College and Research Libraries An Economic ''Commons'' Tragedy for Research Libraries: Scholarly Journal Publishing and Pricing Trends Gary D. Byrd The recent dramatic increases in both the numbers and prices of scholarly journnls are evidence of a dis- torted economic marketplace for scholarly information. They can also be viewed as evidence of an impend- ing economic "commons" tragedy for this nation's research libraries. The general economic principles underlying this tragedy were popularized in a nontechnical and very influential 1968 essay by Garrett Hardin called ''The Tragedy of the Commons. ''The strategies needed to manage our research library com- mons effectively will require a fundamental reshaping of the present system of scholarly communication. The leadership for making the necessary changes must come from the research library community. . • . he venerable, centuries-old sys- tem of scholarly publishing and -· . collection building in research libraries is in very real danger of collapse. Dramatic increases in both the numbers and prices of scholarly journals, especially scientific scholarly journals, are the most obvious manifestation of this im- pending crisis. For the past two decades journal prices in the United States have outpaced general inflation by rates that would be totally unacceptable in other sec- tors of our economy. In 1989, for example, journal prices increased by more than two times the Consumer Price Index (CPI) and, as recently as 1987, the rate of price inflation was more than five times the CPI. 1 Not only do prices increase dramati- cally, but the number of new journals also continues to "proliferate," as Herb White puts it, ''in total disregard of all economic indicators." "One publisher," he notes, ''reported that over a period of time he had started 180 new journals and canceled five.'' 2 To date, none of the attempts by re- search libraries to cope with these dual cri- ses of information glut and spiraling price inflation have been very successful. In fact, the most common strategies, that is, pleading for a larger acquisitions budget, or transferring dollars from monographs to the serials budget, have actually aggra- vated the problem by accommodating or hiding it rather than dealing with underly- ing causes. Even worse, argues White, these tactics destroy ''any remaining ves- tige of a suggestion that library materials allocations as between monographs and serials [follow] any sort of professionally developed library plan."3 Clearly, current distortions in the scholarly information marketplace threaten the ability of re- search libraries to continue to carry out their mission of collecting, preserving, and providing access to the scholarly rec- Gary D. Byrd is Assistant Director for Finance, Planning, and Research at the Health Sciences Library, Uni- versity of North Carolina, Chapel Hill, NC 27599. @1990 Gary D. Byrd. 184 ord. Since research libraries are central to our current system of scholarly communi- cation, a threat to their basic mission must be ·considered a threat to scholarly com- munication in general. A puzzling and troubling aspect of this problem for research librarians has been the fact that until very recently (and to a large extent, even today) the scholars and publishers, who create, market, and ulti- mately consume journals and other rec- ords of scholarship, have seemed to ig- nore or to be unaware of the problems this explosion in scholarly journal numbers and prices is causing for libraries. Herb White suggests that this is because re- search libraries find themselves in a sort of never-never land between the ''true ven- dor/client relationship" of scholarly pub- lishers and the scholars who write for and then read their publications . Thus, "pub- lishers don't treat us as customers, they treat us as purchasing agents, who simply perform the routine tasks we are told to perform." And because scholars, who do not directly experience the price increases, show "no objection to any of [the pub- lishers' pricing] tactics ... , [they] are fu- rious not with the journal publishers but with us [when we] ... hesitantly, apolo- getically . . . apprise them of this prob- lem.''4 The economic relationship among the producers, consumers, and distributors of scholarly publications is central to under- standing both the causes of, and the po- tential solutions to, this impending break- down in our system of scholarly communication through research li- braries. The sections that follow review key papers and monographs dealing with the microeconomic principles that under- lie the relationship among scholars, pub- lishers, and research libraries. The theories of supply and demand and the way these affect the exchange of goods and services among the three direct partic- ipants in the scholarly communication system are important for an understand- ing of the individual components of the system. But I will argue that they must be seen as part of a much more basic and fun- damental economic principle. This princi- ple, usually called the "Tragedy of the Economic 11Commons" Tragedy 185 11Hardin describes how shared pub- lic resources (the 1 Commons') may eventually be destroyed when indi- viduals are allowed unrestrained freedom to exploit them for personal gain." Commons," was vividly described by Garrett Hardin in his 1968 essay by that ti- tle in Science. 