A common view of journal prices is that they go up while, in real and comparative terms, the ability libraries have to afford and develop them goes down. It is more complex than that but nevertheless the concern about journal prices resonates persistently among professionals. Another over‐simplification is that publishers want to push prices up and librarians want to keep prices down, and again things are not as simple. The need, then, for reliable and objective information is of particular value, above all where half‐knowledge and high emotion come into play.
LISU have specialized in providing helpful information for many years. Their statistical series (like trend analyses of annual library statistics, average prices of books, and surveys of libraries) is well‐known. Two readers' guides (one for adults and another for children) are deservedly doing well. There are also research reports covering topics like benchmarking, copyright compliance, and acquisition, and the work under review here, on scholarly journal prices, designated a LISU occasional paper no 37, is one of these reports. It follows up an earlier study by White and Creaser (2004) called Scholarly Journal Prices and this can be accessed at the LISU website (given at the end of the review). It is interesting in itself to see how LISU have carried out this research on behalf of Oxford Journals, revealing it seems that a for‐profit approach will increasingly be taken by LISU given much‐heralded changes in funding.
You cannot cover everything and so White and Creaser have sensibly focused on two types of scholarly journal – biomedical and social science – and on 11 “big” publishers (eight commercial and three university presses). They are Oxford Journals and Blackwell Publishing, Cambridge University Press and the University of Chicago Press, Elsevier, Lippincott Williams & Wilkins, Nature and related titles, Sage and Springer, Taylor & Francis and Wiley. Some 3,000 biomedical journals and some 2,000 social science journals were identified (from data based on sources like Swets and the ISI JCR listings).
The key things of interest were price and impact. Impact was based on citations in the normal way. Attention was paid to the different dissemination behaviours in the two disciplines. Relationships (correlations in fact) fed through to price‐per‐page, and with these and other parameters, median and other comparisons as well as trends were examined. Such information is understandably of particular relevance to those constituencies most implicated in the field as a whole, certainly librarians in academic and specialist libraries, anyone involved with budgets and cost‐effectiveness, and anyone in (or interested in) the publishing dimension. Over‐arching the whole debate, of course, is the shift to electronic/digital journal publishing and online delivery in recent years, a topic not directly examined here (rightly) but relevant for anyone reading the report itself.
The key findings emerge clearly in the executive summary, and this is probably where readers will pick up the initial information (from the website) and then decide whether to buy the paper copy. The advantage of the paper copy is that the graphics are in full colour, enabling easy comparisons in graphs (e.g. between percentile groupings of journal prices, and year break‐downs in comparative correlations for all the publishers). The report itself is structured so that the scene is set by the introduction, this is followed by an overall analysis (both types of journal and all publishers), and then twelve further chapters each one on each publisher. At the end comprehensive listings of biomedical and social science subjects and of the actual journals in each category are provided, providing ample evidential support for the data analysis itself.
Given the interest in prices and impact, and relationships between them (and given too that even the most quantitative relationships are themselves susceptible to some justifiable scepticism, e.g. about what impact means, let alone whether price‐per‐page is a criterion of quality or merely an indicator of physical production), the banner headlines of the report will have the most impact (no pun intended). For biomedical journal titles, median prices during the period 2000‐2006 ranged from £198 to £859. Such things are attributable of course to specific publishers, and in this case the first to Cambridge University Press and the second to Elsevier.
Even from this initial snapshot data, we might find it convenient to start building up a theory about university presses and commercial players, or even about differentials between publisher X and publisher Y – there has after all been bloggery about certain demons in the industry – but White and Creaser remain calmly objective and interpretive, leaving inferences to be made by their readers. Wider contrasts might be deduced from, say, UK‐based and US‐based and world‐wide publishers, and, of course, between biomedical and social science fields in themselves. I can foresee some readers going to this document for merely one of these.
Other key findings (for biomedical journals) include these: increases in median prices varied from 42 (Oxford Journals) to 104 per cent (Sage), more than a few journals had no identified impact factor, and the median price‐per‐point‐of‐impact‐factor ranged from £186 (Oxford Journals) to £552 (Springer) in 2006. Turning to social science journals, median prices ranged from £119 (University of Chicago) to £513 (Wiley), price‐per‐page from 13p (University of Chicago) to 93p (Wiley) in 2006, and changes in the median price‐per‐point‐of‐impact‐factor between 2000 and 2006 varied from a fall of 20 per cent (Elsevier) to an increase of 83 per cent (Springer). Set this within an overall scenario of a 39 per cent increase in journal prices and a rise of 16 per cent in the retail price index, and factor in differentials from each discipline (and, it has to be admitted, a lot of other things, like, say, differences in financial structure between, say, Elsevier on the one hand and Cambridge University Press on the other), and you are starting to build up a coherent picture.
The analysis chapter (chapter 2) gives a good overview for purposes of comparison. Of course, some prices (and median prices) start higher than others, and others rise more steeply than others. Different publishers publish different numbers of journals, too. Impact factors, also, have behaved in a complex way – we might assume (from improved methods of analysing citations, research assessment exercises, and the growth of performance measurement) that the trend would be unanimously upwards for all journals, but, again understandably, this is not so: commercial factors and rivalries creep in, of course, disciplines change, editorial reputations shift about, substitution takes place in crowded fields, and libraries (and individual subscribers) can take only so many. Rightly, subscriptions alone are not regarded as a thoroughly reliable measure of impact since they emanate merely from budgets. The correlation graphs are revealing and allow easy comparison, as indeed to the bar charts of journal rankings (for 2005) into top 10 per cent and then lower quartiles.
The profiles of individual publishers make equally interesting reading. Predictably, Nature publications are high cost‐high value: looking at “most expensive”, an increase of 8.5 per cent in 2000‐2001 itself increased to 25 per cent in 2005‐2006, elevating the price‐per‐impact factor (£) too from 415 to 783 (making, respectively, for the two sets of figures a percentage change between 2000 and 2006 281 and 89 per cent). Turning to Sage social science journals (based on 161 titles included), for the most expensive we find a percentage change between 2000 and 2006 of 146 per cent, while for social science journals for Taylor & Francis a mere 9 per cent. We are then impelled (I think and hope) to consider whether we have apples and oranges in such cases, above all when we consider the profile of journals by price and subject across each publisher.
There is a helpful and concise list of sources on such analyses – most of the work has been done in the USA – and this includes earlier work by White and Creaser. Inevitably, not everything could conceivably have been included – journals with not enough continuity for example, bundling of titles with other products, discounting and membership subscriptions, and all the other factors in play in real life. Crossovers, too, between biomedical and social science journals have presented some challenges. Different pricing models are in use in both libraries and publishing houses, and these models in themselves are undergoing radical revision because of the internet. To sum up, if the trends are upwards for prices, then so are they for impact. Publishers fear negative correlations here, and so, down the supply chain, do librarians. This report, then, sheds valuable light on its field, provides insights that readers can and should take further forward, and demonstrates a new flavour in LISU's own strategy for the future.
