College and Research Libraries UNG CHON KIM A Comparison of Two Out-of-Print Book Buying Methods Two out-of-print book buying methods, searching desiderata files against o.p. book catalogs and advertising want lists in The Library Bookseller, are compared based on the data collected from a sample of 168 titles. The o.p. catalog method requires more staff time, yet yields less success in locating desired titles. The cancellation rate for unfilled orders is lower in the advertisement method. The average price of books quoted on The Library Bookseller is slightly higher than those listed in o.p. catalogs, but the overall acquisition cost of o.p. book catalog searching is more costly than using The Library Bookseller. THE COMMON LY USED OUT-OF-PRINT BOOK BUYING METHODS by college and university libraries are checking desid- erata files against out-of-print booksell- ers' catalogs, sending lists of wanted titles to secondhand book dealers for searching, and advertising the needed titles for quote in publications such as The Library Bookseller or AB Book- man's Weekly.l Also known but less fre- quently used methods are the use of book scouts, book auctions, book buying trips, and book exchanges. Libraries use one or more of these methods to obtain out-of-print books depending on the ex- perience and preference of the librari- an in charge. According to Shirley Hep- pell' s survey in 1966, the most common- ly-used methods of college libraries are, in order of frequency, searching in o.p. book catalogs, direct contacts with spe- cialist dealers, use of search services, and advertising. 2 A similar survey for larger college and university libraries re- ported by Sarah Cook reveals that Mr. Kim is assistant head of the Ac- quisitions Department, Indiana State U ni- versity libraries, Terre Haute . 258/ checking want lists against incoming catalogs is considered most effective by eighty-one libraries, while the circula- tion of a desiderata file to antiquarian book dealers is preferred by sixty-one libraries. The submission of lists to LB (The Library Bookseller) is the first choice by fifty-seven libraries. 3 Several reports in library literature discuss the methods of buying out-of- print books, but the writers express vari- ous opinions and do not agree upon any one method. Checking all incoming lists of out-of-print books is considered a "must" by Felix Reichmann although it is "cumbersome, time-consuming and inefficient."4 Eldred Smith and Betty Mitchell report that the use of the search service of specialized dealers is more efficient than other techniques be- cause of costly staff time, high cancella- tion rate, and the amount of clerical work involved in other methods. 6 Emer- son Jacob considers advertising want lists in LB ccsuperior to other methods in numerous respects."6 The purpose of this study is to com- pare the effectiveness of two of these commonly used out-of-print book buy- 1 1 ing methods based upon the measured statistical data. The method of check- ing a desiderata file against o.p. catalogs and publishing want lists in LB are ex- amined here in terms of: ( 1) rate of success in locating wanted titles; ( 2) price of the book compared with the original publisher's price; ( 3) staff time involved in locating the titles; and ( 4) cancellation rate of unfilled orders. The study 1nay be further expanded for broader generalization. However, the present investigation is limited to a study of obtaining out-of-print books which are generally needed by most col- lege and university libraries. 7 Highly specialized research materials, rare books, and foreign language titles are not included. PROCEDURE The study was conducted at Indiana State University library during the fall semester of the 1971/72 academic year. In September the file of 232 recently re- ceived requests for searching were ex- amined, and 30 slips falling into the following categories were sorted out: ( 1) non-English titles and foreign pub- lication (excluding British output); ( 2) rare books for special collections; ( 3) special reports and pamphlets; and ( 4) government publications. The re- maining 202 slips were alphabetized and checked against the latest editions of Books in Print, British Books in Print, Forthcoming Books, and Whitaker's Books of the Month and Books to Come. A total of 27 titles were eliminat- ed which included two duplicates, five reprinted titles, eight follow-up cancel- lations, and twelve which were "tempo- rarily out-of-stock."8 Those with no original publisher's price were searched in the proper Cumulative Book Index or an earlier edition of Books in Print to obtain the original publisher's price. Seven titles whose price could not be verified were also removed. The selected sample of the remaining Book Buying Methods I 259 168 titles in alphabetical order were di- vided into two groups by alternating ti- tles of Group A and Group B. Group A was used as the sample for checking o.p. catalogs and Group B was utilized as the sample for the LB advertisement method. The eighty-four slips in Group A were kept in alphabetical order, but duplicate slips were made to file by sub- ject areas to facilitate the search of both classified and