Global Resource Sharing: Experiences of American Research Libraries in the Last Decades and Future Directions in the Next Decade

Mary E. Jackson
Senior Program Officer for Access Services
Association of Research Libraries
Washington, DC USA

University Library Directors Conference on Global Resource Sharing
May 22, 2000
Tokyo, Japan



It is a great honor to be invited to talk about global resource sharing to directors of university libraries. I would like to thank the Japan Association of Private University Libraries and its Committee for International Library Cooperations for providing the opportunity to make my second trip to Japan enjoyable and possible. I only regret that I didn't learn to speak Japanese fluently on my first visit; because I didn't, I must deliver this speech in English.

Over the next 90 minutes my goal is to describe the key components of American resource sharing, primarily focusing on interlibrary loan and document delivery. Although I will share many examples of effective strategies from the United States and Canada, I will also focus on some areas that still need our attention and improvement. My presentation will conclude with some suggestions for possible areas of collaboration between libraries in North American and Japan. I will also leave time for questions.
Let me mention the terminology I will use this afternoon. To many American librarians, resource sharing is a general term that includes interlibrary loan, document delivery, cooperative collection development activities, and on-site access. Interlibrary loan is the common term that includes borrowing and lending of books and copies. Document delivery is a term used to describe either photocopy traffic between libraries, use of commercial document suppliers, or physical delivery of books. For this afternoon I will probably use interlibrary loan, or ILL, to mean the requesting and supply of books and photocopies.

What are the Key Components of ILL? 

I will list five key components that enable ILL to function effectively and efficiently. I'm not sure if one is more important than another, but they all are the foundation blocks that permit North American libraries to provide ILL services.
1. Willingness to Cooperate

First, librarians must have a willingness to cooperate. Historically, ILL has been viewed as a service that served only a few faculty members and/or students as research libraries had the means to acquire the vast majority of what local users needed. Research libraries viewed lending as a burden and borrowing as a shame. Research libraries were asked to fill many more ILL requests than they initiated, and requests for items not in local collections was viewed by some as an admission of collection failure. Fortunately, this attitude is no longer prevalant as most administrators of research libraries understand that ILL is mutually beneficial to borrowers and lenders. 

Reciprocity describes two different behaviors. The first is the willingness to lend only if a library borrows. The second is a belief that a library's borrowing should equal its lending. Reciprocity in both forms is a very strongly held belief, especially by ILL managers. Many ILL librarians have expressed concern that certain libraries only borrow; they are not willing to lend. But, the concept of reciprocity has been fading. Over the past five years more and more libraries have begun to look at ILL from a business perspective. They charge for lending, and they are willing to pay when they borrow material for their local patrons. These libraries establish budgets to pay for materials they borrow, and even more importantly, they do not spend staff time searching for lenders that do not charge. However, most library administrators do not view ILL as economic equity because most are net lenders.

2. Knowledge of Holdings

Once a library has committed to the principle of interlibrary loan, then the practical aspects become key. A second key component is knowledge of holdings of other libraries. The print National Union Catalog and regional union catalogs have been replaced by the now-international bibliographic utilities of OCLC and RLIN. Within the past decade state-based union card catalogs have been replaced by CD-ROM catalogs and union lists on OCLC. Libraries are beginning to embrace the virtual union catalog now that Z39.50 searching can link catalogs.

3. Request Transmission

The next step of the process, and the third key element, is the ability to send requests to other libraries. We have moved from mailing or faxing paper request forms to using the OCLC, RLIN, and/or Docline online ILL systems. Libraries in the U.S. are just beginning to implement ILL messaging systems that use the ISO ILL Protocol, the international communication standard that permits different ILL systems to exchange ILL requests and responses. Protocol-compliant systems will be key to international ILL as ILL staff will be able to receive international ILL requests via the method they use to receive domestic requests. I will talk more about the ISO ILL Protocol when I talk about future developments.

4. Effective Internal Procedures

The fourth key component is effective internal procedures. Online requesting has eliminated postal delays in getting requests to potential lenders. In addition to streamlined procedures, libraries need to have appropriate staffing levels to handle the volume of borrowing and lending requests they process. Imagine a circulation desk that was not staffed to handle the number of books checked out in a typical day!