5 The essay deals primarily with the way cultural norms of individual independence and freedom of action en- danger our common environmental re- sources. However, the "tragedy" Hardin describes has much wider economic impli- cations for the way societies manage all shared public resources, including there- sources in our research libraries. Briefly, Hardin describes how shared public re- sources (the ''commons'') may eventually be destroyed when individuals are al- lowed unrestrained freedom to exploit them for personal gain. The current distortions in the scholarly information marketplace are evidence of a commons tragedy centering around our nation's research libraries, which share many of the characteristics of Hardin's commons. But, before describing this tragedy in more detail, we need to review the characteristics of scholarly informa- tion resources and the microeconomic re- lationships among publishers, research li- braries, and scholars. THE NATURE OF SCHOLARLY INFORMATION One essential key to understanding these economic relationships is the re- source that is produced and consumed, that is, scholarly information. When re- garded as a commodity, scholarly infor- mation, or just ''information'' or ''knowl- edge" (synonyms used by many who write about these issues) has many quali- ties that distinguish it from tangible prod- ucts, like automobiles or wheat, on which classic economic theories are based. Harlan Cleveland posits the following list of unusual economic characteristics of in- 186 College & Research Libraries formation: it is expandable, compressible, substitutable, transportable, diffusive, and sharable. 6 The last of these character- istics, he says, the propensity for this shar- ing resource to leak, is ''eroding the doc- trine that knowledge can be owned, exchanged, and monopolized the way 'real' resources can. " 7 Pat Molholt ex- pands on Cleveland's list with the follow- ing additional characteristics of informa- tion. We are dealing with: • a commodity that does not depreciate, with all this implies for financing, tax structures, and accounting principles; • a resource that is freely available unless artifi- cially impounded; whereas economic theo- ries focus on scarcity; • an intangible that can render people jobless and force whole organizations to restructure in order to survive; • a resource that thrives on reuse, repackaging, recycling; one that has been growing expo- nentially and under completely unconven- tional rules; • a resource that has, as an important aspect of its use, something uncontrollable and unpredictable-serendipity. 8 Robert Taylor describes four other impor- tant characteristics of scholarly informa- tion that can help us understand the ways publishers, scholars, and research li- braries produce and exchange this unique economic good: (1) "A person cannot know before seeing (hearing) . . . infor- mation, whether or not it is of use." (2) Knowledge ''about a chunk of information . . . in some ways decreases the need for that chunk because one may already pos- sess a part of the information.'' (3) ''Many people can know the same facts, or own the same chunk of information without depriving others." And (4) "we tend to overproduce and to overconsume infor- mation."9 Together, these characteristics of schol- arly information show it to be a poor ex- ample of the pure private goods that econ- omists use to model the interaction of buyers and sellers in a competitive mar- ket. In a perfectly competitive market, many rival firms produce and sell stan- dardized goods or services at quantities and prices determined by the total supply of, and total demand for, those goods or May 1990 services. Perfect information about the quantity and quality of those goods and services is available to all producers and consumers. Finally, and most importantly for an understanding of the scholarly in- formation "marketplace," the sale and purchase of goods and services in a com- petitive market causes no costs or benefits to third parties not participating directly in the sale or purchase transactions. (In eco- nomic jargon, there are no "externalities" to these transactions.) On the other hand, scholarly informa- tion does share many, but not all, of the characteristics of public or social goods. Social goods are often, but not always, produced and provided to consumers by government or other public agencies, or by employees of these agencies. More im- portantly, they are goods that cannot eas- ily be excluded from individuals who fail to pay for the service or benefit received. This ''free rider'' effect brings to mind Cleveland's observation that information has a propensity to "leak." Pure public goods can provide benefits simulta- neously to more than one individual (of- ten whole populations) without any addi- tional cost. (Again, in technical economic jargon, the production of the good for the benefit of any one person results in posi- tive externalities for all persons in a society and zero marginal costs for those extra benefits.) Fritz Machlup argues that knowledge is, in fact, a public or social good ''of the pur- est type. . . . There may be a cost of the transfer of knowledge, of teaching it and learning it, but there is no additional cost of using it once it has been acquired. " 10 Lawrence White notes that because the marginal cost to disseminate information "is frequently very low or zero, ... we ought to be encouraging ... [its] maxi- mum distribution . . . once it is in exis- tence.'' But, maximum free public distri- bution ''may well interfere with the long run considerations of producing a flow of new information ... [since] people are not going to invest resources in producing information unless they think they are go- ing to get a net gain from it."11 This eco- nomic conflict between the private, mar- ketplace incentives for the production of new information and the public, social in- centives to subsidize wide dissemination of existing information is one of the keys to the tragedy of the commons explored below. While basically a public or social good that is widely available at little or no cost, scholarly information also has char- acteristics that make its production and some forms of distribution, like journals, financially attractive to market for com- mercial profit. This leads directly to the question of economic value. THEVADUEOF SCHOLARLY INFORMATION Defining an accurate economic value for scholarly information is complicated by uncertainty. In contrast to most commodi- ties, like coffee, for which "people usually have a pretty good idea of how much [more] ... would contribute to their hap- piness," notes Machlup, "we cannot know what a piece of knowledge may be worth to us before we know what it is. "u (Remember Taylor's observation that it is impossible to know about information