, and who pays? <ul><li>Scope </li></ul><ul><ul><li>journals only ( not monographs or unpublished data) </li></ul></ul><ul><ul><li>excludes secondary publishing and aggregation </li></ul></ul><ul><li>Methodology </li></ul><ul><ul><li>so far as possible, based on publicly-available information </li></ul></ul><ul><ul><li>some estimates based on expert information </li></ul></ul><ul><ul><li>annual costs, expressed in £s (assuming an exchange rate where necessary of $2 to the £) </li></ul></ul><ul><li>Model available for others to use to test different assumptions with different data </li></ul>
    5. 5. The big picture: overall costs of the current system
    6. 6. Publishing and distribution
    7. 7. Quality assurance <ul><li>Peer review </li></ul><ul><ul><li>a non-cash cost (excludes costs of editorial boards, and of managing peer review) </li></ul></ul><ul><ul><li>30% of total cost of publishing and distribution </li></ul></ul><ul><ul><li>based on estimates of (2.5 reviewers per article) x 4 hours per reviewer (see Tenopir and King 2000, and Mark Ware for PRC 2008) </li></ul></ul><ul><ul><li>takes into account rejection rates of different kinds of journals </li></ul></ul>1.89bn Non-cash Peer review
    8. 8. Publishing <ul><li>Fixed costs for </li></ul><ul><ul><li>article selection </li></ul></ul><ul><ul><li>management and processing of peer review </li></ul></ul><ul><ul><li>editing and proof-reading </li></ul></ul><ul><ul><li>composition and typesetting </li></ul></ul><ul><ul><li>illustrations and graphics </li></ul></ul><ul><ul><li>rights management </li></ul></ul>Direct fixed cost 1.80bn
    9. 9. First copy costs <ul><li>peer review cost plus </li></ul><ul><li>fixed publishing costs </li></ul>First copy cost 3.70bn
    10. 10. Distribution <ul><li>variable costs </li></ul><ul><ul><li>sales administration, online user management, printing, inventory management etc </li></ul></ul><ul><li>indirect costs </li></ul><ul><ul><li>marketing, online hosting, customer service/helpdesk, management & administration </li></ul></ul><ul><li>surplus </li></ul><ul><ul><li>profit or surplus for distribution, investment, or use in other activities </li></ul></ul>Indirect cost 0.96bn Surplus 0.82bn Variable cost 0.97bn
    11. 11. Access and usage Access provision User search And print cost Reading cost 2.1bn 16.4bn 33.9bn
    12. 12. Access <ul><li>Library costs </li></ul><ul><ul><li>extrapolated from UK figures </li></ul></ul><ul><ul><ul><li>SCONUL </li></ul></ul></ul><ul><ul><ul><li>Schonfeld and King 2004 </li></ul></ul></ul><ul><ul><li>fixed and variable costs </li></ul></ul><ul><ul><ul><li>Space and shelving; time spent on acquisitions, registration etc; cataloguing; preservation; IT systems; access management etc </li></ul></ul></ul><ul><ul><li>excludes subscription costs ( to avoid double counting ) </li></ul></ul>2.1bn Access provision
    13. 13. Search, print etc <ul><li>Readers’ costs </li></ul><ul><ul><li>Time spent in searching, gaining access, downloading, printing etc </li></ul></ul><ul><ul><li>( Tenopir and King estimates) </li></ul></ul><ul><ul><li>excludes costs of aggregator and A&I services </li></ul></ul>User search & print cost 16.4bn
    14. 14. Reading <ul><li>Calculation of costs of time spent in reading </li></ul><ul><li>( Tenopir and King estimates ) </li></ul><ul><li>A measure of value as well as of cost </li></ul>Reading cost 33.9bn
    15. 15. How are these costs being met? <ul><li>publishing and distribution </li></ul><ul><ul><li>see next slide </li></ul></ul><ul><li>access </li></ul><ul><ul><li>costs met by libraries and those who fund them </li></ul></ul><ul><li>search, download and reading </li></ul><ul><ul><li>costs met by researchers and those who employ them </li></ul></ul>
    16. 16. Meeting the costs of publication and distribution <ul><li>estimates for the system as a whole </li></ul><ul><li>differences for different types/categories of journal </li></ul>
    17. 17. Meeting the costs of scholarly communications 5.0 40.6 1.2 11.6 0.5 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 HE Libraries HEI's and their funders Non-HE Libraries, special libraries and Individual subscribers Funders/employers of Non HEI researchers other £ Billions
    18. 18. So what? <ul><li>RIN role to question how efficient and effective are the information services and resources provided for and used by the UK research community </li></ul><ul><li>Clearer picture of where major costs arise, and how they are funded, enables us to </li></ul><ul><ul><li>focus attention on key areas where cost efficiencies are most likely to arise (eg peer review??) </li></ul></ul><ul><ul><li>analyse the balance of trade between different sectors and different countries (eg UK or EU contribution) </li></ul></ul><ul><ul><li>develop scenarios of possible changes, and model their impacts both on costs and on how/where those costs are met </li></ul></ul>
    19. 19. Increases in research funding and article production over 10 years: Costs <ul><li>Publishing and distribution costs </li></ul><ul><li>Real terms increase of £1.6bn (25%) </li></ul>
    20. 20. Increases in research funding and article production over 10 years: Funding <ul><li>Sources of funding and other contributions </li></ul>
    21. 21. E-only publication <ul><li>~£1bn cost savings, split between </li></ul><ul><ul><li>publishing/distribu-tion (~5% reduction) </li></ul></ul><ul><ul><li>access provision (~36% reduction) </li></ul></ul>
    22. 22. Some key messages <ul><li>Journal publishing and distribution are pivotal, but only part of, the scholarly communications system </li></ul><ul><ul><li>c 3.6% of the overall costs </li></ul></ul><ul><li>Much larger costs incurred by readers in search, print and reading </li></ul><ul><ul><li>but those costs are also measures of value </li></ul></ul><ul><li>Costs are met overwhelmingly (>78%) by the higher education sector </li></ul><ul><li>Costs will continue to rise in real terms </li></ul><ul><li>There is scope for savings, and improvements in efficiency and effectiveness, across the system </li></ul><ul><ul><li>and they are not just, or even mainly, in publication and distribution </li></ul></ul>
    23. 23. Thank you Michael Jubb Research Information Network Activities, costs and income flows report available at Strategic goals for public policy statement available at