alphabetical o.p. cat- alogs. At the same time a list of eighty- four titles were made from slips in Group B for the LB letter requesting publication. On the same day in late September, the LB letter was mailed out and the check of incoming antiquarian booksellers' catalogs began. All catalogs received were examined by the writer to screen out those which were inappropriate (such as foreign book catalogs, special subject catalogs, rare book catalogs, etc.). Those remain- ing were given to a staff member to check against either the alphabetical or the subject file of Group A. The time spent searching was recorded and the lo- cated titles were reviewed for order. If the price was less than $20.00 and not more than three times the original pub- lisher's price, or more than $20.00 but less than double the publisher's price, and if the condition of the book was acceptable, the title was deemed accept- able for ordering. A form letter asking that the located titles be held was imme- diately mailed to the dealer. This process took place within a span of twenty-four hours from the time the catalog was received in the mail. The official purchase order was sent out with- in four days. The LB quotes for Group B began to be received in the first week of No- vember. Their receipt continued for three weeks and then faded out. Every second day during the period, titles quoted were pulled from the Group B file and marked with the quoted source, price, date, and special condition. The 260 I College & Research Libraries • September 1973 very same considerations, applied to the out-of-print catalog method, were used in deciding acceptableness of quotes. Earlier editions than specified, paper- back substitutes, and books listed as being in poor condition were rejected. When more than one quote was received for a title in the same listed condition, the less expensive title was selected. A form letter requesting that titles be held was sent out immediately for all accept- ed quotes. The titles with an unaccept- able quote were filed back with Group B for future use. At the end of Novem- ber, or after the end of nine weeks from the beginning of the measure- ments, the searching of both o.p. cata- logs for Group A and checking LB quotes for Group B were discontinued. RESULTS During the study period of nine weeks a total of ninety-eight catalogs were checked against the Group A file, fifteen titles (and one which was be- yond the set price limit) were located. The staff time required for searching totaled eighteen hours and forty min- utes. Table 1 shows the number of cat- alogs searched and number of titles spotted per week. In the same period a total of thirty-two acceptable LB quotes were received, mostly during the sixth, seventh, and eighth weeks. Seventeen other titles were also quoted, but not or- dered because the prices exceeded three times the original publication prices. Many titles were quoted more than once, in all a total of 124 quotes were received. Table 2 shows the number of quotes received and accepted each week. The average price of the fifteen titles located in the antiquarian book catalogs was $6.34. Table 3 gives the cost and original prices of these titles. Table 4 is for thirty-two titles accepted from the LB quotes. The average price for these titles was $9.00. The time associated with the process- ing of the eighty-four titles in Group A, including the preparation for search- ing, the searching and other activities totaled twenty-five hours and five min- utes. This does not include the time taken to make the initial selection of slips, checking Books In Print or Cumu- lative Book Index and other tasks which were completed prior to the division of the samples into Group A and Group B. The breakdown time is shown in Ta- ble 5. All measurements were carefully preplanned; however, accurate measure- TABLE 1 NuMBER OF CATALOGS SEARCHED AND TITLES LocATED PER WEEK Week Starting Sept. Oct. Oct. Oct. Oct. Nov. Nov. 27 4 11 18 25 1 8 Number of Catalogs Searched 11 13 10 9 11 14 13 Number of Titles Located 4 3 4 0 1 30 1 0 Includes one title not ordered TABLE 2 NuMBER OF ALL LB QuoTES AND AccEPTED QuoTES PER WEEK Week Starting Number of All Quotes Number of Accepted Quotes Sept. Oct. 27 4 0 0 0 0 Oct. Oct. Oct. 11 18 25 0 0 0 0 0 0 Nov. Nov, 1 8 83 21 27 3 0 Total number of quotes which includes certain titles quoted more than once Nov. Nov. 15 22 Total 10 7 98 0 0 16° Nov. Nov. 15 22 Total 18 2 124° 2 0 32 1 i 00 00 ,..-j lf5lf5 ,......{ oo om C':l 00 Ol OQ') coco ,..