lawrence.p65 Life Cycle Costs of Library Collections 541 Life Cycle Costs of Library Collections: Creation of Effective Performance and Cost Metrics for Library Resources Stephen R. Lawrence, Lynn Silipigni Connaway, and Keith H. Brigham An important issue for research librarians is the life cycle cost of acquir­ ing and maintaining a collection. While purchase costs are easy to iden­ tify, associated acquisition, cataloging, circulation, and maintenance ex­ penses are difficult to measure and attribute to specific collections. This paper develops a methodology to determine the life cycle costs of col­ lections based on readily available statistical data collected annually by the Association of Research Libraries (ARL). ARL cost data (e.g., sala­ ries and wages, materials expenditures, and operating expenses) for a specific library are allocated to collections (e.g., manuscripts, serials, and microforms) based on the size of the collection and its relative space requirements. By aggregating allocated costs, total life cycle costs for a collection can be estimated. Results of this research indicate that life cycle costs of collections are many multiples of their purchase costs. Results further suggest that the life cycle costs of monograph collec­ tions overwhelm the costs of other collections in research libraries—the cost structure of a research library is largely driven by its monograph collection. These results should prove useful in efforts to control costs and improve performance in research libraries. his project investigates the life cycle costs associated with the acquisition and maintenance of various collections in research libraries. The objective is to develop effec­ tive performance and cost metrics for col­ lections in research libraries. Historically, research libraries have had no alternative but to acquire books and periodicals in a bound paper format. However, with the advent of digital technology, research li­ braries now can acquire material for addi- Stephen R. Lawrence is Deming Professor of Entrepreneurship in the Graduate School of Business Ad­ ministration at the University of Colorado at Boulder; e-mail: Stephen.Lawrence@Colorado.edu. Lynn Silipigni Connaway is Vice President, Research and Library Systems, at netLibrary, Inc.; e-mail: lynnc@netlibrary.com. Keith H. Brigham is Assistant Professor of Strategy at Texas Tech University; e- mail: KBrigham@bant2.ba.ttu.edu. This research was commissioned and sponsored by netLibrary under University of Colorado Contract #OCG4165B. The authors gratefully acknowledge the support and assis­ tance of netLibrary and particularly thank Tim Schiewe and Woody Palasek for their helpful information and encouragement. Faculty from the University of Colorado library also aided our research. CU Libraries Dean Jim Williams; Associate Deans Scott Seaman, Susan Anthes, and Janet Swan Hill; and Business Library Director Carol Krissman all provided helpful information, perspectives, and time. 541 mailto:KBrigham@bant2.ba.ttu.edu mailto:lynnc@netlibrary.com mailto:Stephen.Lawrence@Colorado.edu 542 College & Research Libraries November 2001 tion to collections in a variety of formats, including traditional bound paper, CD­ ROM disk, or online delivery via Internet or database services. Currently, few data or little developed methodology exists for comparing the life cycle ownership costs of traditional or alternate media to guide libraries in making purchasing decisions. Thus, the purpose of this project is to re­ view extant literature, collect data, and de­ velop a methodology to allow libraries to effectively measure and compare the costs of acquiring content in alternative forms. This article seeks to: • understand the total life cycle cost of ownership for library collections; • develop metrics for measuring and comparing the life cycle costs of various library media; • develop effective methods for quickly calculating performance and cost metrics. To accomplish these objectives, the au­ thors surveyed the library costing litera­ ture, conducted interviews with a number of knowledgeable library experts, acquired relevant data from the Association of Re­ search Libraries (ARL) and industry sources, and developed an analytic tool to assist in the calculation of life cycle costs. This research has provided an easily accessible and easy-to-use tool, the Li­ brary Interactive Costing Spreadsheet (LICS), which provides a method for re­ search librarians to quickly evaluate the cost structure of their libraries compared with other ARL libraries. Application of the tool indicates that the life cycle cost of maintaining library collections is many multiples of their acquisition cost. Also, the research indicates that monograph collections are the overwhelming life cycle cost components for research librar­ ies—costs in research libraries are largely driven by their monograph collections. Literature Review The research