be- fore seeing or hearing it, and that, ironi- cally, knowledge about a chunk of infor- mation decreases our need for it.) ''Because scholars can never be sure what piece of information may prove useful . . . they have a strong incen- tive to encourage publishers to pub- lish everything which might be use- ful and to encourage research libraries to buy everything pub- lished." In his 1970 dissertation on the scientific journal market, Sanford Berg noted that scholars ''cannot be completely certain about which items in [a] ... stock of scien- tific information will be needed." Thus, ''the result is a tendency to be complete rather than selective when organizing a ref- erence library. '' 13 Because scholars can never be sure what piece of information may prove useful (and because serendip- ity often leads to unexpectedly useful in- Economic "Commons" Tragedy 187 formation), they have a strong incentive to encourage publishers to publish every- thing which might be useful and to en- courage research libraries to buy every- thing published. Another more technical aspect of this uncertainty characteristic of scholarly in- formation has been described more re- cently by Machlup14 and Aatto Repo. 15 They argue that the failure to distinguish between the "use value" (or "value-in- use'') and the ''exchange value'' of schol- arly information leads to inaccurate esti- mates of value. Use value includes the qualitative measures of how successfully individuals use the content of particular in- formation products to accomplish infor- mation tasks. Exchange value includes economic measures of the marketplace where individuals and organizations pro- duce, store, and exchange information products and services. Exchange values can be measured using fairly objective em- pirical data, while use value can only be measured subjectively by individuals, based on what they expect or perceive. The problem in much of the literature on the economics of information, says Repo, is that researchers ''skip from exchange values to value-in-use . . . without realiz- ing the implications of the change.' 116 Ironically, Sanford Berg's study, which accurately describes this uncertainty char- acteristic of scientific information, 17 has been strongly criticized by Machlup for failing to make the distinction between use value and exchange value. Machlup argues that Berg and others are '' overzeal- ous'' in their attempts to overcome the un- certain value of information by trying to quantify or objectively measure the total benefits or value to society of scholarly in- formation. They do this by substituting projected individual estimated benefits (use value) for actual measures of the in- formation marketplace (exchange value). ''Most economists have long since aban- doned as hopeless and irrelevant the aim of measuring the total utility or total bene- fits . . . of a class of good or service,'' says Machlup, because "demand ... for any one good or service is [based on]. . . the assumption that the prices of all other goods and services are given and un- 188 College & Research Libraries changed, and hence the benefits which buyers of a particular good obtain. . . would depend on millions of other prices. ''18 This argument, though rather technical and complex, is also another key to the commons tragedy of research libraries. When scholars, publishers, and librarians speak about the "value of scholarly infor- mation,'' they usually mean the perceived or expected use values to individuals of particular information products, rather than the exchange value of those products in the marketplace. But when these neces- sarily subjective, qualitative measures are used to justify the overall economic im- portance of scholarly information, value estimates (not restrained by market reali- ties) become inflated. When scholars are free to create new in- formation without considering the eco- nomic market for that information, and, at the same time, are free to use published information at little or no cost, they are more likely to overestimate its value. Simi- larly, when publishers and librarians hear from the scholarly community only about the high actual or potential use value of this information and see continuing gen- erous public financial support for its distri- bution and acquisition, their natural ten- dency is to distribute and acquire as much as possible. This surely demonstrates Tay- lor's last characteristic of scholarly infor- mation, its overproduction and overcon- sumption. The next question, then, is exactly how, in microeconomic terms, publishers, research libraries, and scholars interact with each other in the scholarly information marketplace. THE SCHOLARLY INFORMATION MARKETPLACE The public good and uncertainty charac- teristics of scholarly information lead to distortions in the ideal competitive mar- ketplace that economists use to predict supply, demand, and prices of goods and services. In a competitive market one would expect the total production and prices of relatively uniform scholarly jour- nals to be balanced by the limited demand for individual journal titles. Instead, this marketplace is distorted by the dual and May1990 differential pricing strategies of scholarly publishers. Most publishers now charge one subscription rate to individuals and another substantially higher rate to li- braries. And many European publishers charge substantially higher prices to U.S. libraries than to those in the rest of the world. Economists call these practices price discrimination, and define it more precisely as ''selling the same product to different buyers at different prices, where the difference in price does not reflect cost differences in producing or selling the product." 