The Crisis and the Transition from Ownership to Access

Scientific Growth and Price Inflation

The current serials crisis engulfing academic libraries is rooted in the very nature of scientific growth. Price (1986, 4–29) brilliantly described this nature a generation ago. According to Price, the normal mode of scientific growth is exponential, and in this respect it agrees with the common natural law of growth governing the number of human beings in a country, the number of fruit flies growing in a bottle, or the number of miles of railroad built in the early Industrial Revolution. However, the law of scientific growth is marked by two remarkable features. First, the exponential law of scientific growth holds true with high accuracy for long time periods extending for centuries. Second, scientific growth is surprisingly rapid, outstripping that of the size of the population and nonscientific institutions.

It is in the latter feature that the roots of the current serials crisis should be sought. In Price's view, all exponential growth curves must ultimately hit an upper limit and flatten into logistic curves, and such a flattening process is marked by violent fluctuations of the curve and prolonged periods of crisis. With telling prescience, Price (p. 28) predicted for science just such a period of crisis marked by "rapidly increasing concern over those problems of manpower, literature, and expenditure that demand solution by reorganization."

As part of his analysis of scientific growth, Price (1986, 5–8; 1975, 164–73) dealt with the problem of scientific journals. He stated that the exponential increase in the number of scientific periodicals has proceeded with an extraordinary regularity seldom seen in any human-made or natural statistic ever since the earliest surviving such journal, Philosophical Transactions of the Royal Society of London, was first published in 1665. Price estimated that starting from 1750, when there were 10 scientific periodicals, the number of such periodicals has increased by a power of 10 every half century, which has lead to a doubling every 15 years. Taking a longer view, he calculated that this corresponded to a factor of 1,000 in a century and a half and of 1,000,000 since the mid-seventeenth century. Price compared the growth of scientific journals to that of a colony of rabbits breeding among themselves and reproducing ever so often.

Price might have overestimated the growth in the number of scientific journals, because he did not exclude discontinued serials (Line and Roberts 1976, 128). Nevertheless, his estimates take on a frightening reality as soon as one considers the constantly expanding coverage of the standard reference source on serials, Ulrich's International Periodicals Directory. Whereas the 20th edition of Ulrich's for 1981 (vii) together with its companion volume Irregular Serials & Annuals (6th ed. 1980–81) listed some 96,000 titles, the 34th edition of Ulrich's for 1996 (vol. 1, vii) contained information on nearly 165,000 titles including irregulars and annuals–a gain of 71.9%. As a base of comparison, it should be noted that the first edition of this publication (Ulrich 1932, ix) covered 6,000 titles.

By itself, the exponential growth in the number of scientific serials would have been a difficult enough problem for academic libraries to handle. However, the problem has been immensely compounded by an extraordinary inflation in serials prices. An idea of the extent of this inflation and the role of scientific serials in it can be gained from analyzing the data published annually in the U.S. Periodical Price Index (USPPI) (Carpenter and Alexander 1996). Excluding Russian translation journals, the average price of a U.S. periodical rose 154.8% from $65.00 in 1986 to $165.61 in 1996. This rate of inflation exceeded both the general inflation rate as measured by the U.S. Consumer Price Index (CPI) and that in the cost of higher education as measured by Higher Education Price Index (HEPI).

Thus, from 1986 to 1995 the USPPI rose 129.9%, while the CPI increased 39.0%; whereas from 1986 to 1994 the USPPI gained 108.3%, the HEPI went up 41.2%. The most expensive subject category in the 1996 USPPI is Chemistry and Physics. Its inflation rate by far outstripped that of the overall USPPI, and the average price of chemistry and physics periodicals rose 228.4% from $264.05 in 1986 to $867.00 in 1996. The rapid increase in chemistry and physics serials prices greatly affected the overall structure of U.S. periodical prices, and this was evident in an exploding Gap Factor, which was calculated by dividing the average price of the highest priced subject category by the average price of the lowest priced subject category after discarding Russian translations and children's periodicals as constant outliers. In both 1987 and 1996 Chemistry and Physics was the highest cost subject category, but the Gap Factor surged from 11.49 in 1987 to 22.02 in 1996. The ultimate result of this process is evident from the fact that, with again the exclusion of Russian translations and children's periodicals, while chemistry and physics serials comprised only 4.7% the titles of the periodical sample used to construct the 1996 USPPI, these serials accounted for 23.9% of the total cost of this sample.