questions for this study are: • What is the total life cycle cost of a library book or monograph? • What are the relative costs of vari­ ous media? • How do these relative costs com­ pare across media? These questions may seem straightfor­ ward, but the unique nature of libraries makes it difficult to uncover the answers to them. Cost management principles and techniques available for economic library management are among the weakest in the repertoire of library management.1 Possible reasons for the relatively weak development of library costing systems include:2 • non-corporateness (no legal provi­ sions or requirements); • lack of motivation to develop cost­ ing systems (weak requirement for pub­ lic accountability); • lack of historical tradition (related to limited control over limited resources); • lack of a definable product and per­ formance measures; • inherent complexity of the system. Final delivery of service to the client is paramount; librarian and library sponsor are content if some reasonable level of service is accomplished within budgetary targets. Stephen A. Roberts defined five types of cost studies in the library management literature:3 1. cost analysis; 2. cost distribution and/or cost allo­ cation; 3. unit costing and timing; 4. cost-effectiveness; 5. cost benefit. The first two—cost analysis and cost distribution—are methods for determin­ ing costs in libraries. The remaining three—unit costing, cost-effectiveness, and cost benefit—use the results of cost analysis and cost distribution to evaluate performance. Because the purpose of this study is to determine the life cycle costs of library media, the following discussion focuses principally on cost analysis and cost distribution. Cost Analysis Studies Work measurement lies at the heart of cost analysis studies. It strives to measure the precise quantity of factors used (usually Life Cycle Costs of Library Collections 543 labor) and the time devoted to particular functions and tasks.4 This represents a bottom-up approach to determining costs by starting at the smallest identifiable and measurable activity and then relating it to a specific function or task. For example, Joan A. Maier used timed observations to calculate the walking rate (in feet per minute) of the average library employee.5 This figure then was used to help deter­ mine the cost to catalog a volume. Work measurement also was used to determine the costs of different functions, including circulation and tracking overdue items. 6, 7 Estimation methods are often used when exact work measurement of labor is difficult or impractical. The use of fac­ tor inputs is allocated based on estimates derived from either observations of or dis­ cussions with knowledgeable staff mem­ bers. Estimation methods, such as Roberts’s, often provide relatively accu­ rate estimates and may complement work measurement–based studies.8, 9 Although cost analysis studies provide detailed cost information for particular functions in specific libraries within a narrow time pe­ riod, the generality of their results is lim­ ited. Roberts concluded that there was a “yawning gap in the library management literature” with respect to these early cost analysis studies.10 Cost Distribution and Allocation Studies Whereas cost analysis is a bottom-up ap­ proach to determine actual costs in a par­ ticular library function, cost distribution is a top-down method that provides simi­ lar information with less effort. Under cost distribution, actual total expenditures are allocated to various cost centers. Cost distribution often depends on budgetary data and is useful in supplementing nor­ mal budget control. An early study using a cost distribution method was conducted by Ferdinand F. Leimkuhler and Michael D. Cooper.11 In this study, actual budget data from the General Library at the Uni­ versity of California-Berkeley were used to determine the cost of performing par­ ticular services. Through the allocation of budget data to different services, per unit costs for circulation, holding, and acqui­ sition were developed. More recently, Paul M. Gherman and Lynn S. Cochrane used the cost distribution method in their study of The Virginia Tech Libraries.12 By allocating historical costs to cost centers, they determined unit costs for collection development, cataloging, retrieval from storage, and remote housing in storage. Although cost distribution studies may lack the precision of work measurement studies, they do offer valuable diagnos­ tic and prognostic information.13 Other Library Costing Studies The remaining three types of library cost studies are more concerned with the re­ lationship between costs and perfor­ mance. They all rely on cost data gath­ ered through either the cost analysis or the cost distribution method and attempt to establish performance implications. As performance aspects are beyond the pur­ poses of this study, only a brief summary of each