-j I:Olri Book Buying Methods I 261 ment for some activities was not pos- sible in practice and in such cases esti- mated times were used. The time re- quired to prepare the LB letter, the process of handling LB quotes, and or- dering titles in Group B is summarized in Table 6. The total time for the LB method was six hours and twenty-two minutes. Thirty titles out of thirty-two LB quoted titles were actually received. The remaining two were cancelled on a fol- low-up cancellation which took place in April 1972. Of the out-of-print catalog orders, nine out of fifteen were re- ceived; five were reported c'sold" or •cout-of-stock indefinitely" and one was cancelled on the same follow-up. DISCUSSION Rate of Success in Locating Desired Titles: By carefully searching the wanted titles in ninety-eight out-of- print catalogs, fifteen acceptable titles were located. This represents 18 percent of the eighty-four titles. The success achieved by placing ads in LB was high- er than the o.p. catalogs check method. Overall thirty-two out of eighty-four titles, or 38 percent of the titles were found. The results of this study indi- cate that by publishing a want list in LB it is possible to locate more titles than by checking o.p. catalogs; more- over, the success rate was about double. When one has certain urgently needed titles, or is ready to pay more than triple the original publisher's price, the success rate is even higher. If seventeen titles which were quoted at high prices are added to the thirty-two titles, the per- centage of titles located rises to 58 per- cent. It is undeniable that some of the titles quoted were very expensive; how- ever, when considering the price of re- prints or photo duplication which are usually more than three times the orig- inal price, an acquisition librarian may wish to pay the high price for out-of- 262 I College & Research Libraries • September 1973 TABLE 5 TIME SPENT FOR O.P. CATALOG METHOD Activities Duplication of slips to be used for subject £le Classification and £le of titles by subject areas Measured or Estimated Time 1 hour 45 min. (measured) 50 min. (measured) Preexamination of catalogs (approximately 1 minute each for about 200 3 hours 20 min. ( es timated ) catalogs) Searching time (for 98 catalogs) Preparation of form letters (approximately 2 minutes each for 15 titles ) Total 18 hours 40 min. (measured ) 30 min. (estimated) 25 hours 5 min. TABLE 6 TIME SPENT FOR LB QuoTE METHOD Activities Preparation of LB list and letter Measured or Estimated Time Process of LB quotes-to pull out quoted titles from the £le, evaluation of quotes, price check, and £le back nonacceptable quotes 1 hour 10 min. (measured) 4 hours 8 min. ( estimated ) ( approximately 2 minutes each for 124 quotes) Preparation of form letters (approximately 2 minutes each for 32 titles ) 1 hour 4 min. ( estimated ) Total print books, even when an extra cost is required for rebinding. Price of Out-of-Print Books Com- pared with Original Price: The total price of books found in out-of-print catalogs was $95.05 and was 1.56 times the total original price of $60.85. The total quoted price for titles listed in the LB advertisement was $287.96 or 1.78 times of the original price of $161.60. When considered solely in terms of the book price, checking o.p. catalogs is more economical than using LB. The difference in average price, however, was not critical and, if necessary, the differ- ential could be reduced by holding the received quotes several days and com- paring all the quotes. Of course, the delay introduces the chance of losing the quoted title, espe- cially when only one quote is received. Considering the fact that desiderata 6 hours 22 min. files should include only those which are urgently needed, this process is not high- ly recommended. The diversity of the quoted prices for the same title with similar condition in the LB method is notable. One title was quoted eight times with a price ranging from $4.00 to $15.00. Quotes are based presumably on the dealer's perception of how difficult it was to ob- tain a particular title. Therefore, no li- brarian should criticize the judgment of what constitutes a fair price. On the other hand, acquisition librarians must also spend tax money as wisely as pos- sible by paying the lowest price possible. Most dealer prices are usually reason- able, but the prices of some dealers were found to be consistently high. It is possible to establish a list of ccreason- able" dealers based on an average price index for each bookseller by dividing l v t the average quoted price by the average original price. This is possible only when more than several quotes are re- ceived from one dealer. The index can be used as a helpful reference when there is only one quote, when original price is not known, or when handling a large number of out-of-print titles. Many sellers' indexes are more or less 1.5; however, the indexes for some deal- ers were found to be as high as 4.75. (However, this index should be used only as a guide, not as an absolute tool.) Staff Time Needed to Locate Titles: The method of checking out-of-print catalogs is more time consuming than the LB advertisement method. The total staff time needed to locate fifteen valid titles in o.p. catalogs was twenty-five hours and five minutes, the average being one hour and forty minutes per title. On the other hand, only about twelve minutes of staff time were need- ed per title located by the placement of ads, or a total of six hours and twenty- two minutes for thirty-two titles. When these are ·converted into dollar costs, the out-of-print catalog method costs $4.17 per title, calculated on the basis $2.50 for the hourly wage of a staff member. The LB advertisement method costs about $.50 to locate one title. There is, however, a subscription charge for LB of $25.00 per year to become eligible to place