19 Price discrimination can only take place in noncompetitive markets (such as mo- nopolies or competitive monopolies) where sellers can separate potential buy- ers into submarkets that respond differ- ently to higher prices. Each submarket has a different ''elasticity of demand'' (or the product has different "price elasticities" in each submarket). Price elasticity is mathematically expressed as the percent- age change in quantity demanded divided by the percentage change in the product price. It measures the responsiveness of buyers to changes in price. Highly elastic markets are those where higher prices quickly reduce demand or where lower prices quickly increase demand. Inelastic markets are those where price changes have little effect on demand. 20 Current laws provide for the free trans- fer of copyright ownership from the scholars who write journal articles to the publishers who print, market, and distrib- ute them. This effectively removes any marketplace incentives from the scholarly community. It also provides scholarly journal publishers with enough monop- oly power that they can divide the market into separate submarkets with different price elasticities. Margaret Quinlin, an ed- itor at Aspen Publishers, recently wrote a special report for the Society for Scholarly Publishing. (SSP), in which she states clearly how important publishers feel price elasticity is in setting prices for their publications. "Price elasticity has a direct bearing on the pricing of publications," she says. "The publisher who lowers the price of . . . [a specialized publication with a 'price-inelastic demand schedule'] will not increase [sales] volume significantly. . . . In short, total revenues will be re- duced and circulation will not be in- creased. ''21 Individual demand for scholarly jour- nals is more elastic than library demand. As early as 1977, Fritz Machlup pointed out the extent to which this is true. ''Pub- lishers . . . have given up the idea of sell- ing to individual buyers [only],'' he said, ''and are determined to charge what the traffic will bear in the supposed inelastic range of the demand curve-the research libraries."22 In meetings with librarians, publishers argue that price increases to libraries result from overall increases in the fixed and variable costs of publishing, 23 the fluc- tuating value of the dollar abroad, which makes it necessary to compensate for ex- pectations of a weakened dollar b/' setting higher prices in the U.S. market, 2 and the many new and increasingly narrow scien- tific and scholarly specializations, which force fixed costs to be distributed over a narrower base. 25 Publishers also argue that, although the marginal costs of pro- ducing journals for these different mar- kets are about the same, the value or mar- ginal benefit of the journal to each submarket is different enough to justify a higher price to research libraries. They reason that in libraries journals will be used by many readers, but the individual subscriber's copy will only benefit one, or, at most, a few. The magnitude of recent price increases and the size of the price differentials, how- ever, have convinced most research librar- ians that "publisher profit is the driving force behind the recent escalation of serial prices," 26 and that "publishers of aca- demic journals have sufficient monopoly power to engage in price discrimina- tion."27 In a recent editorial, James Thompson singled out for particular criti- cism those "few publishers, owned and directed by large multinational holding companies,'' which openly seek monop- oly power. He quotes Robert Maxwell, the owner of the Pergamon journals, who out- lined his business strategy thus: ''If Perga- mon could win the trust of scientists it could establish the ·standard journal in Economic "Commons" Tragedy 189 each specialization, and that would give it a series of publishing monopolies . . . sci- entists are not generally as price- conscious as other professionals, mainly because they are not spending their own money.'' 28 At a fall 1988 seminar on the future of scholarly journals sponsored by the Soci- ety for Scholarly Publishing (SSP); librari- ans, publishers, and scholars exchanged views on the present and future economic viability of the journal system as it pres- . ently exists. Mr. Maxwell's assessment of the price-consciousness of scholar/scien- tists is corroborated in large part by the re- marks of Robert Peet, a biologist who par- ticipated in the SSP seminar. ''As a scholar," said Peet, "I am reasonably happy with the current status of scholarly journals." He did acknowledge the exis- tence of both a ''library problem'' and a "publisher problem" caused by the in- creasing "store of human knowledge" and the exponential growth in the number of researchers worldwide and their result- ing publications. But he argued for "one fundamental principle ... [to] constrain all solutions to these problems. That is, there must be free access for all scholars to journal articles. We will tolerate, grudg- ingly, travel as a necessity. But, we will not tolerate a strict pay-for-use system.'' 29 To the extent that this view is held by most scholars, any hope that a competitive, free-market, vendor/consumer relation- ship can be established between pub- lishers and scholars is probably unrealis- tic. In summary then, the scholarly infor- mation marketplace is characterized by producers (academic scholars) who turn over gratis, through copyright transfer, the ownership of their products (scholarly journal articles) to sellers (scholarly pub- lishers), who in turn earn a profit, not by selling to the ultimate consumers (again academic scholars), but to publicly sup- ported agencies (research libraries) acting on behalf of the entire scholarly commu- nity to organize, store, and provide free or low-cost access to these products. As Herb White puts it, "natural selection and the pressures of the marketplace simply do not apply here." 30 190 College & Research Libraries The product exchanged in this market- place, scholarly information, has many characteristics which make it quite differ- ent from free-market goods and services. Most notably, it easily "leaks," providing extra free or low-cost benefits to the schol- arly community at large; and yet wide public distribution will in the long run dis- courage the primary mechanism now available for its distribution (scholarly publishing). Finally, the economic value of this product is