is provided. Unit Costing and Timing Unit costing requires the use of work stan­ dards, or norms. These are derived from data provided by work measurement or cost distribution studies. Expected out­ puts based on these standards then can be compared with actual outputs. The variance between expected outputs and actual outputs can provide a simple mea­ sure of performance for a specific branch or service. Eileen G. Abels, Paul B. Kantor, and Tefko Saracevic applied functional cost analysis to reference services across libraries.14 They found wide degrees of variation in unit costs, implying that there were still substantial challenges to under­ standing the cost basis of library services. Cost-effectiveness In a cost-effectiveness analysis, the focus is on the level of output achieved or on satisfying certain goals. Cost-effectiveness and performance measurement are cen­ tral topics in the library management de­ bate.15 The cost-effective library, one in which the highest level of performance http:libraries.14 http:information.13 http:Libraries.12 http:Cooper.11 http:studies.10 544 College & Research Libraries is achieved given limited levels of expen­ diture and resources, is the ultimate goal for library managers.16 Cost-benefit Measures Cost-benefit analysis is a method for de­ termining the social costs and benefits of an investment to decide whether the project should be undertaken. Within cost- benefit analysis, the concept of value is compared with costs. Karen Svenningsen developed a model that incorporates the economic value of information and cost and benefit concepts.17 In many cases, in­ terest in costing specific library functions was driven by increased availability of li­ brary information systems and automation and the consequent need to undertake cost-benefit analyses to justify the acquisi­ tion of new technologies.18 Perhaps because the promise of library cost studies has yet to be realized, the call for additional studies of library functions and services continues.19 This may be at­ tributed to the escalating costs of library resources and stagnant or declining library budgets. In addition, rapidly evolving in­ formation technologies and media options make it essential for librarians to under­ stand their costs in order to make informed decisions about acquiring new technolo­ gies. As Martin M. Cummings concluded, “Analysis of costs in meaningful detail is essential to management, i.e., answering the question, ‘Are there alternate, less costly ways to do the same thing?’”20 Methodology This article presents a new and unique method for determining library costs. Al­ though this method could be categorized as a cost distribution study (following the top-down allocation method), it is novel in its approach. Whereas previous cost dis­ tribution studies have allocated expenses and resources to specific cost centers, this study allocates expenses and resources to different types of library media (mono­ graphs, serials, audio, etc.).21 Through the creation of a costing spreadsheet that uses readily available ARL data and industry parameters, this study provides librarians November 2001 with a new method for ascertaining the costs of library collections. Sources of Data This study uses the easily accessible and available 1998 and 1999 data collected by the ARL and reported on its Web site (www.arl.org). Every year, the ARL sur­ veys its members (121 in 1999) to solicit statistics on a number of dimensions im­ portant to research libraries. A summary of the data types collected by the ARL is included in appendix A. The authors used ARL data from its 1998 and 1999 surveys and, consistent with these surveys, grouped library collections into ten catego­ ries: • Monographs (books) • Current Serials • Microfilms • Government Documents • Manuscripts and Archives • Cartographic Materials • Graphic Materials • Sound Recordings • Video and Film • Computer Files In addition to ARL data, published data, statistics, and formulae relating to library design and construction were used.22 Cost Categories Library expenses were estimated in six broad categories: • Purchase Cost of Holdings • Operating Expenses • Wages and Salaries • Building and Facilities • Building Maintenance • Fixtures and Equipment Member libraries in the annual ARL surveys report expenses for the first three categories. In contrast, expenses for the second three categories must be estimated and derived from both external industry data and ARL survey data. The method­ ology used to allocate expenses to these categories is explained below. Allocation Methodology In any costing study, the allocation of ex­ penses to the service or product of interest http:www.arl.org http:etc.).21 http:continues.19 http:technologies.18 http:concepts.17 http:managers.16 Life Cycle Costs of Library Collections 545 is of fundamental importance. The princi­ pal allocation method used was the physi­ cal area occupied by