advertisements. Since a cycle of the LB advertisement sequence takes about three months, Indiana State Uni- versity library sends lists of out-of-print titles about four times a year. The ser- vice charge is then calculated at $6.25 per issue, in this study, or about $.20 per located title. When comparing $4.17 for the o.p. catalogs searching with $.70 ($.50 for wage and $.20 for subscrip- tion charge) for the LB method (or one hour-forty minutes per title against twelve minutes), the latter is far more favorable. The major problem with the out-of- Book Buying Me.thods I 263 print catalog method is the lengthy, less productive, and tedious process of searching wanted titles against numer- ous secondhand book catalogs which can be compared with ''looking for a needle in a haystack." The arrangement of catalogs and the different scheme of classification in subject catalogs makes searching even more difficult. Some an- tiquarian book dealer catalogs list titles alphabetically under broad subject fields such as economics, history, literature, etc.; while others use smaller classifica- tions and list titles in several alphabet- ical groups such as American history, European history, ancient history, or modern history. Some booksellers fur- ther subdivide European history, for ex- ample, by countries, others subgroup these into centuries. It is conceivable at this time to think of storing desiderata lists in a machine-readable data file such as on punched cards, tapes, or discs. All incoming catalogs then could be con- verted into machine-readable form, and a computer can be used to search desired titles against received out-of-print book lists. Unfortunately, most antiquarian booksellers' catalogs are printed or mim- eographed and libraries would be re- quired to convert the printed pages into machine-readable form. This is an ex- pensive process, and until o.p. dealers start publishing available out-of-;print titles in a new form, or libraries initiate a large scale cooperative venture, the procedure is not yet a practical one. Cancellation Rate for Unfilled Or- ders: A notable difference exists between the cancellation rate of the two out-of- print book buying methods. In spite of immediate notification to dealers after the decision to order, only nine titles of fifteen titles were actually received, when the o.p. catalog method was used. With the LB orders, thirty out of a to- tal of thirty-two orders were received. Since out-of-print catalogs are printed by antiquarian booksellers and hundreds 264 I College & Research Libraries • September 1973 of copies are distributed to libraries throughout the country, an acquisition librarian must compete with his fellow librarians who may have received the same catalog in the mail sooner. In this respect the LB quote method is a direct communication between a particular li- brarian and a dealer and is comparative- ly safe unless the librarian delays his re- sponse too long in the hope of receiving a less ·expensive quote later. CoNCLUSION The result of the study shows that ob- taining out-of-print books by searching o.p. catalogs requires more staff time and yields a lower success rate than ad- vertising in The Library Bookseller. The cancellation rate of unfilled orders is higher in the catalog method. The higher success rate and lower cancella- tion rate with the LB method makes it more favorable than the out-of-print~ catalog search method. The price of books listed in anti- quarian booksellers' catalogs are general- ly less expensive than those quoted in LB advertisements. However, when con- sidering this in relation to the much larger staff costs needed to search o. p. catalogs, out-of-print titles are obtain- able at a lower cost by advertising. REFERENCES 1. The Library Bookseller is known as T AAB to many librarians. But the more proper ti- tle, The Library Bookseller, is used and ab- breviated as LB in this article. 2. Shirley G. Heppell, "A Survey of OP Buy- ing Practices," Library Resources and Tech- nical Services 10: 28-30 ( Winter 1966 ) . 3. Sarah A. Cook, "The Selective Purchase of Out-of-Print Books: A Survey of Practices," Library Resources and Technical Services 10:31-37 (Winter 1966). 4. Felix Reichmann, "Purchase of Out-of-Print Material in American University Libraries," Library Trends 18: 328-53 (Jan. 1970). 5. Eldred Smith, "Out-of-Print Booksearching," CRL 29:303- 9 (July 1968). Betty J. Mitch- ell, "A Systematic Approach to Performance Evaluation of Out-of-Print Book Dealers: The San Fernando Valley State College Ex- perience," Library Resources and Technical Services 15:215-22 (Spring 1971). 6. Emerson Jacob, "The Use of TAAB in Out- of-Print Book Searching," C RL 17: 16-18 (Jan. 1956). 7. Titles included are not limited to specially selected basic titles such as those listed in Books For College Libraries. 8. Re: A follow-up cancellation: Indiana State University library periodically sends out fol- low-up letters for all outstanding orders which are more than six months old and can- cels out those for which no response has been received. These titles cancelled on a follow-up were not included in the study be- cause they may or may not be an out-of- print book. '(