distorted because its ex- pected or perceived use value for individ- ual consumers is confused with exchange value in the scholarly information market- place. How has this economic situation become so dangerously distorted when it is in our collective best interest to keep it healthy? The principle underlying the "tragedy of the commons" may provide a key to the answer. THE TRAGEDY OF THE COMMONS The central story and dilemma of Gar- rett Hardin's 1968 essay in Science can be briefly summarized as follows. All the herdsmen of an agrarian community use a free open pasture or "commons" to graze their cattle. This arrangement works rea- sonably well for centuries because wars, disease, etc. keep the number of herds- men and their cattle well below the capac- ity of the land. Finally, however, a day of reckoning comes when the pasture is full and the ''inherent logic of the commons remorselessly generates tragedy.'' Each economically "rational" herds- man seeks to get the most benefit possible from the commons and asks himself, ''What value will I gain or lose by adding one more animal to my herd?" The posi- tive value will equal all the proceeds from the eventual sale of the additional animal. The negative value will equal the over- grazing damage to the commons caused by one more cow. But, since this negative value is shared by all the herdsmen, the rational herdsman concludes that his share of this damage is much smaller than the profit one more cow will generate. The tragedy inevitably follows because each and every rational herdsman sharing the commons reaches the same conclusion. May1990 The end of Hardin's little drama is worth quoting verbatim: Therein is the tragedy. Each man is locked into a system that compels him to increase his herd without limit-in a world that is limited. Ruin is the destination toward which all men rush, each pursuing his own best interest in a society that believes in the freedom of the commons. Freedom in a commons brings ruin to all. 31 Hardin goes on to point out that problems of pollution are examples of the tragedy of the commons in a reverse way. Instead of exploiting and thereby, depleting or tak- ing something out of the commons, ra- tional users decide to put in damaging things like sewage, toxic wastes, noise, or unpleasant advertising signs. 32 · "Perhaps most sadly of all, we re- search librarians, who are charged with managing this commons, also encourage the exploitation and pollu- tion of the research library commons, and compound the tragedy, by insist- ing on comprehensive collections and unlimited free access for any and all potential users.'' Unfortunately, both the depletion and pollution manifestations of the tragedy of the commons can be applied to the current economic crisis in our system of scholarly communications. Publicly supported re- search libraries share many of the charac- teristics of the commons. Publishers, scholars, and even research librarians are all guilty of depleting and/or polluting this commons in the ways they "rationally" seek maximum benefit from its economic (budget) and scholarly information (books and journals) resources. The price dis- criminating publisher, like the herdsman who chooses to add more cows to his herd, exploits the research library com- mons by rationally deciding to maximize profits and charge whatever the traffic will bear in the inelastic range of the demand curve. The scholar who furthers his or her career or research goals by publishing as much as possible, and then insisting on free access to all other published informa- tion that might be of use, shares attributes with the independent, free-enterpriser who fouls his own nest with pollution. Perhaps most sadly of all, we research li- brarians, who are charged with managing this commons, also encourage the exploi- tation and pollution of the research library commons, and compound the tragedy, by insisting on comprehensive collections and unlimited free access for any and all potential users. A deep-seated belief in the efficacy of the free-market system and Adam Smith's "invisible hand" (that is supposed to cre- ate greater prosperity for all if we each fol- low our own best interest) underlies the actions and views of each of these partici- pants in the scholarly information market- place, including research librarians. Katina Strauch, the head of Collection De- velopment at the College of Charleston Li- brary, in a recent talk on the economic re- lationships among librarians, vendors, and publishers, argued passionately and effectively for the need to preserve our competitive system where ''the govern- ment does not dictate whom you must deal with or at what price. ''33 But when we face the problem of managing a commons, says Hardin, ''we can make little progress . . . until we explicitly exorcise the spirit of Adam Smith ... [who] contributed to a dominant tendency of thought that has ever since interfered with positive action based on rational analysis, namely, the tendency to assume that decisions reached individually will, in fact, be the best decisions for an entire society.' '34 Economists, on the other hand, realize that conditions of free choice for con- sumers and profit maximization by pro- ducers do not always result in socially de- sirable consequences. In 1977, economist Fritz Machlup pointed out the long-term consequences of making the scholarly publishing industry dependent on what he called the "grants economy," that is, the resources produced and paid for by re- cipients of public or private grants instead of private consumers paying out of their incomes. "Should trends of the recent past continue," Machlup said, "it would be impossible for the private sector of the Economic "Commons" Tragedy 191 [publishing] industry to survive if it had to rely on . . . sales to private buyers not aided by grants. Whether the industry can be viable in the grants economy at the present rates of growth of costs on the one hand, and of gt:ants on the other hand, is the question. " 35 In a 1979 review of the theoretical basics of economic analysis for Library Trends, Richard