various types of hold­ ings. The area storage requirements of the various media were used to calculate a “book equivalent” factor for each media type that represents the fraction of space required by the media type relative to books and manuscripts. For example, the media type “government documents” has a book equivalent (BE) of 20 percent be­ cause “govdocs” takes only 20 percent of storage space per unit compared to books. The methodology used to allocate ARL- reported expenses for each of these catego­ ries is explained below. A fundamental assumption of this study is that media BE figures are the pri­ mary drivers of costs in research libraries. BE estimates are used to allocate operat­ ing expenses, facility costs, maintenance expenses, asset inventories, and salary and wage expenses. The use of BE to allocate costs is based on a review of the library costing literature and on extensive discus­ sions with professional librarians. The re­ search confirms that the space consumed by a collection is an excellent first-order proxy for the costs associated with main­ taining and circulating the collection. For example, a collection that occupies twice the space of another collection will gener­ ally require twice the labor and twice the assets (shelving, tables, etc.) and incur twice the expenses. Estimated space re­ quirements for various cat­ egories of library media are Purchase Expenses The ARL collects statistics on total annual expenditures for monographs (EXPMONO) and serials (EXPSER) but aggregates pur­ chases of other media types (EXPOTH). Esti­ mates for expenditures of other media types (e.g., manuscripts, videos and film, etc.) were derived as follows: The growth of individual collections in a specific library were calculated by subtracting the 1999 size of the collection from its reported size in the 1998 ARL survey. In cases where the size of a collection declined, growth was set to zero. The relative size of the collec­ tion relative to other collections (exclud­ ing monographs and serials) then was used to allocate expenditures on other media. For example, if graphic materials were found to grow by 2,000 documents from 1998 to 1999, and this represented five per­ cent of all nonbook and nonserials docu­ ments acquired by a library for that year, five percent of “other expenditures” (EXPOTH) would be allocated to graphic materials. Appendix B shows the median allocation percentages for other expendi­ tures of the 121 libraries that participated in the 1999 ARL survey. Operating Expenses Aggregate library operating expenses were taken as the sum of ARL expense categories: • Operating Expenses (OPEXP) • Miscellaneous Expenses (EXPMISC) TABLE 1 shown in table 1. BE Figures for Library Media Types Holdings Holdings data by collection type were taken directly from ARL survey statistics. The holdings statistics then were adjusted to BE hold­ ings by applying the BE multiplier described above. Appendix B shows the me­ dian holding and BE-ad­ justed holdings by media category for the 121 librar­ ies in the 1999 ARL survey. Area Occupied Book Medium (sq. ft. 1 1,000) Equivalent Monographs (books) Current serials Microforms Government documents Manuscripts and archives Cartographic materials Graphic materials Sound recordings Video and film Computer files 100 100.0% 67 67.0% 0.17 0.17% 20 20.0% 20 20.0% 20 20.0% 2 2.0% 20 20.0% 143 143.0% 0.20 0.2% 546 College & Research Libraries The aggregate operating expenses for a library then were allocated to collections using the BE-adjusted holding percent­ ages for the library. Appendix B tabulates the median operating expenses by collec­ tion for ARL libraries in 1999. Wages and Salaries Total salaries and wages (TOTSAL) reported in ARL statistics again were allocated to library collections using BE-adjusted holding percentages. Median library wages and salaries by collection are listed in appendix B. Building and Facilities The ARL survey does not collect data on the actual size of facilities occupied by its member libraries nor does it solicit the replacement cost of these facilities. The facility size of individual libraries was estimated from Metcalf, in which neces­ sary space requirements for libraries of various sizes were listed.23 Table 2 is almost perfectly estimated (R2 = 99.9%) using the regression equation: Area = 5,750 + 0.148 * VOLS where VOLS is the total number of volumes in a library as reported in the ARL survey. This equation was used to estimate library facility sizes measured in square feet. Facility costs then were cal­ culated using national estimates for the re­ placement cost (per square foot) of public buildings. Because buildings have an ex­ pected life far longer than one year, amor­ tized annual replacement costs were calcu­ lated assuming a facility life expectancy of fifty years. This annual amortized replace­ ment cost then was allocated to a collection using its relative BE-adjusted