High-performing lenders have developed procedures and policies that permit them to fill requests in a timely manner and with minimal staffing. Lenders also maintain and publicize detailed lending policies so borrowing libraries will not send them requests that the lender would never fill.

5. Transport

Transport is the fifth key component. Transport is an oft-neglected, but critically important, component of the ILL process. Transport is physical and electronic. Many U.S. libraries prefer not to ship ILL books via the U.S. Postal Service because, although it is the most inexpensive option, it has been characterized as the least reliable. State, regional, and/or national commercial carriers are generally more expensive, but provide much more reliable service. Some U.S. libraries are reluctant to ship their books outside the U.S., but many libraries are willing to ship internationally as long as the material is shipped and returned via a commercial carrier. 

For electronic delivery, most research libraries use the Ariel software developed by the Research Libraries Group to send and receive photocopies. Ariel has been likened to fax on the Internet as it sends a compressed page image to another Ariel workstation or to the user's email address. Fax is also used to send articles, but some libraries are not as willing to fax articles to international libraries because of the potentially large phone charges.

ILL in the United States: A Very Brief History

Interlibrary loan has held a distinguished, if rather specialized, place in American library services. In 1876 Samuel Green, director of the Worcester Massachusetts Free Public Library, published a letter in the first issue of Library Journal suggesting interlibrary loan. The Library of Congress established the print-based National Union Catalog in 1901. In the early 1900s the head of the Princeton University Library called for the development of union catalogs, rationalization of collections, and interlibrary loan, all ideally served by a national lending library. In 1917, the American Library Association developed the first Interlibrary Loan Code.

Other Key ILL Values

As you can see, the concepts and values of interlibrary loan have roots firmly established in the earliest library development in the United States. This very brief history has illustrated the long-standing need for effective interlibrary loan services. But, there are seven other values that I would like to share with you.

1. Centralized vs. Decentralized

For many years, librarians debated whether ILL should be centralized like the United Kingdom's British Library Document Supply Centre or expand the decentralized model where all libraries act as borrowers and lenders. The Association of Research Libraries' 1998 ILL/DD Performance Measures Study found a significant correlation between the volume of lending and the unit cost of lending. This finding suggested that a centralized model might be more cost effective, but is contrary to our current model. We do not have a national clearinghouse for ILL transactions, and are unlikely to establish one in the near future. 

2. The Importance of ILL Codes

In the U.S., the American Library Association assumes responsibility for maintaining our national ILL code. This code is designed to govern transactions between libraries that have not established any other agreements. Our national code articulates responsibilities of borrowers and lenders. Our 1993 code is currently under revision and I expect that the revised code will less restrictive than the 1993 code.

The IFLA Code for International Lending governs requests between countries, and is the code that governs transactions between Japan and the U.S. It is also somewhat dates, and there has been discussion in the IFLA Document Delivery and Interlending Committee that it is time to revise and update the code.

3. The Role of Standards

Technical and performance standards are critical to effective ILL services, but for the most part, they should be invisible to the user placing the request. The ISO ILL Protocol, which I mentioned above, governs communication between different ILL systems. Implementation of the ISO ILL Protocol by key international vendors, including NACSIS, will permit libraries to send and receive international ILL requests via the system they use to send and receive national requests.
A second key standard is being developed by the National Information Standards Organization, but with some international input. The NISO Circulation Interchange Protocol is being designed to support four models: self-service circulation, interchange between circulation and interlibrary loan systems, direct consortial borrowing, and access to electronic resources.

4. Finance

Most research libraries charge to lend a book or supply a photocopy. The challenge in an international setting is to develop a method that permits lenders to charge and borrowers to pay, but without the costly fees charged by banks to convert currencies. One model we are testing in the Japan Journal Access Project is to waive fees, but that is not a long-term solution. A second model being tested by the German Resources Project is for the Association of Research Libraries to act as the "banker" that permits North American libraries to deposit money with ARL and ARL to pay the lending/supply fees to the German suppliers. Both options are costly and can not be considered as long-term options. The plastic IFLA voucher works for requests sent by post, but is also not a long-term solution. We need to work together to find payment options for ILL transactions that are cost-effective for the borrower and lender.

5. Consortia

Another key value is the concept of working together in library consortia. Research libraries belong to an average of seven library consortia for ILL purposes alone, some of which are regional and some national. Consortia are assuming an increasingly important role in negotiating contracts for ILL messaging systems as well as license agreements to electronic resources. Consortia often establish performance goals for ILL services, such as RLG's four-day limit to respond to lending requests.