McKenzie in- cluded the following statement as a major element in his ''economist's paradigm'': There is a tendency for individuals within very large groups to fail to pursue "common goals" even when the goals are agreed upon by all group members. Therefore, voluntary collective action is not likely in very large groups. [italics in the original]36 This is a very elegant synopsis of the crux of the tragedy of the commons. Unfortu- nately, McKenzie did not go on to explore the implications for library economics of this part of his paradigm. , Herb White has been one of the few li- brarians to state clearly some of these im- plications. For instance, here are his con- cluding remarks to a group of scholarly publishing executives: ''unbridled growth in the number of publications, in their size, and in their price, with all of this dumped on one pliant customer commu- nity, cannot work in the future. Not neces- sarily because librarians will become more assertive ... but because there is no growth in the resources of what continues to be the one and only customer commu- nity. " 37 In other words, voluntary collec- tive assertiveness among research librari- ans is unlikely, even though we may all agree that collective assertiveness would help. Instead, White argues, only the ex- haustion of our collective budget re- sources will bring an end to the spiraling growth in scholarly journal numbers and prices. If this happens, the most important tragic consequence for our current system of scholarly communication will be the slow death of research libraries as much more than archives of the past. Their abil- ity to collect and provide access to the cur- rent scholarly record will be gradually de- stroyed. The publishers and scholars who depend on the research library's budget and scholarly information resources will be forced to find alternative, and probably 192 College & Research Libraries more expensive, ways to disseminate and gain access to these resources. In the face of such discouraging predictions and the inexorable and seemingly inevitable forces leading to a research library com- mons tragedy, what can be done to insure that research libraries (or some reasonable alternative) will continue to carry out the mission of collecting, preserving, and pro- viding access to the scholarly record? The concluding section of this essay will ex- plore a few tentative possibilities for pre- venting this tragedy and managing the re- search library commons for the long-term future. MANAGING THE RESEARCH LIBRARY COMMONS Over the past ten years or so, probably the most frequently suggested strategy for research libraries to deal with the spiraling quantities and costs of scholarly informa- tion has been to use new electronic and optical information storage and retrieval systems tied together in regional, na- tional, and international networks. The combination of these new information storage and sharing technologies, pro- ponets argue, should make possible effec- tive storage and retrieval of much larger quantities of information at much lower unit costs. In addition, networks would permit direct access to remote resources, reducing the necessity for every research library to collect comprehensively. A very influential statement of this vi- sion of a technological future for research libraries was the 1982 Matheson-Cooper Report sponsored by the Association of American Medical Colleges and the Na- tional Library of Medicine. The report called for strategic planning initiatives and major investments in information technol- ogy to manage information resources within integrated networks. 38 However, an important characteristic of the class of human problems represented by the trag- edy of the commons, says Hardin, is that they are not particularly amenable to tech- nical solutions. ''A technical solution,'' he says, ''may be defined as one that requires a change only in the techniques of the nat- ural sciences, demanding little or nothing in the way of change in human values or May 1990 ideas of morality."39 To the degree that technology simply changes the dimen- sions or capacity of the research library commons to deal with larger volumes of information, or transforms that informa- tion into formats with lower unit costs, we may be only delaying the inevitable trag- edy (or perhaps compounding it because of the high cost of the new technologies!). This is because the basic human patterns of overproduction and overconsumption of scholarly information described above will remain unchanged. Matheson and others have argued, however, that this report's recommenda- tions are not, in their essence, a set of pro- posed technological solutions, but rather an ''attempt to define the basis for a differ- ent library paradigm for the future.' '40 She quotes Thomas Kuhn41 who remarked that the process of knowledge transformation may really mean the "reconstruction of group commitments among the commu- nity of scientists.'' Such a reconstruction, Matheson says, "with respect to the man- agement of information in the health sci- ences would be an exceedingly useful thing.'' 42 Virginia Holtz takes this Kuhn- ian analogy of the paradigm shift further by suggesting that the Matheson-Cooper Report provides a focus for a new world view where the "information user, rather than the library, and information per se, rather than the instruments which carry it, have become the central concerns of our discipline. " 43 In essence, what Matheson and others of this school are proposing is "the deinstitutionalization of [research] li- braries" so they will shift away from building definitive collections of books, journals, and bibliographic data and, in- stead, ''conceive of different ways to en- hance the utility of our major [scholarly in- formation] assets, to improve the productivity of the academic commu- nity.''44 Short of the kind of major rethinking of the mission of research libraries Matheson proposes, other useful suggestions have been made to manage research libraries in ways to avoid their over-exploitation by information entrepreneurs. Stephen Fein- man