size. November 2001 Facility Maintenance The ARL survey does not collect informa­ tion on the ongoing maintenance costs necessary to maintain library facilities. Ongoing facility maintenance expenses were estimated using the library space requirements derived above and multi­ plied by a national estimate for the an­ nual maintenance cost (per square foot) of public buildings. As usual, mainte­ nance costs were allocated to collections using BE statistics. Fixtures and Equipment The fixtures and equipment expense cat­ egory includes shelving, tables and chairs, computer terminals, equipment, and any other physical items used in a library. The ARL does not collect statis­ tics on fixtures and equipment, so equip­ ment requirements had to be estimated by library. This was accomplished by multiplying estimated library facility size (measured in square feet) by a national estimate for fixture and equipment assets (cost per square foot) in public buildings. Because the life of these assets is typically greater than one year, their cost must be amortized over their expected lives. The expected life of various types of furnish­ ings and equipment varies greatly by type and use, so an aggregate uniform ex­ pected life estimate is possible and nec­ essary. For the purposes of this study, the authors estimated the average expected life for library fixtures and equipment to be ten years. Amortized annual replace­ ment costs of library fixture and equip­ ment assets were allocated to collections by their relative BE size. TABLE 2 Space Requirements of Monographs Volumes Sq. ft. 100,000 20,000 250,000 45,000 500,000 80,000 1,000,000 150,000 2,700,000 405,000 Life Cycle Cost Calculations The unit life cycle costs of maintaining library collections were found by deter­ mining the annual cost of maintaining an entire collection and then dividing it by the number of bibliographic units in the collection, providing the annual cost of maintaining a single unit in the collection. To obtain a life cycle cost estimate, two calculation methods were used—total costs and discounted costs. http:listed.23 Life Cycle Costs of Library Collections 547 TABLE 3 Expected Life of Library Media Medium Expected Physical Life (years) Manuscripts (books) 50 Current serials 50 Microforms 25 Government documents 50 Manuscripts and archives 100 Cartographic materials 50 Graphic materials 20 Sound recordings 15 Video and film 10 Computer files 5 AdaptedffromfMetcalff1986 The discounted life cycle cost of li­ brary media was calculated by dividing annual ongoing expenses by the discount rate and then adding the average pur­ chase cost of a unit in the collection. It should be noted that this calculation does not require an estimate of the life expect­ ancy of library media. Instead, the au­ thors assumed that the annual expense of maintaining a unit in a collection is incurred in perpetuity. That is, if it cur­ rently costs $10 per year to maintain a book in a particular library, this cost will be incurred for all time into the future. Although it may appear unreasonable to assume that a book will last forever and that its maintenance costs will be as­ sumed forever, the discounted cost of maintaining the book at the end of its ex­ pected life fifty years from now is only: $10.00 x 0.07550 = $0.20 which is a rela­ tively small amount. Thus, discounted cost estimates provide an alternative per­ spective on the life cycle cost of owning library media that takes into account the time value of money. Results Data from the 1998 and 1999 ARL surveys were used to calculate the life cycle costs of library collections using the allocation methodology described above. Details of these calculations are tabulated in appen- Total Costs The total life cycle cost of owning a unit in a collection was calculated by simply mul­ tiplying the annual cost of maintaining a unit in a collection by its expected physi­ cal life and adding the unit’s purchase cost. This calculation requires that the expected physical life of various library media be estimated. In research libraries, “physical life” is defined as the time period between the acquisition of a new document and its eventual reconditioning, restoration, or re­ placement. It should be noted that physi­ cal life is different from useful life. A document’s useful life is the time until its content is obsolete; its physical life is the time until the document no longer func­ tions materially. Because research librar­ ies typically retain collections in perpetu­ ity, the focus here is on the physical life of library media. The authors used the esti­ mates of expected physical life displayed in table 3. The total cost of a bibliographic unit represents an aggregation of all direct and indirect costs over the expected life span of the unit. The research indicates that the purchase price of acquiring a collection is a small fraction of the life cycle cost of maintaining