6. Copyright

The U.S. copyright law permits libraries to engage in interlibrary loan transactions as long as the ILL requests do not substitute for a subscription to the journal. Our law does not explicitly prohibit the use of fax or Ariel to transmit articles. On an international level, the Berne Convention governs uses of copyrighted materials in member countries. Member countries of the Berne Convention agree to provide the same level of copyright protection to other member countries as it provides to its own copyright owners. Varying copyright laws may be one of the most challenging aspects of international interlibrary loan.

7. User-initiated Requesting

Many libraries in the U. S. and Canada are interested in how they can implement systems that permit users to search union or virtual catalogs and then place their own ILL requests. Informal studies suggest that user-initiated requesting is much more cost-effective than mediated ILL transactions. The OhioLINK model ISOne that is being replicated by other consortia in the U.S. and Canada.

Other Components of Resource Sharing

I would like to say a few words about the other components of resource sharing. I can cite examples of many successful local and regional cooperative collection development initiatives, but few of these have succeeded at a national level. ARL's Global Resources Program is a new effort to rationalize collection purchases, but it has been interesting to watch several of the projects evolve into document delivery projects. The Japan Journal Access Project's document delivery pilot with several ANUL libraries ISOne such example and another example is the German Resources Project that permits North American libraries to order articles from German libraries. We are trying to find a way to permit German libraries to order from North American libraries.

International ILL

I would also like to make a few comments about international ILL. For U.S. libraries, international ILL represents less than one percent of their borrowing or lending. Most libraries are more willing to supply photocopies than to lend books or other non-returnables. I've already mentioned the expense of currency conversion to pay or collect lending fees. A final concern relates to slow and unreliable delivery of material; we are reluctant to pay for expedited delivery, but we worry that the U.S. Postal Service's Airmail delivery is slow and unreliable.

Looking to the Future

If I could look into a crystal ball and forecast accurately the first decade of the 21st Century for North American libraries, I would probably project a future in which much of the current, mediated ILL traffic moves to the user-initiated systems. As a result, the volume of mediated ILL transactions will not grow at the current rate of 8%, but may only grow at a rate of 1-2% because of the increased use of user-initiated systems and wider availability of electronic resources. U.S. ILL borrowing and lending will become more international in scope, but the percentage sent or received from libraries outside the U.S. will still be less than 10%. Turnaround time will be less than 10 days and borrowing fill rates will exceed 85%. We will measure success by how quickly we fill request for our local users.

Possible Areas of Collaboration

I hope I've given you an accurate picture of interlibrary loan activities in the United States. We can be proud of our many achievements, but there are also many areas in which we need to improve. But, I would like to conclude my presentation by suggestion several areas of potential cooperation between libraries in the United States and Japan that would improve sharing of materials between our countries.

The first area of collaboration is in the area of standards. I can envision a future in which Japanese libraries use NACSIS to send ILL requests to U.S. libraries, and libraries in the U.S. receive requests via ILL Protocol-compliant software such as RLG's ILL Manager, the OCLC ILL System, or any of the growing number of Protocol-compliant ILL systems. Libraries in each country will continue to use the messaging system they prefer, but we will expand international ILL.
A second area relates to ILL charges. We need to find a way to pay the lending fees charged by U.S. or Japanese libraries. We cannot waive our lending fees, but we cannot pay the steep bank charges to convert Japanese Yen into U.S. dollars and vice versa. The IFLA Office of International Lending is exploring how they can convert the plastic voucher into an electronic equivalent, and perhaps our libraries can test options developed by IFLA.

A final area of possible cooperation relates to delivery. We need to find ways to send material, both books and photocopies, by the most effective means possible. We should be willing to pay for expedited delivery, and we need to find ways to permit the use of electronic delivery methods for copies of journal articles.


In conclusion, I believe I have given you an accurate overview of interlibrary loan activities in libraries in the United States. I am hopeful that the suggestions I have made for possible areas of cooperation are ones that you also feel are important.

It has been a real pleasure to talk about the current interlibrary loan environment in the United States. I hope that we will find ways of continuing the collaboration first established by Waseda University and the seven participants of the ANUL document delivery project. 
Thank you for your attention.