argues that some sort of government . regulation of the scientific and technical information industry may be needed, and that this may best be justified if we cease to focus on the public goods aspects of infor- mation and think instead of "knowledge- information-communication packages'' as a form of capital. Feinman envisions "an information industry structured much like the banking industry'': Information banks would be chartered at both the national and state levels . . . The producers of knowledge, which would be captured by these banks, might be treated in the same man- ner as individuals with time deposits .... The users of the system . . . might well be treated as people or in~titutions that obtain a loan [to be] paid back in terms of principal and interest .... Access to the banks would be broadly regulated by the central bank but controlled by the indi- vidual institutions .... The central bank, con- trolled by the government, would also define the limits of research, since knowledge that is not captured cannot be communicated. 45 As Feinman notes, this scenario would have dramatic implications for all parts of the scholarly communications process, in- cluding elimination of the need for or pur- pose of copyright. Since copyright is one of the keystones supporting the commer- cial exploitation of scholarly information resources and products, most proposals to manage these resources better involve changes in copyright laws. Professor Peet, at the SSP seminar, proposed changes in these laws that would reduce ''copyright on journal articles . . . to a short period of time, perhaps one or two years." This would preserve incentives for publishers to provide the immediate access ''critical to the active researcher, [but] allow li- braries to acquire backruns of . . . journals at little or no cost. " 46 A growing number of research librari- ans now argue that ways should be found to reduce the role of commercial pub- lishers in the scholarly communication process dramatically. For instance, Pat Battin argues that the library should be- come the center of a new kind of restruc- tured university that controls and man- ages the whole process of creating and disseminating ideas to advance knowl- edge for the public good. 47 Similarly, Rich- ard Dougherty and Brenda Johnson sug- gest that "if scholars and librarians can't Economic "Commons" Tragedy 193 communicate easily through publishers, we must learn to communicate around them. " 48 And James Thompson thinks that ''the idea of the academy retaking control of the bulk of scholarly publishing is being forced into consideration b~ the practices of commercial publishers.'' 9 But even if these librarians are all correct, and ''there is no technical or economic rea- son'' why commercial publishers must re- main a part of the "information conduit" for scholarly communication, 50 formidable barriers remain to changing the centuries- old incentives and economic structures of our current scholarly information market- place. Individual publishers and scholars, like the individual herdsmen in Garrett Hardin's story, are convinced that their journals and articles are all outstanding and should easily get a share of the limited budget resources of research libraries. "Each of them thinks," as Herb White puts it, ''that while indeed other and more marginal publishers [and scholars] might suffer, they would be unaffected. " 51 Each of the solutions proposed above, as with other ''commons'' problems faced by our so~iety, would involve a radical re- structuring of the ways scholars, pub- lishers, and librarians think about and use research library resources. Appeals to in- dividual publishers, scholars, or even li- brarians to restrain themselves for the common good will not work. As Hardin puts it, appeals to conscience ''set up a se- lective system that works towards the elimination of conscience from the race ... [since] to conjure up a conscience in others is tempting to anyone who wishes to extend his control beyond the legal lim- its.'' What we need instead, says Hardin, is "mutual coercion mutually agreed upon, " 52 that is, we must eventually give up altogether the idea of a commons open to unrestrained exploitation. In place of individual freedom to exploit, we must learn what Kenneth Boulding (1977) calls simply a sense of "community." But this is a ''long and painful learning process.'' 53 If we are going to bring about the changes needed to avoid the gradual de- struction of our research library com- mons, it will take leadership and a com- mon understanding of the grave 194 College & Research Libraries implications of continuing on our present course. It seems clear that current market incentives are much too strong to expect the initial leadership or understanding to come from publishers or research scholars. Because research libraries have been charged by society to manage our na- tion's scholarly information resources, re- search librarians must assume this leader- ship role and develop strategies to avoid the approaching tragedy. The first step, I would argue, is to educate ourselves and our publisher and academic scholar col- leagues about the true economic implica- tions of the current scholarly information marketplace. It is encouraging to note that profes- May1990 sionallibrary associations such as the As- sociation of Research Libraries have re- cently taken major steps to begin this process of education. Witness the two 1989 studies published as part of ARL' s Project on Serial Prices. 54 These studies present their recommendations backed with convincing empirical evidence and solid economic analysis. They also em- phasize the importance of an "ongoing program of education and publicity.'' 55 Only when the impending commons trag- edy of our nation's research libraries is widely understood will we be able to build the necessary political consensus for change. REFERENCES 1. Peter R. Young and Kathryn Hammell Carpenter, "Price Index for 1989: U.S. Periodicals," Library Journal114(7):43-49 (Apr . 15, 1989), p.47. 