it. Discounted Costs Discounted cost calculations take into ac­ count the time value of money (i.e., a dol­ lar expended today is dearer than a dol­ lar expended twenty years from now). Discounted cost calculations require the estimation of a “discount rate” to adjust the value of future expenditures. A dis­ count rate of 7.5 percent, the long-term average interest rate delivered by munici­ pal and state bonds, was used for the cal­ culations. The assumption is that most research libraries are supported by gov­ ernment and university agencies that can obtain funds at this rate. In the case of a library supported by an endowment, the discount rate represents the “opportunity cost” of money—in other words, the re­ turn the endowment could earn alterna­ tively if invested conservatively. 548 College & Research Libraries dix B, and the median estimated life cycle costs for various media in ARL libraries are summarized in table 4. Discounted life cycle costs are gener­ ally less than total costs because future dollars for the former are heavily dis­ counted. However, this relationship is not true for video, film, and computer files because of their relatively short life ex­ pectancy. Table 4 also includes compari­ son life cycle cost data using 1998 ARL data. The close congruence between the life cycle costs between the two years pro­ vides evidence that life cycle cost esti­ mates are consistent across time. November 2001 Life Cycle Costs of Library Collections The data in table 3 clearly indicate that, on average, the purchase price of library media is a small fraction of the life cycle ownership costs of li­ brary collections. For example, the re­ sults of this project demonstrate that the expected cost of owning a mono­ graph is more than seven times the monograph’s purchase price. This has important financial implications for li­ braries that are building their collec­ tions: the initial cost of populating a collection is a small fraction of the on­ going cost of maintaining one. TABLE 4 Median Life Cycle Cost Summaries Using 1998 and 1999 ARL Data 1999 ARL data Medium Discounted Total Purchase Total Cost I Life Cycle Cost Life Cycle Cost Cost Purchase Cost (per unit) (per unit) (per unit) (per unit) Monographs (books) $ 119.56 $ 343.03 $ 47.78 718 % Current serials 643.91 801.78 590.97 134 Microforms 0.27 0.45 0.11 256 Government documents 14.13 55.40 0.00 311 Manuscripts and archives 20.26 126.79 4.46 1,130 Cartographic materials 26.78 73.82 11.05 247 Graphic materials 1.65 2.91 0.06 216 Sound recordings 22.64 24.77 6.80 219 Video and film 128.95 107.50 15.70 307 Computer files 0.17 0.07 0.01 331 1998 ARL data Medium Discounted Total Purchase Total Cost I Life Cycle Cost Life Cycle Cost Cost Purchase Cost (per unit) (per unit) (per unit) (per unit) Monographs (books) $ 121.28 $ 340.17 $ 47.90 742 % Current serials 594.50 731.20 541.87 139 Microforms 0.26 0.42 0.10 284 Government documents 14.25 56.69 0.00 300 Manuscripts and archives 18.59 126.09 3.11 1,251 Cartographic materials 28.44 74.58 13.00 227 Graphic materials 1.64 2.89 0.09 185 Sound recordings 20.85 22.58 5.36 203 Video and film 122.60 105.02 10.85 290 Computer files 0.28 0.13 0.01 384 Life Cycle Costs of Library Collections 549 TABLE 5 Relative Cost of Collections Book Reported BE Relative BE Medium Equivalent Holdings Holdings Holdings Monographs (books) 100.0% 2,716,117 2,716,117 95.1 % Current serials 67.0% 22,172 14,855 0.5 Microforms 0.17% 3,700,071 7,400 0.3 Government documents 20.0% 282,321 56,464 2.0 Manuscripts and archives 20.0% 12,941 2,588 0.1 Cartographic materials 20.0% 179,190 35,838 1.3 Graphic materials 2.0% 149,171 2,983 0.1 Sound recordings 20.0% 31,251 6,250 0.2 Video and film 143.0% 8,464 12,094 0.4 Computer files 0.2% 3,919 8 0.0 Relative Life Cycle Costs Another important question answered by this study is, What are the relative costs of maintaining different library collections? Table 5 demonstrates that monographs are overwhelmingly the largest source or driver of library costs, consuming 95 per­ cent of library life cycle expenditures. This result indicates that if research libraries want to control their costs, they must work to control and reduce the life cycle costs of maintaining their monograph collections. Summary and Conclusions This study used easily accessible and widely distributed survey data from the ARL to estimate the life cycle costs of li­ brary collections. The research indicates that the purchase price of acquiring a col­ lection is a small fraction of the life cycle cost of maintaining it. Although this re­ sult may be intuitively obvious to those familiar with library operations, it is be­ lieved that this is the first study to rigor­ ously measure the life cycle cost of library collections and to quantify the relation­ ship between a collection’s purchase cost and its subsequent maintenance costs. A second contribution of this research was to measure the relative cost impact of various library collections among ARL research libraries. The results demon­ strate that monographs dominate the cost structure of a research library, consum­ ing approximately 95 percent of library resources. For research libraries that want to control expenses over time, the control of monographs is essential. Although the initial acquisition expense for some col­ lections may rival that of monographs, the life cycle cost of maintaining monographs dwarfs that of other collections. Further Research Because electronic books (eBooks) are be­ coming more prevalent in libraries, they may be one method for controlling the costs of acquiring and maintaining hard-copy monographs. The next step is to determine eBook life cycle costs and to compare them with hard-copy book life cycle costs. The BE costs identified in this study can be used to calculate the costs of eBooks. eBook access fees will be included as mis­ cellaneous expenses and added to the total operating expenses. Printer paper and tele­ communications costs also will be added to the total operating expenses. The purchase and maintenance costs of computer termi­ nals and printers will be included in the fix­ tures and equipment cost category. The space allocations for computer workstations and printers will be calculated in the build­ ing and facilities cost category. A formula will need to be developed to calculate a percentage of all expenses associated with computer workstations and printers for eBook access because this equipment also is used to access the online public access catalog, electronic 550 College & Research Libraries November 2001 journals, databases, and the Internet. Moreover, a percentage of the telecommu­ nications costs will need to be allocated to the staff’s use of electronic resources for their work associated with acquisi­ tions, collection development, cataloging, circulation, interlibrary loan, and refer­ ence services. As computer equipment and telecommunications costs become more affordable and microwave technol­ ogy becomes more accessible, the costs as­ sociated with acquiring, accessing, and maintaining eBook collections may be less than the costs of maintaining hard-copy monograph collections. The determina­ tion of eBook life cycle costs and a com­ parison of these costs to the hard-copy life cycle books costs is the next logical step in calculating and identifying li­ brary material life cycle costs. Note: A copy of the analysis tool, the Library Interactive Costing Spreadsheet (LICS), used in this study is available online from http://bus.colorado.edu/faculty/lawrence/ LICS. Notes 1. Michael D. Cooper, “A Cost Comparison of Alternative Book Storage Strategies,” Library Quarterly 59, no. 3 (July 1989): 239–60. 2. Stephen A. Roberts, Cost Management for Library and Information Services (London: Butterworths, 1985). 3. Ibid. 4. Ibid. 5. Joan A. Maier, “Analyzing Acquisitions and Cataloging Costs,” in Costing and the Econom­ ics of Library and Information Services, ed. S. A. Roberts (London: Aslib, 1984). 6. John Ross and Jane Brooks, “Costing Manual and Computerised Library Circulation Sys­ tems,” in Costing and the Economics of Library and Information Services, ed. Stephen. A. Roberts (London: Aslib, 1984). 7. Joel A. Nachlas and Anton R. Pierce, “Determination of Unit Costs for Library Services,” College & Research Libraries 40, no. 3 (May 1979): 240–47. 8. Roberts, “Internal Costs of Inter-library Lending in British University Libraries,” Interlending Review 9, no. 3. (July 1981): 101–3. 9. ———, Cost Management for Library and Information Services. 10. Ibid., v. 11. Ferdinand F. Leimkuhler and Michael D. Cooper, “Cost Accounting and Analysis for Uni­ versity Libraries,” College & Research Libraries 32, no. 6 (Nov. 1971): 449–64. 12. Paul M. Gherman and Lynn S. Cochrane, “Developing and Using Unit Costs: The Virginia Tech Experience,” Library Administration & Management 3, no. 2 (spring 1989): 93–96. 13. Roberts, Cost Management for Library and Information Services. 14. Eileen G. Abels, Paul B. Kantor, and Tefko Saracevic, “Studying the Cost and Value of Library and Information Services: Applying Functional Cost Analysis to the Library in Transi­ tion,” Journal of the American Society for Information Science 47, no. 3 (Mar. 1996): 217–27. 15. Stephen A. Roberts, ed., Costing and the Economics of Library and Information Services (Lon­ don: Aslib, 1984). 16. ———, Cost Management for Library and Information Services. 17. Karen Svenningsen, “An Evaluation Model for Electronic Resources Utilizing Cost Analy­ sis,” Bottom Line 11, no. 1 (1998): 18–23. 18. Roberts, Cost Management for Library and Information Services. 19. Malcolm Getz, “An Economic Perspective on E-Publishing in Academia,” Journal of Elec­ tronic Publishing [online] 3, no. 1 (Sept. 1997). Available from . ISSN 1080-2711. 20. Martin M. Cummings, “Cost Analysis: Methods and Realities,” Library Administration & Management 3, no. 4 (fall 1989): 181. 