2. Herbert S. White, "The Journal That Ate the Library," Library Journalll3(9):62-63 (May 15, 1988), p.63. 3. Ibid., p.62. 4. Ibid. 5. Garrett Hardin, "The Tragedy ofthe Commons," Science 162:1243-48 (Dec. 13, 1968) reprinted in Managing the Commons, ed. by Garrett Hardin and John Baden (San Francisco, Calif.: W.H. Free- man and Company, 1977), p.16-30. 6. Harlan Cleveland, The Knowledge Executive (New York: E.P. Dutton, 1985), p.29-35. 7. Ibid., p.73. 8. Pat Molholt, "On Converging Paths: The Computer Center and the Library," Journal of Academic Librarianship 11:284-88 (Nov. 1985), p.284. 9. Robert S. Taylor, Value Added Processes in Information Systems (Norwood, N.J.: Ablex, 1986), p.20-21. 10. Fritz Machlup, Knowledge: Its Creation, Distribution, and Economic Significance, Vol. III, The Economics of Information and Human Capital (Princeton, N.J.: Princeton Univ. Pr., 1984), p.159-60. 11. Lawrence White, "The Sensible Economist's Guide to the Economics of Information," in The Eco- nomics of Information, ed. by Jana Varlejs Gefferson, N.C.: McFarland & Co., 1982), p.20. 12. Fritz Machlup, Knowledge: Its Creation, Distribution, and Economic Significance, Vol. I, Knowledge and Knowledge Production (Princeton, N.J.: Princeton Univ. Pr., 1980), p .208. 13. Sanford V. Berg, "Structure, Behavior, and Performance in the Scientific Journal Market" (Yale University, unpublished doctoral dissertation, 1970), p.6. 14. Machlup, Knowledge and Knowledge Production, p.207-208. 15. Aatto J. Repo, "The Value of Information: Approaches in Economics, Accounting, and Manage- ment Science," Journal of the American Society for Information Science40:68-85 (Mar. 1989), p.81-82. 16. Ibid., p.81. 17. Berg, "Structure, Behavior, and Performance." See also Sanford V. Berg, "An Economic Analysis of the Demand for Scientific Journals," Journal of the American Socif ty for Information Science 23:23-29 Ganuary-February 1972). 18. Machlup, Knowledge and Knowledge Production, p.220-22. 19. Patrick Joyce and Thomas E. Merz, "Price Discrimination in Academic Journals," Library Quarterly 55(3):273-83 (1985), p.274. 20. David N. Hyman, Modern Microeconomics: Analysis and Applications (St. Louis: Times Mirror/ Mosby, 1986), p.33-35. 21. Margaret M. Quinlin, ''Pricing: Its Place in the Marketing Mix & Its Contribution to Profit,'' Society Economic ~~commons" Tragedy 195 for Scholarly Publishing Special Report Series 3:1-12 (1988), p.2-3. 22. Fritz Machlup, "Publishing Scholarly Books and Journals: Is It Economically Viable?" Journal of Political Economy 85:217-25 (1977), p.221. 23. Brian Cox, "Scholarly Journal Prices," The Serials Librarian 13:135-38 (Oct./Nov. 1987). 24. Charles Hamaker, "Serials Costs and the Carrying Ability of Serials Budgets 1987," The Serials Librarian 13:129-34 (Oct./Nov. 1987). 25. Richard M. Dougherty and Brenda L. Johnson, "Periodical Price Escalation: A Library Re- sponse," Library Journal113(9):27-29 (May 15, 1988). 26. Ibid., p.28. 27. Joyce and Merz, "Price Discrimination," p.224. 28. James C. Thompson, "Journal Costs: Perception and Reality in the Dialogue," College & Research Libraries 49:481-82 (Nov. 1988), p.481. 29. Robert K. Peet, "Response to Ms. Karen Hunter, 'The Future of the Journal: A Publisher's Per- spective,'" Unpublished manuscript of remarks presented at the Seminar on the Future of the Scholarly Journal (Chapel Hill, N.C.: Society for Scholarly Publishing, 1988), p.l-4. 30. Herbert S. White, ''Scholarly Publishers and Libraries: A Strained Marriage,'' Scholarly Publishing 125-29 (April, 1988), p.127. 31. Hardin, "The Tragedy of the Commons," p.20. 32. Ibid., p.21-22. 33. Katina Strauch, "Response: The Image Maker Meets the Competition," The Serials Librarian 13:85-90 (Oct./Nov. 1988), p.87. 34. Hardin, "The Tragedy of the Commons," p.19. 35. Machlup, "Publishing Scholarly Books and Journals," p.221-22. 36. Richard B. McKenzie, "The Economist's Paradigm," Library Trends 28:7-24 (Summer 1979), p.15. 37. White, "Scholarly Publishers and Libraries," p.129. 38. Nina W. Matheson and John A.D. Cooper, "Academic Information in the Academic Health Sci- ences Center: Roles for the Library in Information Management," Journal of Medical Education 57(10, pt 2):1-93 (Oct. 1982). 39. Hardin, "The Tragedy of the Commons," p.16-17. 40. Nina W. Matheson, "The Author Replies: Neither True Nor False, but More or Less Useful," Bul- letin of the Medical Library Association 72:32-34 Gan. 1984), p.33. 41. Thomas S. Kuhn, The Structure of Scientific Revolutions, 2nd ed., enlarged (Chicago: Univ. of Chi- cago Pr., 1970). 42. Matheson, "The Author Replies," p.34. 43. Virginia H. Holtz, "Measures of Excellence: The Search for the Gold Standard," Bulletin of the Medical Library Association 74:305-14 (Oct. 1986), p.312. 44. Nina W. Matheson, "The Academic Library Nexus," College & Research Libraries 45:207-13 (May 1984), p.207, 212. 45. Stephen Feinman, "Regulation of the STI Industry: A Historical Basis and Some Possibilities," in Information Seroices: Economics, Management, and Technology, ed. by Robert M. Mason and John E. Creps, Jr. (Boulder, Colo.: Westview Press, 1981), p.60. 46. Peet, "Response to Ms. Karen Hunter," p.7-8. 47. Patricia Battin, ''The Library: Center of the Restructured University,'' College & Research Libraries 45:170-76 (May 1984). 48. Dougherty and Johnson, "Periodical Price Escalation," p.28. 49. Thompson, "Journal Costs," p.482. 50. Ibid. 51. White, "The Journal That Ate the Library," p.63. 52. Hardin, "The Tragedy of the Commons," p.26. 53. Kenneth E. Boulding, "Commons and Community: The Idea of a Public," in Managing the Com- mons, ed. by Garrett Hardin and John Baden (San Francisco, Calif.: W .H. Freeman and Company, 1977), p.287. 54. Economic Consulting Services, Inc. "A Study of Trends in Average Prices and Costs of Certain Serials Over Time" (Washington, D.C.: 1989) and Ann Okerson, "Of Making Books There is No End: Report of Serials Prices for the Association of Research Libraries" (Eastchester, N.Y.: 1989). 55. Association of Research Libraries, "Association of Research Libraries' Project on Serial Prices: Overview and Summary" (Washington, D.C.: 1989).