21. Gherman and Cochrane, “Developing and Using Unit Costs.” 22. Madeline J. Daubert, Analyzing Library Costs for Decision-Making and Cost Recovery (Wash­ ington, D.C.: Special Libraries Association, 1997); Gherman and Cochrane, “Developing and Using Unit Costs”; Keyes D. Metcalf, Planning Academic and Research Library Buildings, 2d ed., eds. Philip D. Leighton and David C. Weber (Chicago: ALA, 1986); Malcolm Getz and Doug Phelps, “Labor Costs in the Technical Operation of Three Research Libraries,” Journal of Academic Librarianship 10, no. 4 (Sept. 1984): 209–19. 23. Metcalf, Planning Academic and Research Library Buildings, 564. http:http://www.press.umich.edu http://bus.colorado.edu/faculty/lawrence Life Cycle Costs of Library Collections 551 APPENDIX A ARL Survey Data The table below summarizes data that is collected annually by the Association of Research Libraries from its members. Variables in boldface type represent data used in this study. More extensive descriptions of the data can be found on the ARL Web site at address http:llwww.arl.orglstatslarlstatl98doc.html. Library Characteristics YEAR Survey of year INSTNO Library number INAM Library name TYPE Type of library REGION Geographic region MEMBYR Year library oined ARL LAW Law library included? MED Medical library included? EXCH Canadian exchange rate Collections VOLS Volumes held VOLSADG Volumes added gross VOLSADN Volumes added net MONO Monographs purchased SERPUR Current serials purchased SERNPUR Current serials not purchased CURRSER Total current serials MICROF Microforms GOVDOCS Government documents MSS Manuscripts and archives MAPS Cartographic materials GRAPHIC Graphic materials AUDIO Audio materials VIDEO Video and film COMPFIL Computer files Service Activities ILLTOT Total lending ILBTOT Total borrowing GRPPRES Group presentations PRESPTCP Presentation participants REFTRANS Reference transactions INITCIRC Initial circulations TOTCIRC Total circulations Personnel PRFSTF Professional staff NPRFSTF Support staff STUDAST Student assistants TOTSTFX Total staff and students (PRFSTF+NPRFSTF+STUDAST) 552 College & Research Libraries November 2001 APPENDIX A ARL Survey Data (continued) EXPMONO EXPSER EXPOTH EXPMISC EXPLM EXPBND SALPRF SALNPRF SALSTUD TOTSAL OPEXP TOTEXP TOTSTU GRADSTU Expenditures Expenditures for monographs Expenditures for current serials Expenditures for other materials Expenditures for non-materials Total expenditures for materials Expenditures for binding Professional salaries Non-professional salaries Student assistant wages Total salaries and wages (SALPRF+SALNPRF+SALSTUD) Other operating expenses Total expenditures (EXPLAM EXPBND+TOTSAL+OPEXP) Total full-time student enrollment Total graduate school enrollment PHDAWD PHDFLD FAC University Data Ph.D. degrees awarded Ph.D. fields Instructional faculty L ife C ycle C osts of L ib rary C ollection s 553 APPENDIX B Median Life Cycle Cost Estimates (1999 ARL Statistics) Library Collection Monographs Serials Microforms GovDocs Manuscripts Maps Graphics Au�io �i�eo Comp�iles Reported Holdings 2,772,663 22,913 3,944,886 283,661 13,507 183,706166,932 Book Equivalent Holdings 2,772,663 15,352 7,890 56,732 2,701 36,741 3,339 Relative BE Holdings (percent) 95.1% 0.5% 0.3% 1.9% 0.1% 1.3% 0.1% Unit Cost $ 47.78 $ 590.97 $0.11 $0.00 $4.46 $11.05 $0.06 Total Life Cycle Costs Monographs Serials Microforms GovDocs Manuscripts Maps Graphics 32,484 6,497 0.2% $6.80 Au�io 9,175 4,509 13,111 9 0.4% 0.0% $15.70 $0.01 �i�eo Comp�iles Operating Expenses 39.08 26.18 0.04 8.19 15.59 7.77 0.32 Wages & Salaries 113.68 76.16 0.11 22.55 45.81 22.33 0.95 Buildings & Facilities 35.88 24.04 0.04 7.05 14.32 7.18 0.29 Facility Maintenance 71.77 48.08 0.07 14.10 28.64 14.36 0.57 Fixtures & Equipment 35.88 24.04 0.04 7.05 14.32 7.18 0.29 Total Ongoing Expenses $299.74 $200.83 $0.34 $55.40 $122.33 $62.76 $2.84 Total Life Cycle Costs $343.03 $801.78 $0.45 $55.40 $126.79 $73.82 $2.91 2.37 6.95 2.15 4.31 2.15 $17.97 $24.77 11.41 0.01 33.28 0.02 10.26 0.01 20.52 0.01 10.26 0.01 $91.80 $0.06 $107.50 $0.07 Life Cycle Cost I Unit Cost 718% 134% 256% 311% 1130% 247% 216% Discounted Life Cycle Costs Monographs Serials Microforms GovDocs Manuscripts Maps Graphics 219% Audio 307% 331% �ideo Comp�iles Operating Expenses 7.51 6.98 0.02 2.18 2.08 2.07 0.21 Wages & Salaries 10.12 20.31 0.06 6.01 6.11 5.95 0.64 Buildings & Facilities 27.94 6.41 0.02 1.88 1.91 1.92 0.19 Facility Maintenance 8.74 12.82 0.04 3.76 3.82 3.83 0.38 Fixtures & Equipment 17.48 6.41 0.02 1.88 1.91 1.92 0.19 Total Ongoing Expenses $71.78 $52.94 $0.16 $14.13 $15.80 $15.72 $1.58 Discounted Life Cycle Costs $119.56 $643.91 $0.27 $14.13 $20.26 $26.78 $1.65 2.10 6.17 1.92 3.83 1.92 $15.85 $22.64 15.22 0.02 44.37 0.06 13.68 0.02 27.35 0.04 13.68 0.02 $113.25 $0.16 $128.95 $0.17 Discounted LC Cost I Unit Cost 265% 109% 183% 176% 237% 139% 178% 206% 376% 716%